MEMOIRS

OF" THE

ToRREY Botanical Club

VOL. XVII

PROCEEDINGS

OF THE

SEMI-CENTENNIAL ANNIVERSARY

OF THE

TORREY BOTANICAL CLUB

October i8, 19 and 20, 191 7

New York
1918

PROCEEDINGS

OF THE

SEMI-CENTENNIAL ANNIVERSARY

OF thb:

TORREY BOTANICAL CLUB

October i8, 19 and 20, 1917

Issued June io, 1918

PRESS OF
THE NEW ERA PRINTING COMPANY
LANCASTER, PA.

CONTENTS

Meeting OF Thursday, October i8 i

Discussion of a proposed Botanical Abstract Journal. ... 3

Meeting of Friday, October 19 7

Meeting of Saturday, October 20 8

Barnhart, John Hendley. Historical sketch of the Torrey

Botanical Club 12

Denslow, Herbert McKenzie. Reminiscences 22

Britton, Nathaniel Lord. Torrey Botanical Club reminiscences 24

HoLLiCK, Arthur. Torrey Botanical Club reminiscences 29

Demarest, Sarah Austin. A sketch of the life of Coe Finch

Austin 31

RusBY, H. H. Recent botanical collecting in the Republic of

Colombia 39

MuRRiLL, William A. Collecting Fungi at Delaware Water Gap. 48

Burgess, Edward S. A method of teaching economic botany. . 52

Graff, Paul Weidemeyer. Philippine Micromycetous Fungi. . 56
Blodgett, Frederick H. Weather conditions and crop diseases

in Texas 74

Scott, James Grimshaw. Early horticultural journalism in the

United States 79

Farwell, Oliver Atkins. Sisyrinchium Bermudiana 82

Lloyd, Francis E. The effect of acids and alkalies on the growth

ot the protoplasm in pollen tubes 84

Campbell, Douglas Houghton. The origin of the Hawaiian

flora 90

Arthur, J. C, and Johnston, J. R. Uredinales of Cuba 97

Levine, Michael. The physiological properties of two species of

poisonous mushrooms (with plates i and 2) 176

Seaver, F. J., and Horne, W. T. Life-history studies in Sclero-

tinia (with plate 3) 202

Boas, Helene M. The individuality of the bean pod as compared

with that of the bean plant 207

Harper, R. A. The evolution of cell types and contact and pres-
sure responses in Pediastriim 210

Richards, Herbert M. Determination of acidity in' plant

tissues 241

V

vi Contents

Atkinson, George F. Six misunderstood species of Amanita . . 246

Dodge, B. O., and Adams, J. F. Some observations on the de-
velopment of Peridermium Cerebrum (with plates 4-6) 253

Harper, Roland M. The vegetation of the Hempstead Plains
(with plate 7) 262

Steinberg, R. A. A study of some factorr influencing the stimu-
lative action of zinc sulphate on the growth of Aspergillus
niger. I. The effect of the presence of zinc in the cultural flasks 287

Medsger. Oliver P. Two months in the southern Catskills. . . . 294

Burlingham, Gertrude S. A preliminary report on the Russu-
lae of Long Island 301

Harris, J, Arthur. On the osmotic concentration of the tissue
fluids of desert Loranthaceae 307

White, Orland E. Inheritance studies in Pisum. III. The

inheritance of height in peas 316

Graham, Margaret. Centrosomes during early fertilization

stages in Preissia quadrata (with plate 8) 323

Adams, J. F. Origin and development of the lamellae in Schizo-
phyllum commune (with plate 9) 326

Stout, A. B., and Boas, Helene M. Statistical studies of flower
number per head in Cichorium Intybus: kinds of variability,
heredity, and efl^ecte of selection (with plates 10-13) 334

Hazen, Tracy E. The trimorphism and insect visitors of Pon-
tederia (with plates 14 and 15) 459

PREFACE

Of the papers presented at the celebration of the Semi-cen-
tennial Anniversary of the Torrey Botanical Club, one, "The
ferns of tropical Florida" by John K. Small, has already been
published elsewhere (Am. Mus. Jour. i8: 126-134. 1918); an-
other, on ''Bermuda Algae" by Marshall A. Howe, was, in sub-
stance, an illustrated summary of a contribution to Britton's
"Flora of Bermuda," recently published, and is not here repro-
duced; in a somewhat similar way, the paper on the "Flora of the
Rocky Mountains and adjacent plains" by P. A. Rydberg was
in considerable part a summary of the principal results attained
in the author's book of that title, then issuing from the press.
A few other papers actually presented or read by title are omitted
either because copy failed to reach the editor in season or because
arrangements had been made for publication elsewhere. Such
papers include "Fossil plants from Porto Rico" by Arthur Hollick^
"A cotton-rust epidemic in Texas" by E. W. Olive, " The flora of
the Isle of Pines, Cuba," by N. L. Britton and Percy Wilson,
"The route taken by Capt. Nathaniel J. Wyeth and Mr. Thomas
Nuttall from the Mississippi River to the Columbia River in 1843"
by W. W. Eggleston, "Comparative cultures of seed-plants in
desert valley, desert mountain, and coastal locations" by D. T.
MacDougal, "The Vegetation of Montauk, Long Island" by
Norman Taylor, and " Parthenocarpy in cucumbers" by A. F.
Blakeslee and P. A. Warren.

Marshall A. Howe,

Editor

vii

ERRATA

Page 7, following line 23, insert "Torrey Botanical Club reminis-
cences" by Dr. Arthur Hollick.
Page 49, line 4, for Transchelia, read Tranzschelia.
Page 58, line 4 from bottom, for Triblidiella read Tryblidiella.
Page 58, line 2 from bottom, for rufula, read rufulum.
Page 69, line 2, for paradisica, read paradisiaca.
Page 176, last line, ior fasiculare, redid fascicular e.
Page 178, line 14, for decepiens, read decipiens.

viii

PROCEEDINGS

OF THE

SEMI-CENTENNIAL ANNIVERSARY

OF THE

TORREY BOTANICAL CLUB

OCTOBER 18, 19 and 20, 1917

Meeting of Thursday, October i8

The meeting of October i8 was held in the lecture-room of the
Department of Botany of Columbia University in Schermerhorn
Hall, beginning at 2:io P. M., with Professor Herbert M. Rich-
ards, President of the Club, in the chair. Fifty-five persons,
mostly members of the Club, were present.

President Richards, in his opening remarks, referred to the im-
portance of the occasion and expressed his appreciation of the
honor of presiding at such an anniversary gathering.

Dr. R. A. Harper, Torrey Professor of Botany in Columbia
University, recalled the intimate associations of the Torrey Bo-
tanical Club and the Department of Botany of Columbia Uni-
versity, and on behalf of the University extended to the Club
felicitations on its Fiftieth Anniversary and a most cordial wel-
come to the University.

The following historical, reminiscent, and scientific papers,*
most of which are published in full in this Anniversary Memoir,
were then presented:

"History of the Torrey Botanical Club," by Dr. John Hendley
Barnhart.

* Invitations to present papers were, in the intent of the committee, restricted
to actual members of the Club. In the program as rendered there were few deviations
from this limitation.

1

2

Semi-centennial of Torrey Botanical Club

"Reminiscences," by Rev. Dr. H. M. Denslow.

"A sketch of the hfe of Coe Finch Austin," by Mrs. Abraham
Demarest (Sarah Elizabeth Austin).

"Torrey Botanical Club reminiscences," by Dr. N. L. Britton.

"Contact and pressure reactions in Pediastrum simplex,'' by
Professor R. A. Harper. Illustrated by lantern-slides. Dis-
cussion by Drs. N. L. Britton, A. F. Blakeslee, E. W. Olive,
and J. H. Barnhart and Professor G. F. Atkinson.

"The origin of the Hawaiian flora," by Professor Douglas Hough-
ton Campbell. (Read by title in absence of the author.)

"Uredinales of Cuba," by Professor J. C. Arthur and Mr. J. R.
Johnston. (Read by title in absence of authors.)

"Six misunderstood species of Amanita,'" by Professor George F.
Atkinson. Illustrated by lantern-slides.

** Sisyrinchium Bermudiana Linnaeus," by Oliver Atkins Farwell.
(Read by title in absence of the author.)

"The individuality of the bean-pod as compared with that of the
bean-plant," by Helene M. Boas. (Illustrated by charts.)

"Two months in the southern Catskills," by Mr. 0. P. Medsger.
Illustrated by photographs. Discussion by Dr. N. L. Britton.

"The ferns of tropical Florida," by Dr. John K. Small. Illus-
trated by map and photographs. Discussion by Dr. N. L.
Britton. (Published in Am. Mus. Jour. i8: 127-135. 1918
[Illust.].)

"Fossil plants from Porto Rico," by Dr. Arthur Hollick. (Read
by title in absence of the author.)

"A cotton-rust epidemic in Texas," by Dr. E. W. Olive. Dis-
cussion by Professor G. F. Atkinson.

"Bermuda Algae," by Dr. Marshall A. Howe. Illustrated by
lantern-slides. Discussion by Drs. Tracy E. Hazen and N. L.
Britton. (Published in more extended form in Britton, N. L.,
"Flora of Bermuda," in press at the time of presentation.)

"Some factors influencing the stimulative action of zinc sulphate
on the growth of Aspergillus niger," by Mr. R. A. Steinberg.
(Introduced by Professor R. A. Harper.) Discussion by Pro-
fessors George F. Atkinson and Herbert M. Richards and Drs.
N. L. Britton and A. F. Blakeslee.

"Centrosomes in fertilization stages of Preissia quadrata," by Dr.
Margaret A. Graham. Illustrated by lantern-slides.

Meeting of Thursday, October i8

3

"Philippine Micromycetous Fungi," by Professor Paul Weide-

meyer Graff. (Read by title in absence of the author.)
"A method of teaching economic botany," by Professor Edward
S. Burgess. (Read by title in absence of the author.)
After the presentation of the above papers, adjournment was
made to the Columbia University Faculty Club, where dinner was
served to thirty-seven persons. Following the dinner, there was
an informal discussion of the proposal to establish in America a
Botanical Abstract Journal.

Dr. W. a. Murrill, editor of Mycologia, introduced the sub-
ject of the evening by referring to correspondence with Dr.
Donald Reddick, editor of Phytopathology, and by outlining vari-
ous phases of the subject that seemed to need discussion at this
time. He spoke of the work already done by the Bulletin of the
Torrey Botanical Club, Mycologia, and Phytopathology , in con-
nection with the indexing and abstracting of botanical literature.
Applied botany, according to Dr. Murrill, is now looked after
pretty well by the U. S. Department of Agriculture, but the Gov-
ernment's work in this line could hardly be extended to include
purely scientific articles. Original articles are now going begging,
especially costly illustrated ones. It might be possible to reduce
the pressure on existing journals by greatly condensing some of
these articles and preventing the duplication of material in various
publications. To make a new abstract journal a success would
require a paid board of editors, and a prompt, complete, and ac-
curate account of all botanical publications. This would be a
very expensive and arduous undertaking and such a journal would
have to compete with the Botanisches Centralblatt when the war
is over.

Dr. C. Stuart Gager, in emphasizing the need of such a pub-
lication as "Botanical Abstracts," referred to the inadequacy of
the botanical portion of Experiment Station Record, on account of
its omissions, and its custom of translating all foreign-language
titles into English, thereby making citations difficult. Reference
was also made to the recent enormous increase in the bulk of bo-
tanical publication, rendering it practically impossible for any one
to keep in touch with the literature by depending on original
sources. The opinion was expressed that the "Abstracts" should
not endeavor to be full enough to render the reading of the original
papers unnecessary (as one speaker had suggested). Lack of
funds seems to be the only obstacle to launching the enterprise,
and no adequate source of funds seems to be in sight.

Mr. Norman Taylor, editor of Torreya, made the suggestion

4 Semi-centennial of Torrey Botanical Club

that the Torrey Botanical Club so change the character of Torreya
that it would be entirely of an Abstract nature. The desirability
of enlarging that journal to accommodate the greater amount
of material involved in the proposed scheme was urged, par-
ticularly in view of the facts that the necessary steps in the inau-
guration of a new journal must be somewhat slow and halting,
and that Torreya, with a slight modification of existing conditions,
could be made available at once. The income now set aside from
Club dues and that derived from subscriptions would have to be
augmented from other sources. The proposal was made that
financial assistance for the first two years should be guaranteed
by the Club and that subsequently the greatly increased circulation
of Torreya, in its new form, would tend to make it a self-sustaining
journal.

Professor R. A. Harper emphasized the need of critical
reviews, as well as colorless abstracts, with a view to raising the
general standards of our work and perhaps in some cases to reduc-
ing the length and number of papers published.

Dr. N. L. Britton gave a tentative estimate of the number of
pages of such an Abstract Journal and of the probable annual cost
of printing.

Professor George F. Atkinson remarked : L have given this
subject of an Abstract Journal, to be published in this country,
very little thought, although I have known for several months that
it has been under consideration by some botanists. What I shall
now say on the subject is of course not the result of careful delibera-
tion nor have I had the opportunity of hearing any very definite
suggestions concerning the plan, aside from the remarks I have
heard this evening, except that the abstracts should be quite full,
that the journal should be published in the English language, and
that the abstracts should be published within a reasonable time
after the appearance of the original contribution.

I have felt that there was a great waste of effort and money in
our present plan of abstracts in our current botanical journals, in
this country as well as in Europe. The same contribution is re-
viewed in from five to ten or more journals. To have the journals
for their original contributions, we are obliged to pay for all this
reduplication of abstracts, or if we take the journal for the ab-
stracts chiefly, we must subscribe to a dozen or more and still pay
for this needless duplication.

It has occurred to me that if we cannot have one complete
Abstract Journal, perhaps the managers of our present journals,
in this country at least, might arrange for a classification and
division of abstracts and thus avoid duplication. For example
one journal might confine its abstracts to morphology and phys-

Meeting of Thursday, October i8

5

iology, another to taxonomy, another to plant pathology, and so
on.

I do not think an Abstract Journal for American botany alone
would meet the situation. Nearly every botanist in America now
has access to nearly all, or all, of American publications. What we
need is a journal which is complete and will give us suitable ab-
stracts of botanical contributions (original) from all parts of the
world no matter what language the original is published in. I
believe practically all of our botanists in this country would be
willing to pay $10 a year for such an Abstract Journal. I think
bacteriology should be excluded, for bacteriologists will have their
own abstracts, and I would exclude abstracts of remedial measures
for plant diseases.

I believe such an Abstract Journal should aim to be the organ
of communication for the botanists of the world. I believe it
should admit abstracts in at least three different languages
(English, French and German) according to the wish of the writer
of the review, unless the journal can be put on a sufficient financial
basis to have a strongly centralized organization.

I doubt if the Torrey Botanical Club alone could swing such
a proposition. If such an Abstract Journal could receive a suffi-
cient foundation, either from a personal donor, or from some exist-
ing institution, as has been suggested, so that a central bureau in
New York or Washington could prepare the abstracts and publish
the journal, it would render a great service to the botany of the
world and particularly of this country.

Dr. J. Hendley Barnhart said: In this matter, as in some
others, I fear that I am a pessimist. This is a great project, and
an attractive one, and I do not like to seem to be trying to throw
cold water on it. But I fear that those who are backing the ven-
ture do not realize a§ fully as I do the difficulties to be met. If
the proposed abstract journal is to be undertaken, it should be
commenced at once, for the need is great, the time is ripe, and the
opportunity, if lost, may never come again. But the attempt to
abstract American botanical literature only is not worth while;
the world-field is vast; the journal would inevitably exceed in size
and cost the limits estimated in advance. If done at all, it should
be well done. And it must not be forgotten that, as a rule, per-
sons competent to prepare reliable abstracts are also fitted for
original investigation, and it is always difficult to induce an investi-
gator to lay aside his chosen work long enough to write out reports
upon the work of others.

After all, however, adequate financial subsidy at the beginning,
continued as long as necessary, perhaps indefinitely, is the most
essential requirement. Given this, and competent editorial

6 Semi-centennial of Torrey Botanical Club

management, and it can be done. And in that case it surely ought
to he done.

I am strongly in favor of absolutely ''colorless" abstracts as
contrasted with critical reviews. Indeed, in an undertaking like
this, which depends for its success upon universal cooperation,
any other course is sure to lead to friction, then to open breaks,
and eventually to complete ruin. Opinions expressed would be,
after all, merely individual opinions, and liable to greater error
than the views criticized. Form and method of presentation
might be safely, and perhaps should be fearlessly, criticized; but
criticism of statements of facts and soundness of theories should
be studiously avoided. Each abstract should attempt to mirror,
in miniature, the contents of the book or paper abstracted. An
abstract journal is no place for reviews.

Dr. Alfred Gundersen inquired if it would be practicable
for the editors of the proposed journal to cooperate with such pub-
lications as the International Catalogue of Scientific Literature or
the Botanisches Centralblatt. In France, in several cases, the same
botanist has been reporting to both of these.

,Dr. Marshall A. Howe remarked that the matters of pub-
lishing more reviews and abstracts in Torreya and of increasing
the number of titles relating to phytopathology in the Index to
American Botanical Literature published in the Bulletin had al-
ready been referred to the Board of Editors for consideration, and
that while the editors had approved moderate increases in these
directions, they had felt somewhat appalled by the magnitude of
the more recent and more ambitious proposals, which, however,
they had thus far discussed without the formality of a regular
meeting of the Board.

Dr. E. W. Olive suggested that the proposed publishing of
Botanical Abstracts might possibly be subsidized, at least in part,
by the two botanical gardens of New York City. In this great
financial capital of the nation, it ought to be easy to secure a sub-
stantial backing for such an important venture and one fraught
with such obvious possibilities for scientific progress. Such a
proposition should readily draw from the commercial world strong
financial support, especially if in the appeal to them the more
economic phases of the broad field of botany were emphasized.
It may be that all that is necessary for the securing of the necessary
funds for the founding of the proposed new journal is that Director
Britton and Director Gager should ask their friends of Wall
Street for their support of the project.

On the motion of Dr. Gager it was voted that a committee
consisting of the Board of Editors and four other persons be ap-

Meeting of Friday, October 19

7

pointed by the President to consider the matter of estabhshing
an American Botanical Abstract Journal.

President Richards stated that he would announce at a later
meeting the names of the four persons to serve with the Board of
Editors to constitute such a special committee.

On the motion of Dr. Gager a vote of thanks to Columbia
University and members of its botanical staff for the hospitalities
of the day was unanimously passed.

Meeting of Friday, October 19

The meeting of October 19 was held at the Mansion, New
York Botanical Garden, beginning at 2:20 P. M., with Dr, John
Hendley Barnhart, Vice-president of the Club, in the chair.
Fifty-four persons were present.

It was announced that President Richards had appointed Dr.
N. L. Britton, Professor R. A. Harper, Dr. C. Stuart Gager, and
Dr. J. Hendley Barnhart to serve with the Board of Editors as a
special committee on a proposed American Botanical Abstract
Journal.

Dr. N. L, Britton, Director-in-Chief of the New York Botani-
cal Garden, in extending a welcome to the Club, referred to the
fact that the existence of the Garden was due to the activities of a
special committee of the Torrey Botanical Club.

The following' papers were then presented :
"The flora of the Isle of Pines, Cuba," by Dr. N. L. Britton and

Mr. Percy Wilson. Illustrated by lantern-slides.
"Observations on the development of Peridermium Cerebrum,''

by Dr. B. O. Dodge and Professor James F. Adams.
"Collecting Fungi at Delaware Water Gap," by Dr. W. A. Mur-

rill. Illustrated by photographs.
"The physiological properties of two species of poisonous mush-
rooms," by Dr. Michael Levine. Questions and discussion by
Professors Harper, Atkinson, and Rusby, and Drs. Gager and
Murrill.

"Flora of the Rocky Mountains and adjacent plains," by Dr. P.
A. Rydberg. Illustrated by maps. (This was a historical
sketch of the work of the various botanical collectors and
writers who had dealt with the flora of the region named, con-

8

Semi-centennial of Torrey Botanical Club

eluding with a summary of the results shown in large book on
the subject written by the speaker and soon to be published.)
"The route taken by Capt. Nathaniel J. Wyeth and Mr. Thomas
Nuttall from the Mississippi River to the Columbia River in
1834," by W. W. Eggleston. (Read by title in absence of the
author. To be published by the U. S. Department of Agri-
culture),

"Statistical studies in Cichorium,'" by Dr. A. B. Stout.

"The origin and development of the lamellae in Schizophyllum

commune/' by Professor James F. Adams.
"The effect of acids and alkalies on the growth of the protoplasm

of pollen-tubes," by Professor Francis E. Lloyd. (Read by

title in absence of the author.)
"Recent botanical exploration in Colombia," by Professor H. H.

Rusby.

At the close of the arranged program. Dr. N. L. Britton ex-
hibited architects ' plans for recent and future constructions in the
development of the New York Botanical Garden. Refreshments
were then served, after which visits were made to the new Rose
Garden and other points of special interest in the vicinity of the
Mansion.

Meeting of Saturday, October 20
The meeting of October 20 was held in the lecture-hall of the
Brooklyn Botanic Garden, beginning at 2:20 P. M., with Dr. C.
Stuart Gager, Vice-president of the Club, in the chair. Seventy
persons were present. Dr. Gager, in behalf of the Brooklyn
Botanic Garden, made a brief address of welcome, after which the
following papers were presented:

"Determination of acidity in plant tissues," by Professor Herbert
M. Richards.

"The osmotic concentration of the tissue fluids of desert Loran-
thaceae," by Dr. J. Arthur Harris. Discussion by Professor
R. A. Harper.

"Early horticultural journalism in the United States," by Mr.
James G. Scott.

"A preliminary report on the Russulae of Long Island," by Dr.
Gertrude S. Burlingham. Illustrated by specimens. Dis-
cussion by Dr. W. A. Murrill and Mr. Norman Taylor.

Meeting of Saturday, October 20 9

"Comparative cultures of seed-plants in desert valley, desert
mountain, and coastal locations," by Dr. D. T. MacDougal.
(Read by title in absence of the author.)

"Life-history studies in Sclerotinia,'" by Dr. F. J. Seaver and Pro-
fessor W. T. Horne. Illustrated by photographs and draw-
ings.

''The vegetation of the Hempstead Plains, Long Island," by Dr.

Roland M. Harper. Illustrated by lantern-slides.
"The vegetation of Montauk, Long Island," by Mr. Norman

Taylor. Illustrated by lantern-slides. (To be published, in

a more extended form, by the Boooklyn Botanic Garden.)
" Weather conditions and crop diseases in Texas," by Dr. F. H.

Blodgett. (Read by title in absence of the author.)
" Inheritance of height in peas," by Dr. O. E. White. (Read by

title in absence of the author.)
" Parthenocarpy in cucumbers," by Dr. A. F. Blakeslee and Mr.

P. A. Warren. Illustrated by lantern-slides.
"Trimorphism and insect visitors of Pontederia,'' by Dr. Tracy

E. Hazen. Illustrated by lantern-slides.

Dr. Marshall A. Howe, in behalf of Dr. N. L. Britton, chair-
man of the Committee on the Fiftieth Anniversary, reported that
subscriptions to the special Anniversary fund amounted, up to
date, to $1,854, with about one third of the enrolled members
subscribing.

On motion of Professor Robert A. Harper, a vote of thanks was
extended to the Director and other officers of the Brooklyn Bo-
tanic Garden for their hospitality in connection with the Club's
Anniversary.

Professor Herbert M. Richards, in moving a special vote of
thanks to the chairman of the Committee on the Fiftieth Anni-
versary, made the following remarks:

Mr. Chairman: I should like to ofifer another motion for a vote
of thanks and at the same time to make a few remarks. If I may
be permitted I will begin in a reminiscent vein. It has just oc-
curred to me that I have now attained my majority as one of the
botanical group in New York City. It was twenty-one years ago
this month when I attended my first meeting of this Club. That
same autumn a little dinner was given to some of the newcomers, of
whom I was one. I remember that it struck me at the time and

10 Semi-centennial of Torrey Botanical Club

has been even more forcibly impressed on me since that the acti-
vating force, the activating energy, in the botanical development
of this region centered to a preponderating extent in one individual.
You all know whom I mean. It was he who really organized the
botanical department of Columbia University and who was its
first professor. It was he also who was largely responsible for the
organization of the New York Botanical Garden. As members
of the Torrey Botanical Club we all rightly pride ourselves upon
the influence of the Club as a whole in bringing about the founda-
tion of that institution. It is, however, no disparagement to the
devotion of the non-professional members to say that had there
not been a professional botanist back of the movement who was
not only a man of vision, but also one of unusual organizing ability,
the Garden would not have developed at the time or in the manner
in which it did. I think that you will agree with me, Mr. Chair-
man, that even this Garden where we now are was influenced in its
initiation in no small measure by the importance and success of its
elder sister in Bronx Park. Finally, in a less important way per-
haps, the same efficient loyalty to botany has been shown in the
arrangement and the carrying through of this Semi-centennial
Celebration of our Club, which we owe to the chairman of our
special committee.

I would, therefore, like to offer a motion for a vote of thanks to
be extended by the club to Dr. N. L. Britton, Director-in-Chief
of the New York Botanical Garden and Dean of our botanical
fraternity here.

The motion was unanimously carried.

After the serving of tea in the rotunda of the library, the mem-
bers of the Club inspected some of the more attractive and inter-
esting features of the grounds of the Brooklyn Botanic Garden
under the guidance of Dr. C. Stuart Gager and other members of
the Garden staff.

The following members of the Club were present at one or more
of the three meetings held in connection with the Semi-centennial
Anniversary.

Adams, James F. Britton, E. G.

Atkinson, George F. Britton, N. L.

Banker, H. J. Broadhurst, Jean

Barnhart, John H. Burgess, Edward S.

Benedict, Ralph C. Burlingham, Gertrude S.

Blakeslee, A. F. Cassebeer, H. A., Jr.

Boas, Helene M. Chamberlain, Edward B.

Boynton, K. R. Concanon, J. J.

Meeting of Saturday, October 20

Coutant, Mary W.
Denslow, H. M.
Dodge, B. O.
Douglas, H. B.
Enquist, John
Evans, A. W.
Gager, C. Stuart
Graham, Margaret A.
Gundersen, Alfred L.
Harper, Robert A.
Harper, Roland M.
Harris, J. A.
Hazen, Tracy E.
Hollick, Arthur
Howe, Marshall A.
Jones, James H.
Jud, Friedolina C.
Kaufman, Pauline
Keeler, Mrs. L. M.
Levine, Michael
Levy, Daisy
Lorenz, Annie
Mansfield, William

Marquette, William
Medsger, O. P.
Mulford, Fanny A.
Murrill, William A.
Nash, George V.
Olive, E. W.
Richards, Mrs. H. M.
Richards, H. M.
Robinson, Winifred J.
Rusby, H. H.
Rydberg, Per Axel
Scholl, Edith
Scott, James G.
Seaver, F. J.
Small, John K.
Smith, Annie Morrill
Stewart, Grace
Stout, A. B.
Taylor, Norman
Thomas, Mrs. H. Mark
Tweedy, Alice B.
Williams, R. S.
Wilson, Percy

HISTORICAL SKETCH OF THE TORREY
BOTANICAL CLUB

By John Hendley Barnhart

The New York Botanical Garden

The Torrey Botanical Club developed so gradually from a
mere group of associated botanical enthusiasts into a full-fledged
scientific society that it is quite impossible to fix upon an exact
date of origin which might not be honestly disputed. In an early
number of the Bulletin, the beginnings of the Club are traced to
"the summer of 1866";* a few years later the editor remarked in
a footnote "not later than 1865 ";t in later years Dr. Allen, one of
the earliest members, is said to have claimed as early a date as
1858,1 but there is nothing to verify this claim, and it is possible
that he has been misquoted. Dr. Thurber, in his inaugural
address as president of the Club, in 1873, when many if not most
of the original members of the Club must have been among his
auditors, declared frankly: "We have no record of the date of the
beginning of the Club."§

The earliest positive evidence of the existence of an association
which can be definitely connected with our present organization
seems to be a small printed notice preserved in our archives. It
is dated at the office of the American Agriculturist, December 10,
1867, signed by George Thurber and Thomas Hogg, and calls a
meeting of "the Botanical Club, to be held at this office on Satur-
day, the 14th inst., at 2 o'clock P. M. A full attendance is de-
sired, in order that final arrangements may be made for the festival
of the 20th." Presumably the meeting thus called was duly held,
for on the 20th the Club gathered, with various botanically dis-
tinguished guests from out of town, to enjoy a supper at the Astor
House. The occasion was the celebration of the fiftieth anni-

* Bull. Torrey Club i: 45. 1870.
t Bull. Torrey Club 4: 26. 1873.
} Bull. Torrey Club 27: 552. 1900.
§ Bull. Torrey Club 4: 26. 1873.

12

Barnhart: Historical sketch

13

versary of the presentation (December 22, 1817) by Dr. Torrey, to
the Lyceum of Natural History, of the manuscript of his catalogue
of plants growing spontaneously within thirty miles of the city of
New York. An account of the celebration was published at the
time in the American Naturalist,"^ and in the Bulletin, a few years
later, it is clearly stated that it "still further united the members,
and the present organization was effected."! It is at least from as
early as December 20, 1867, therefore, that we may without dis-
pute date the establishment of the Club; and it is that date of
which we are now (somewhat prematurely) celebrating the fiftieth
anniversary. Unfortunately, no complete list of those present at
the meeting of organization has been preserved, and if we do not
count the guests, only eleven members of the Club are positively
known to have been there; it is probable, however, that there were
fifteen or more.

For the years 1868 and 1869 there are no records of the Club
extant, although regular monthly meetings seem to have been
held throughout the year, both winter and summer. The speaker
has in his possession an original written notice referring to the
meeting of February, 1868, which reads: "The meeting of the
Botanical Club is unavoidably postponed until Friday eve. next,
Feby 21st." The same year, June 30, the Club lost its first mem-
ber by death: William Wallace Denslow, "one of the earliest, most
enthusiastic, and, with the disadvantage of feeble health, one of
the most indefatigable."!

The beginning of the year 1870 found the Club still a very
informal association. It had no written constitution, no officers,
no formal list of its membership. It was even without a name,
being known to its members familiarly as "the Club," or more
formally as "the Botanical Club," and to outsiders as the "Bo-
tanical Club of New York." At this time William H. Leggett,
one of the earliest and most faithful of the founders, started, as a
private venture, a modest four-page monthly sheet to which he
gave the name "Bulletin of the Torrey Botanical Club." Of
course this journal, which was the first botanical periodical in
America, was issued with the approval of the Club, but the entire

* Am. Nat. 2: 41-47. 1868.

t Bull. Torrey Club i : 45. 1870.

X Bull. Torrey Club i: 45. 1870.

14 Semi-centennial of Torrey Botanical Club

financial responsibility rested with the editor for many years,
and American botany will ever be indebted to the memory of
Leggett for the powerful and helpful influence thus exerted by him
upon its development at a critical period. It is not by accident
that even the latest volume of the Bulletin bears upon its title-
page the inscription ''Founded by William Henry Leggett, 1870.'*

The name "Torrey Botanical Club" made its first appearance
in public upon the first page of the first number of the Bulletin,
and it is a tradition among us that this name was selected and
applied to the Club, by the editor, in order to have what he re-
garded as a satisfactory name for his periodical, and was thereupon
accepted without question (except for mild protest on the part of
the modest president, Dr. Torrey) by the other members. In any
event, and however it originated, the name "stuck," and has
never been altered to this day, in spite of the very different kind of
associations to which the name "Club" is now commonly applied.

The first list of officers and members was published in the
Bulletin for December, 1870. The editor mentions that "the
association is rather informal, and somewhat fluctuating," and
apologizes for any consequent "errors and deficiencies." The
list (including W. W. Denslow, mentioned as already deceased)
comprises thirty names; and these persons have ever since been
regarded as the founders of the Club, although it is certain that
not all of them had been members from 1867. There is one notable
omission from the list; namely, Thomas Hogg, whose name was
one of those signed to the printed call of December 10, 1867 (as
already mentioned), and who was certainly a member of the Club
both before and after the date of the printed list, so that the
omission of his name was probably an oversight. The addition
of his name makes the number of "founders" (that is, members
prior to 1 871) thirty-one. When Dr. Timothy Field Allen died
in 1902, it was supposed by most of the members of the Club that
he was the last surviving founder; when James Hyatt died in 1904,
it was stated in Torreya that he was the last. James Sheldon
Merriam, however, did not die until 1908, and at least two of
those whose names appear on the list of December, 1870, are still
living. These are Charles Belknap Gerard, now of Muskogee,
Oklahoma, and Rev. Dr. Herbert McKenzie Denslow, who has

Barnhart: Historical sketch

15

again, after a long interval, been one of our fellow-members for
the last seven years.

In 1 87 1 the Club decided upon incorporation, desiring, as it
was facetiously expressed by Dr. Thurber, to "enjoy the privilege
of sueing and being sued." An act of incorporation was passed
by the Assembly of the State of New York, April 21, 1871, but it
was seriously defective, and the Club refused to accept it and
failed to organize in compliance with its provisions. The first
defect, for which it is not unlikely that Dr. Torrey was responsible,
was that the corporate name was given as the "New York Bo-
tanical Club." The other defect was a mere reflection of the
corrupt politics of that period, and consisted in the inclusion
among the incorporators of two members of the infamous "Tweed
ring." To remedy the defects the act was amended by the legis-
lature, April 29, 1872, but even then the Club was very slow to
effect organization under its provisions. The charter was ac-
cepted unanimously at the meeting of January 7, 1873, and a com-
mittee appointed to draft a constitution and by-laws; but these
were not adopted until March 25, and the first officers were not
elected in accordance with them until April 29.

Meanwhile, on the tenth of March, the beloved president,
whose inspiring influence had brought the Club into existence,
and whose name it bore. Professor John Torrey, had breathed his
last, and the Club had become a monument to his memory. Few
are the botanists now living who remember Torrey, but his kind-
ness, his gentleness, his patience, his earnestness, his scholarship — •
these still seem vividly real, even to us who know them only by
the recorded testimony of those who both respected and loved
him. The Torrey Botanical Club could bear no name more de-
serving of honor.

The first president elected under the provisions of the new con-
stitution was Dr. George Thurber, well known as a student of
grasses. He had been botanist to the Mexican Boundary Survey,
and first head of the department of botany at the Michigan agri-
cultural college, before coming to New York; and he was at this
time, and for many years afterward, editor of the American Agri-
culturist. Fortunate it is that his inaugural address, rich in
reminiscence, was printed in full in the Bulletin. The vice-presi-

16 Semi-centennial of Torrey Botanical Club

dency, a new office at this time, was filled by the election of Dr.
Timothy Field Allen, to whom tradition credits the first suggestion
looking toward the formation of the Club.

Dr. Thurber's presidency covered a period of about seven
years. The meeting-place of the Club continued to be the Her-
barium of Columbia College, with which Dr. Torrey *s memory
was so indissolubly associated. The Bulletin grew from a four-
page to a twelve-page monthly, and the scope of the papers pub-
lished broadened noticeably. In this connection it may be re-
marked, however, that although the founders of the Club were
mostly collectors, and their efforts were primarily devoted to the
botanical exploration of the vicinity of New York City, it is evident
that their work was limited only by the meagerness of their knowl-
edge and the narrowness of their opportunity. Their interest in
botany was as broad as the science itself, and their concept of the
science no narrower, at least, than that of their contemporaries.
The early pages of the Bulletin were devoted chiefly, it is true, to
placing upon record stations for the flowering plants of the local
flora; but even before the end of the first volume there was an im-
portant illustrated paper on the structure of the flowers and fruit
of Spirodela, and within a few years the taxonomy of the lower
plants began to occupy a conspicuous place. There is no reason
to believe that, from the very beginning, any botanical paper was
ever excluded from the pages of the Bulletin because foreign to its
field.

The need of a publication which would serve to assist corre-
spondence between American botanists was filled by the appear-
ance in the Bulletin for November, 1873, of a botanical directory
for North America; additions and corrections were published in
the Bulletin from time to time, and two supplements appeared
separately; in 1878 a new edition was issued in pamphlet form.
The reestablishment of Cassino's "Naturalists' Directory"
rendered further efforts in this direction superfluous. Dr. Thur-
ber was followed in the presidency, in 1880, by John Strong New-
berry, professor of geology at Columbia, and famous as a palaeo-
botanist. Professor Newberry was the president of the Club for
ten prosperous years — although the success of the organization
then, as before and since, has been due rather to the faithful and

Barnhart: Historical sketch

17

correlated labors of devoted members than to the efforts of any
one man. This decade saw many changes, recorded in and in
some cases reflected by the Bulletin. The history of the Bulletin
before and during this period is admirably summed up in a report
printed in the second number of the seventeenth volume.* The
journal which had been established as a private venture, and main-
tained as such for twelve years, was formally taken over by the
Club at the beginning of the year 1882, and an associate editor
chosen; just in time, for "the morning of April 11 witnessed the
death of the genial, talented and earnest editor." Succeeding
years saw a rather kaleidoscopic change in the editorial board —
there were ten different members in eight years, the largest number
at any one time being six — yet the publication showed steady
growth and improvement. In 1886, under the editorship of
Elizabeth Gertrude Britton and Frederick James Hamilton Mer-
rill, an index to recent American botanical literature was under-
taken, which greatly enhanced the value of the Bulletin to its
readers, and has been maintained in modified form until the
present time.

In April, 1888, appeared the Preliminary Catalogue of Antho-
phyta and Pteridophyta within one hundred miles of New York
City, based upon the work of the members of the Torrey Club up
to that time. It was a mere check-list, intended as a manual to
be used by members for manuscript records of further work, but
it was much too extensive for publication in the Bulletin, and was
issued separately in pamphlet form. Before the end of the year,
however, the need for a regular series of Club publications in
monographic form had made itself felt, and the establishment of the
Memoirs of the Torrey Botanical Club had been authorized; the
first number made its appearance the following May.

At about the same time the Club began the promotion of a
project for a botanical garden. An appeal for such an institution
in New York City was adopted January 8, 1889, and distributed
with the number of the Bulletin for that month. The effort met
with various setbacks, and the plans required much modification,
but it was the movement inaugurated at this time that eventually
resulted in the establishment of the New York Botanical Garden
in Bronx Park.

* Bull. Torrey Club 17: 48-52. 1890.

18 Semi-centennial of Torrey Botanical Club

It was in the fall of 1888, too, that the Club began to hold
meetings twice instead of once each month. At first one meeting
was called the ''regular" one and the other the "adjourned" one,
but at the end of the following year (December 10, 1889) a con-
stitutional amendment made the distinction unnecessary.

During the decade, 1880 to 1889, the Bulletin had more than
doubled in size, the Memoirs had been begun, and the active mem-
bership of the Club had increased to more than twice its former
size. Insufficiency of funds interfered with the development of
the Club's activities then as it has ever since; but this very need
of financial aid furnished a stimulus to further effort.

In January, 1890, Hon. Addison Brown was elected president.
Unlike his predecessors, he was never a professional botanist, but
as an amateur had long devoted as much time to his favorite
science as could be spared from the responsibilities of his judicial
career. He had been vice-president for many years, even during
Thurber's presidency, and his elevation to the highest office in the
gift of the Club was but a recognition of his faithful interest in its
welfare. His services in this office were retained for fifteen years,
and terminated only by his insistence upon retirement.

From the beginning of the year 1889, Nathaniel Lord Britton,
then instructor in geology and botany (there was at that time no
department of botany) at Columbia University, was the editor-in-
chief of the publications of the Club, and his invaluable services
in that capacity continued for nine years. The Bulletin had long
held a conspicuous place in American botany, and its prestige was
now further strengthened. The reputation of the Club and its
editor grew together, and interacted upon each other. Professor
Lucien Marcus Underwood, Dr. Britton's successor as professor
of botany at Columbia, also followed him, two years later, as
editor, and so served for five years, 1898 to 1902 ; the present speak-
er's first two years of editorship, 1903 and 1904, coinciding with
the last two years of the presidency of Judge Brown.

The summer of 1891 was made notable in our history by the
organization of the Scientific Alliance of New York, with the
Torrey Botanical Club as one of its constituent societies. This
cooperative scheme proved of mutual advantage. The Club
benefited by it no less than the others, and remained a member
throughout the sixteen years of the Alliance's continuance.

Barnhart: Historical sketch

19

For the first thirty years of its existence, the headquarters of
the Club had always remained at Columbia University. At first
the meetings were held at the herbarium and afterward, when at
last that was outgrown, the Club met for years in Hamilton Hall.
In the summer of 1897, Columbia removed from the Madison
Avenue and Forty-ninth Street location to the new site on Morn-
ingside Heights, and the Torrey Botanical Club at that time trans-
ferred its herbarium and changed its meeting-place to the College
of Pharmacy, at 115 West 68th Street.

At the beginning of the year 1900 the Club assumed the pub-
lication of the Card Index of American botanical literature. This
had been issued for the preceding six years by the Cambridge
Botanical Supply Company, but had merely been reprinted by
them from the pages of the Bulletin, and it seemed only reasonable
that the Club under whose supervision the catalogue was prepared
and first printed should also issue it in card form. The Card Index
thus became the third series of Club publications.

Until 1900 it had been customary for the Club to hold all its
meetings in the evening. The first meeting of May, in that year,
however, was held in the afternoon at the New York Botanical
Garden; the Club joined with Section G (Botany) of the American
Association for the Advancement of Science in its celebration of
"Torrey Day" at the Garden, June 27,* and after the summer
vacation began the custom, continued until this day, of holding
one of the two meetings each month at the Garden, and in the
afternoon instead of evening.

The commencement of the new century was marked by several
important changes. The Bulletin had grown until its annual
volume comprised nearly '700 pages and many plates, and the
pressure for publication of technical papers tended to exclude
brief communications and those of a popular character. A new
monthly journal was therefore established by a vote of January 8,
1901, under the editorship of Dr. Marshall Avery Howe, and the
first number of Torreya made its appearance before the end of the
same month. The following year the publication of the proceed-

* The historical papers read on this occasion were published in the Bulletin (27:
540-565. 1900) ; the one by Professor Burgess on " The work of the Torrey Botanical
Club " was prepared with much care, and contains various details of the Club's history
which are not repeated here.

20 Semi-centennial of Torrey Botanical Club

ings of the Club was transferred to Torreya from the Bulletin, but
otherwise its scope as "a monthly journal of botanical notes and
news" has remained unchanged under successive editors.

At the same meeting which authorized the establishment of
Torreya, the Club voted to present its herbarium, subject to cer-
tain conditions, to the New York Botanical Garden; and at the
following meeting the privilege long enjoyed by Columbia Uni-
versity of incorporating Torrey Club exchanges into its library
was transferred to the library of the Garden.

Judge Brown's long term in the presidency was followed by the
election of Dr. Henry Hurd Rusby, who held the office for the
seven years 1905 to 1911. In spite of this comparatively long
tenure, however, the tendency has been, perhaps more from acci-
dent than by design, toward rotation in office, and during the past
twelve years there have been four presidents and five editors. I
shall not dwell upon these later years, for many of my hearers have
been familiar with their history and contributed in an important
measure to it; moreover, it is so fully recorded in printed form that
my omission of it need not embarrass the future historian. The
expansion of the Club from a purely local association to a body of
almost national scope can scarcely be emphasized, however, by
anything more than the wide geographic distribution of our present
active membership, and the fact that our main editorial office is
now in New Haven and our editor a professor in Yale University.

About the beginning of the year 1905, the Club began to hold
its evening meetings at the Museum of Natural History instead
of the College of Pharmacy; and some two years later, upon the
dissolution of the Scientific Alliance, the Club joined with the
other members of the Alliance in becoming affiliated with the New
York Academy of Sciences. Organic union with the Academy is
not close, but the Club has a representative in the Council of the
Academy, and the Club's meetings are announced in the Acad-
emy's weekly bulletin.

Time fails me, on an occasion like this, to refer in detail to the
vast amount of valuable scientific work accomplished by members
of the Torrey Botanical Club, and presented in our meetings or
published in the Bulletin, the Memoirs, and Torreya. Nor have I
found an opportunity to mention, even by name, the many who

Barnhart: Historical sketch

21

have devoted years of faithful service to the Club's interests, as
vice-presidents, secretaries, treasurers, curators, librarians, asso-
ciate editors, and members of important committees. Of one
standing committee, however, I feel that I must speak.

The early work of the Club was, as already pointed out, largely
in the field. At first Manhattan Island furnished many interesting
localities for rare plants; but, with the growth of our metropolis,
the wild conditions that so delight the field-worker were pushed
farther and farther from the center of the city and with this in-
creased inaccessibility and an accompanying diversification in the
interests of the members came a decline in the field-activity of the
Club. All through its history, however, the Club has clung to the
idea that it was its duty to arrange field-meetings for those who
desired to avail themselves of such a privilege. There have been
times when the field-activities of the Club seemed on the verge of
extinction; but always some one has been found to serve on the
field-committee, and the present committee has in the past few
years seen a renewal of interest under the stimulus of its efforts,
particularly those of its chairman. May the Club never fail to
retain a strong hold upon the soil from which it sprung!

REMINISCENCES

By Herbert McKenzie Denslow

The General Theological Seminary

The Rev. Dr. Denslow spoke somewhat as follows:

Mr. President and Fellow-Members of the Torrey Botanical
Club : I am quite aware that I appear to-day as a relic. When I
tell you that I attended early meetings of the Torrey Club while a
school-boy in Brooklyn, you will readily understand that there will
be little of scientific accuracy in my recollections of that distant
time. That I was allowed to attend the meetings in the Her-
barium at the School of Mines was due in part to the fact that my
uncle, W. W. Denslow, was a member of the Club and in part to
the great kindness of Dr. Torrey. I was present at the dinner on
December 20, 1867, but I recall distinctly only that Dr. Gray was
present, as well as Dr. Torrey, and that the occasion was most
impressive to my boyish imagination. It was my first function
of that sort and I probably exaggerate the number present; but my
memory has always reported a long table and a goodly company.

It is to my uncle that I owe my introduction to botany. I
made many field-excursions with him and his friends and worked
in holiday times on his herbarium. My beginnings of botanical
knowledge were thus practical and concrete. Whether there is
pedagogical suggestion in this, I do not assert. Probably a cer-
tain intensity of interest and application, which is a family trait,
contributed to my early enthusiasm. Certainly I gained a life-
long interest in the study of plants; and this avocation has not
only contributed much of pleasure but has helped distinctly in
shaping my mental life.

It was for little more than three years that I was able to attend
with some regularity the meetings of the Club. Then college life
in New Haven, followed by teaching and professional study, kept
me fully occupied elsewhere. I bound up my few volumes of the
Bulletin and found scant time for botany. Still ' I continued to

22

Denslow: Reminiscences

23

collect and exchange, having inherited my uncle's duplicates and
some of his correspondents, until my herbarium became too large
to be lodged conveniently in a rectory. I sold it for a nominal
sum to Hobart College, reserving only the Orchidaceae. But I
continued to study this family, from time to time, in the midst of a
busy parochial life. After being away from the vicinity of New
York for more than twenty years, I came to my present position
in 1902 ; and it was a great pleasure, after getting fitted to my new
harness, to renew my active association with the Torrey Botanical
Club. It is seldom that I can attend a meeting, but I have and
read its publications; and you will readily understand that I get
to the Botanical Garden as often as I can and that I find there
always the kindest welcome from all whom I meet.

TORREY BOTANICAL CLUB REMINISCENCES

By Nathaniel Lord Britton

The New York Botanical Garden

I gladly contribute reminiscences of the years immediately
following my election to the Club in 1877, while I was a student
in the School of Mines, with especial reference to members of the
Club known to me during that period.

Dr. George Thurber was president in 1877, and for many years
afterward the meetings were held at the herbarium rooms of
Columbia College, at 49th Street and Madison Avenue. Dr.
Thurber, long editor of the American Agriculturist, had a fertile
fund of botanical information of all kinds and stimulated discus-
sion on nearly every topic presented at the meetings. Mr. P. V.
LeRoy, who had been an assistant of Dr. Torrey, was curator of
the Torrey Herbarium and he carefully guarded the collection
during his incumbency and increased it by the purchase of many
valuable sets of plants.

I entered the School of Mines in 1875. Dr. Torrey died in
January, 1873. While I was being prepared for the School of
Mines at the Staten Island Academy, I was taken on several
occasions to Columbia College by my father to see Dr. Drisler,
and on one of these visits I was told where the herbarium was lo-
cated and a professor was pointed out to me as the renowned Dr.
Torrey. This did not make much impression on me as a boy, but
later, when becoming familiar with Dr. Torrey's portrait, I recol-
lected the incident. Mr. J. J. Crooke, then resident of Great
Kills, Staten Island, an all-around naturalist who accumulated
large collections, induced my parents to send me to the School of
Mines, and told me much about Dr. Torrey. Mr. Crooke was
Treasurer of the Club for a period.

Mr. William H. Leggett, who founded the Bulletin of the Club
in 1870, was still its editor; he was an enthusiastic field and her-
barium botanist, a highly successful teacher, and a fine linguist

24

Britton: Torrey Botanical Club reminiscences 25

who inspired all his associates. His herbarium forms the principal
part of the nucleus of the Local Herbarium of the Club, which has
in later years been expanded into the Local Herbarium of The New
York Botanical Garden. Dr. Timothy F. Allen, a physician of
prominence and a man of delightful personality, had already
commenced his long-continued studies in Characeae, and was in-
terested in obtaining specimens of these plants from all parts of
the world. Mr. J. M. Wilbur, Secretary of the Club for many
years, seldom missed either a field or an herbarium meeting. Dr.
J. W. Barstow, resident of Flushing, was a frequent attendant at
meetings and brought in many specimens. Mr. John L. Wall,
an active microscopist, attended many field-meetings and sub-
sequently was one of the founders of the New York Microscopical
Society; I was closely associated with him for a period of years.
Mr. William Bower, who had a garden of native plants at Newark,
New Jersey, and was keen on their cultivation, rarely missed a
field-meeting. Mr. W. R. Gerard, subsequently editor of the
Bulletin and a man of great erudition, was pursuing his mycological
studies. Messrs. Isaac Buchanan and James Hogg, both nursery-
men, occasionally came to meetings and brought specimens of
cultivated plants; Thomas Hogg, brother of James, joined the
Club in 1882, after returning from Japan, whence he sent many
Japanese shrubs and trees for their first introduction into the
United States; he subsequently became a Vice-president of the
Club and was active at its meetings; I saw a great deal of him in
later years. Professor James Hyatt was already an enthusiastic
microscopist, and I well remember collecting diatoms with him in
marshy grounds now occupied by the systematic herbaceous
plantations of The New York Botanical Garden. Mr. C. F.
Austin, who resided at Closter, was in the midst of his important
bryological work, but it was never my good fortune to meet him ;
after his death in 1880, Dr. Newberry sent me to Closter to buy
his bryological collections, which thus became the nucleus of the
great moss herbarium subsequently built up by Mrs. Britton at
The New York Botanical Garden. Mr. M. Ruger was a regular
attendant at all meetings and a diligent collector; a considerable
number of his specimens are preserved in the local herbarium ; one
of the first field-meetings I remember was under his guidance at

26 Semi-centennial of Torrey Botanical Club

Train's Meadows, Long Island, especially to collect Scleria ver-
ticillata, first found by him there in 1874. Dr. Denslow we still
have with us; Mr. H. A. Cassebeer, Jr., will attend the dinner
this evening.

Professor D. C. Eaton was vigorously pursuing his fern studies
at Yale and sent communications for the Bulletin; I do not re-
member seeing him at any meeting, but I visited him at New
Haven. Mr. G. W. Wright and Mr. William Chorlton, both of
Staten Island, contributed much to the interest of meetings by
bringing specimens of both wild and cultivated plants. Judge
Addison Brown had already commenced his active participation
in the affairs of the Club and his important influence on American
botany by forming an herbarium, and for a number of years
attended nearly every field-meeting. Mr. J. H. Redfield, of
Philadelphia, made occasional contributions to the Bulletin; I do
not think that he was ever present at a meeting which I attended,
but I visited him later in Philadelphia. Professor Joseph Schrenk,
almost our only plant anatomist and physiologist of those years,
was at the Hoboken Academy and subsequently at the College
of Pharmacy; he was a keen observer, and I recollect searching for
Schizaea with him at Tom's River for a day without finding any,
but he took in a sod of Drosera for experimental purposes, and
shortly afterward found that he had Schizaea in the same sod!
Mr. Cornelius Van Brunt was pursuing studies of diatoms and his
collections subsequently came to the New York Botanical Garden ;
he was present at a number of field-meetings and helped found the
New York Microscopical Society.

Professor Alphonso Wood, of the College of Pharmacy, re-
sided at West Farms, where I once visited him with others of the
Club and the party walked up the Bronx Valley through the whole
length of the present New York Botanical Garden reservation, on
which occasion I first saw the pot-holes, located near the west end
of the present Boulder Bridge, which I described in the Transac-
tions of the New York Academy of Sciences in 1881; it was this
trip that gave me my first knowledge of the natural beauties of the
Bronx Valley. Professor Wood had at that time about completed
his long series of noteworthy text-books. Dr. O. R. Willis, a
diligent student of the local flora, resided at White Plains and at-

Brixton: Torrey Botanical Club reminiscences 27

tended many meetings of the Club, contributing notes and speci-
mens. Dr. L. Schoeney, a practicing physician, was in those
years perhaps the most constant attendant at both field and her-
barium meetings, and continued his interest over a long series of
years. Dr. Arthur Hollick, who was elected in 1877, at the same
time I was, is one of the few living persons who has maintained
continuous membership in the Club since that time; he was active
with me in the study of Staten Island plants and our collections
of those years are preserved in the herbarium of the Staten Island
Association of Arts and Sciences.

Dr. Newberry, then in the midst of his paleobotanical studies,
occasionally came to the meetings, being elected to membership in
1878, and became President to succeed Dr. Thurber in January,
1880. He had an enormous fund of botanical information and
was able to throw light on almost every topic brought up for con-
sideration; as his assistant in the School of Mines for a series of
years subsequent to 1879, it was my good fortune to be closely
associated with a naturalist of his renown. Miss Elizabeth G.
Knight (subsequently Mrs. Britton), elected in 1879, had already
absorbed enthusiastic interest in plants and animals, through
association with Dr. Newberry and with Professor Edward H. Day,
of the Normal College, who was elected to the Club in 1880 and
who subsequently participated in many field-meetings. Pro-
fessor Day was a most jovial naturalist, a pupil of Huxley, and of
very broad information. He led a field-meeting once into Mon-
mouth County to see Lygodium, and perpetrated a pun in wanting
to know why that tramp was like rum, which he expounded by
maintaining that it was a sandy cruise (Santa Cruz) ! As a popu-
lar professor of the Normal College, Professor Day is remembered
with affection by a large number of students. Mr. Eugene P.
Bicknell, who was elected in 1880, had already commenced his
critical studies of the local flora, especially of what is now the
Borough of the Bronx, and had begun the formation of his exten-
sive herbarium ; he attended herbarium meetings with much regu-
larity and contributed frequent notes and specimens. Mr.
William H. Rudkin, subsequently and for many years Treasurer
of the Club, became a member in 1878, and for a long period con-
tributed important aid to the work of the organization, attending

28 Semi-centennial of Torrey Botanical Club

both field and herbarium meetings and aiding Mr. Leggett in the
publication of the Bulletin; both Mrs. Britton and I were with
Mr. and Mrs. Rudkin much during those years.

My first botanical contribution was made to the Club in Sep-
tember, 1877, and printed in the October Bulletin of that year; it
is upon Rarer Plants of Staten Island, including a note on the
sensitive stamens of the Opuntia of what is now Crooke's Point.
Was this a premonition of my subsequent interest in Cactaceae?

TORREY BOTANICAL CLUB REMINISCENCES

By Arthur Hollick

Staten Island Association of Arts and Sciences

If I remember correctly, it was in 1876 or 1877 that Doctor
Britton and I joined the Torrey Botanical Club. We were class-
mates in the Columbia College School of Mines at the time and
had collected plants together in a more or less desultory way.
Each of us had a small local herbarium and we did the best we
could to identify and name our specimens with the aid of Gray's
Manual. The only instruction we received in botany was one
lecture a week during one term, by Professor Newberry, who also
lectured on zoology, paleontology and geology. Practically, he
was professor of "natural history." There was no laboratory
work of any kind and we were left entirely to our own devices so
far as assistance in securing botanical information or knowledge
was concerned.

We had heard vague rumors to the effect that somewhere in
the recesses of the old college buildings an herbarium was housed,
and after making several inquiries we finally located it, and found
Mr. P. V. Leroy in charge as curator. I believe his salary was
paid by Mr. John J. Crooke and not by Columbia. Certainly
Columbia made no use of the herbarium. The Torrey Botanical
Club met there and in this way we became acquainted with some
of the members and soon ventured to apply for admission.

I shall never forget the first meeting I attended. I felt that
I was under indictment for the crime of being a young man. There
were no young botanists in those days. Many of those whose
acquaintance I made at these early meetings were as old as I am
now and others were older, and that was forty years ago. I recall
particularly Alphonso Wood, William H. Leggett, P. V. Leroy,
O. R. Willis, Bowers, Ruger, and several others. I never met Dr.
Torrey, of course, as he died in 1873. No woman had yet been
elected to membership in the Club. Any such innovation would

29

30 Semi-centennial of Torrey Botanical Club

have been unthinkable at that time. We brought specimens to
the meetings, discussed them, helped each other to identify them,
described how, when and where they were collected, and then
arranged informally for a field-meeting — perhaps for more than
one — before the next meeting of the Club. If I remember correctly
the dues were one or two dollars a year. Subscription to the
Bulletin was a dollar. It was not published by the Club, but by
Mr. Leggett personally. There were no expenses, except in con-
nection with the small number of postal cards to announce the
meetings. The money in the treasury was mostly spent for re-
freshments, and after each meeting we had a pleasant, sociable
time, drinking coffee and eating cakes and sandwiches and occa-
sionally fruit when in season.

Attending meetings in those days was not so easy as it is now —
I mean for out-of-town members. I lived at Port Richmond on
Staten Island. The last boat to the island was at 9 P. M. I used
to take the midnight train on the Central Railroad of New Jersey
at Liberty Street, get off at Bergen Point Station, walk three
quarters of a mile to the shore of the Kill van Kull, wake up a
man who lived in a little shanty there, and hire him to ferry me
over to Staten Island in a rowboat, arriving home about 1 130 A. M.
Sometimes, in winter, the trip was not a comfortable one; but I do
not recall that I ever thought it a hardship, and, to the best of my
recollection, I think I merely regarded it all as a matter of course.

I still live on Staten Island, but I can attend this meeting in
the Bronx to-day far more easily and with less waste of time in
coming and going than was formerly the case when I attended the
meetings held at 49th Street.

A SKETCH OF THE LIFE OF COE FINCH

AUSTIN

By Sarah Austin Demarest

^ Englewood, New Jersey

Coe Finch Austin, the subject of this sketch, was born June
20, 1 83 1, at Finchville, Orange County, New York. Grand-
parents on father's side were Abraham Austin, EngHsh descent,
and Mehitable Campbell, Scotch, and on his mother's side,
William Cortright and Jemima Huff, both Hollanders. He was
the second of a family of ten children, born to James C. and
Elizabeth Cortright Austin, thrifty farmers of that period. When
he was quite young his parents moved to Greenville, N. Y., where
the lad entered the public school. He made rapid progress in the
fundamentals. A few years later the family moved to a fine farm
near what was then called Brookfield, now Slate Hill, Orange
County, N. Y., where his early life was spent, much the same as
other boys, assisting on the farm during the summer, and attending
public school in the winter, where he was generally at the head of
his class and at the forefront in the various games played on the
school grounds.

Early in life he manifested an interest in floriculture, and was
his mother's constant companion as she cared for her flower garden
(of which she was extremely fond), this same being often enriched
by choice specimens which the boy gathered from neighbors of
like tastes. Nor was he less interested in arboriculture, and the
lawn of his paternal home was ornamented by trees of many
varieties collected from forest and field, whose generous shade now
gives pleasure and comfort to another generation occupying the
homestead.

In character he was independent and aggressive, and impatient
of restraint. When about eighteen years of age, being repri-
manded by his father for some neglect of duty, he rebelled, and
as a result was compelled to take the world for his parish; he went

31

32 Semi-centennial of Torrey Botanical Club

through the full experience of the ''Prodigal Son," returning after
a few months, if not a sadder, a wiser boy. He partook of the
fatted calf with evident relish, and there was joy in the household
again, there being no elder brother in evidence to mar the occasion.
This experience wrought a wholesome change in his character; he
cut off some of his old companions, and started life on a new track.
Soon after this he took up teaching and lecturing in neighboring
schools during the winter months.

In the early 50's he entered Rankin's Classical School, at what
is now Sussex, in Sussex County, New Jersey. Here he met a
congenial spirit in the person of Mrs. Rankin, a botanist of some
note, and to this chance meeting his choice of life work is un-
doubtedly due. He came from this school a working botanist.
At first he took up the subject in its broadest sense, but after a
time realized that the field was too broad to accomplish much in a
lifetime, and becoming infatuated with microscopic revelations,
he dropped all except mosses and lichens. In his search for speci-
mens of them he was most indefatigable, letting no obstacles,
however formidable, deter him from their pursuit. No distance
was too great, no jungle too dense, no mountain too high or steep,
no toil too great, to turn him aside in his eager search for new forms.
Accompanied by an Indian he would spend whole days in the
forests and field, from early morning till night, with no other food
than berries and roots, which his knowledge of botany disclosed as
of food value. He was heard to say that the botanists have sources
of food supply that the world knows not of.

During the winter of 1856-57, in company with Edward Swift,
of Marathon, N. Y., he toured New England, lecturing on elec-
tricity and chemistry, with apparatus to illustrate, and it was said
by those competent to judge that his brilliant experiments were
the most striking of their kind, and rarely, if ever, surpassed by any
one. He was heard to say that if man ever learned to control elec-
tricity of sufficient power they would see horseless carriages and
the steam engine would be succeeded by electric engines.

Later, in 1857, he accepted a position as school teacher at
Tappan, N. Y. There he met Hannah Campbell, daughter of
David P. Campbell, a farmer, living on the Alpine Road, about
one quarter mile from Closter, N. J., to whom he was married,

Demarest: a sketch of the life of Coe Finch Austin 33

May II, 1858. In the fall of that year he went to Dennisville,
Cape May County, N. J., taught school for a year, and it was while
there that he met in his wanderings through woods and marshes
Charles F. Parker, of Philadelphia. In the early days of their
acquaintance the latter often visited him, and it was he who as-
sisted him in the purchase of his microscope.

Miss Warwick, a resident of Dennisville at that time, told the
writer a few years ago that she had vivid recollection of the lectures
delivered in her town by Professor Austin. He drew large gather-
ings of people from the surrounding country, for the subjects were
new to them, and his experiments were always very successful,
which made the lectures attractive.

In the middle of June, 1859, he, with his family, returned to
Closter. About this time he became acquainted with Dr. Torrey,
and through Torrey 's influence became curator of the Columbia
College Herbarium. He moved, in the autumn of 1861, to an
apartment in the college. During this period he devoted himself
to intensive study; his power of concentration was remarkably
great. He remained at the college until after the spring of 1863,
when he again returned to Closter, again taking up lecturing, going
through the country for miles around, traveling with a little pony
and a carry-all wagon.

In 1865 and 1866 he taught school at Demarest, N. J., one
mile south of Closter, and spent his noon hours in nearby woods,
gathering mosses and other plants, which he often showed to his
pupils, pointing out and' explaining the peculiarities of each.
Boys of neighboring villages were always on the lookout for him,
and, when spied by one, soon there would be a group of them around
him, for his pockets always contained something to interest them.
Those were days when children were not noticed by grown-ups,
as they are to-day, and the attention he paid to them was very
acceptable.

As a teacher he was fond of children, and if they could not
keep up with their classes and showed a willingness to do so, he
would help them after school hours, but he had no patience with
those who could be termed lazy.

His last lectures were delivered at the Englewood Institute in
the winter of 1871-72. His interest in botany increased as time

34 Semi-centennial of Torrey Botanical Club

went on, possibly due to his finding new species of mosses, hepatics,
and lichens. Some time in the latter half of the 6o's a piece of
swampy meadowland, lying due west of what was then the Re-
formed Church Parsonage of Closter, had been cultivated and
seeded down by a neighbor living a little distance away. One day
the pastor's little daughter came rushing to her father excitedly,
saying a man was stealing Freddie's turnips. Rev. Hammond,
for that was the pastor's name, went out to see who the intruder
was. He saw him on his knees, scraping earth with his hands.
As he came near he found it to be the botanist, who joyously ex-
claimed, " I have found a new lichen." He walked with his friend
up to the parsonage, telling about his discovery as they went.
When he reached there he was bubbling over, and he asked Miss
Isabelle, who had caught some of his enthusiasm, what he should
name it and she replied, "Austini." It was this pastor who as-
sisted him in his study of classical Latin, but he had to study out
botanical terms without assistance. He maintained a large cor-
respondence with scientists in all parts of America and Europe,
his knowledge of Latin being a benefit to him, but desirous of
corresponding with a noted German botanist, and being ignorant
of the language, he laid aside everything until he had so far mas-
tered it that he was able to communicate w^th this person.

Many specimens were sent to him to be named, from foreign
countries as well as the United States and Canada, in which he
was so intensely interested that with his beloved microscope he
would work until two or three o'clock in the morning, seeming
oblivious of the passing of time, God giving him to see in the
humble mosses and lichens which the world tramples under foot
oceans of beauty and interest. He seemed to have a contempt for
the riches of this world, his gold mine being the dense forest or the
rugged mountain, rich with his beloved mosses.

I recently received a letter from Dr. J. J. Haring, of Toledo,
Ohio, formerly of Tenafly, N. J., an old friend of his, still living,
from which I quote the following: "I remember many conversa-
tions with your father in his best years — upon botanical matters,
especially in relation to that of mosses, of which he was a close
and enthusiastic student, devoting to them most of his time, of
years not a few. I remember his pride in having the opportunity

Demarest: a sketch of the life of Coe Finch Austin 35

of correspondence with leading authorities on mosses, and his
exchange of rare specimens of them. Especially do I recollect his
enthusiastic announcement that he had discovered varieties un-
known to writers and students in his particular line. . . . Had
his life been spared for some years longer, and could he have been
more favorably situated financially, it is my belief that as an
authority on mosses and lichens he would have been surpassed by
only a few distinguished specialists in his chosen department of
botanical work and study."

Many positions of profit were offered him, but all were rejected
for fear that their acceptance would interfere with his favorite
study, ignoring the fact that his family would be benefited thereby,
yet no one who knew him would think of calling him selfish, for his
knowledge of any subject was theirs for the asking.

He made the most of every day allotted to him, but not always,
in fact seldom, for his own benefit. He was kind-hearted, and
always ready to give a helping hand to those who needed his assist-
ance, invariably without remuneration. He was happy, and fond
of his family, proud of the progress made by his children in their
studies, never refusing, no matter how busy, to help them over
difficult problems.

He accepted his circumstances uncomplainingly, for he was
so infatuated with the love of nature in all forms that if he could
study unmolested he knew no cold, heat, fatigue, or hunger.
Family needs, trials, and troubles he left for others to care for.
He was of a cheerful disposition, and could always find something
to smile about, often on account of the clumsy way some one tried
to do a piece of work, for he could mend a plow, or repair a watch,
as well as any one.

During the last decade of his life he would leave his studies
and make excursions into the country on foot, in search of speci-
mens. To illustrate his intrepidity and fearlessness on these
excursions, the following incident may be given : While visiting a
brother at Haverstraw, N. Y., he expressed a desire to ascend the
"High Tor," a rocky and precipitous eminence of the mountain,
setting back landward near the town. This peak rises to a height
of about 800 feet. Its front is rocky and rises in a succession of
sheer faces of from 25 to 50 feet. A rugged path called the " Deer

36 Semi-centennial of Torrey Botanical Club

Path" affords a comparatively easy means of ascent, the difficult
places being bordered by scant shrubbery, which being grasped
by the hands enabled one to ascend. Here and there on the way
up specimens were added to his shoulder-bag. On reaching the
top, and after viewing the magnificent scenery, he surprised his
brother by handing him the specimens, with instructions to meet
him at a certain point at the base of the mountain, saying that he
would descend the face of the peak. The brother was horrified
at the suggestion and tried to dissuade him from so hazardous an
undertaking, but failed. Arriving at the point of meeting agreed
upon, no sound of his approach could be heard, and no answer was
made to his brother's vociferous call. An approaching thunder-
storm added to the brother's apprehension of disaster. Suddenly
the botanist appeared, loaded with specimens, and a smile on his
face which was intended to convey a rebuke for faint-heartedness.

In 1870 he began the task of mounting in book form his Musci
Appalachiani. This work was carefully and artistically done.
One who did not see the work being done could not possibly form
any idea of the amount of time and painstaking labor required to
arrange each set.

At this time Closter was rapidly building up, and owners of
new homes were desirous of beautifying their grounds with shade-
trees, but met with much disappointment because so many of the
trees died. He, being successful in planting trees and shrubs, was
asked by a neighbor if he would take up the work. He felt sure of
success, and in a modest way began the work, selecting his trees
from woods of the farmers on the outskirts of Closter, until the
demand was so great that he started a nursery. The beautiful
shade-trees stand to-day as monuments to his memory. It was
he who brought pond lilies to this region and planted them where-
ever he found suitable place for them. No matter how hard he
worked during the day he was always ready to handle mosses in the
evening.

He took several trips in the interest of botany. He went to
Ohio to see Mr. Sullivant, and to the White Mountains, his last
trip being to Florida. There in a rowboat, with a negro oarsman,
he ransacked the banks of southern rivers and morasses, looking
for new treasures. He enjoyed remarkably good health until the

Demarest: a sketch of the life of Coe Finch Austin 37

southern trip, for there he contracted a sort of malarial fever, which
undermined his health, and he seldom visited his familiar haunts
on the Palisades after that.

The last time he went there was to meet a prominent man at
Col. Miles's. He wore a new suit, just completed by his wife, and
much against her wishes, for she knew his failing, but he promised
her he would not go botanizing, simply going for a call. On his
return he was tired, and stopped at his father-in-law's to rest,
looking more like a tramp than the gentleman of a few hours before,
and behold he had cut the lining of his coat and used his coat for
a bag to carry mosses — he had found such beautiful specimens that
he could not pass them by!

Although for several months he gradually grew weaker, he did
not give up his work entirely, for he realized he was near the end
of his life's journey, and was anxious to complete the arranging of
several sets of his supplement to Musci Appalachiani, so that they
could be disposed of profitably by his family. He continued until
he was getting the numbers mixed because his mind could not
endure the strain, laying the work aside, unfinished, three days
before his death.

On March i8, 1880, he was called from the scenes of toil and
study. His parents, then residing in Ridgebury, two miles from
Slate Hill, being too feeble to come to Closter, appropriate funeral
services were held in the Ridgebury Methodist Church, and inter-
ment was in the family plotin the cemetery adjoining the church.
He left a widow and six children, one son and five daughters, the
youngest being then seven years old. His widow was called to her
rest December 12, 1916. The six children are still living. They
are all married.

Sarah Elizabeth, wife of Abraham Demarest, Englewood, N. J.

David C. Austin, of Westfield, N. J.

Annie, wife of Walter G. Warner, of New York.

Kate, wife of Henry Scott, of New York.

Marietta, wife of Harry L. V. Warner, of Bloomfield, N. J.

Ella, wife of Edward W. Dorey, of New York.

Large quantities of mosses continued to come to his address
for fully two years, the senders not knowing of his decease. Five
years later, the minister who officiated at his funeral, while travel-

38 Semi-centennial of Torrey Botanical Club

ing on pony-back over the Rockies, alighting to take a rest, found
a man gathering mosses. As he was in a measure interested in
them he opened up a conversation with him in regard to what he
was collecting, and much to his surprise he found him to be an
amateur, who expected to send his unfamiliar varieties to Professor
Austin of Closter, N. J., to be named. Our friend told him there
was no Professor Austin in Closter, whereupon he insisted he had
the name correct and that our friend was mistaken, until told that
such was the fact, as our friend had officiated at the funeral.

Mrs. Isabelle Hammond Demarest, of Closter, N. J., whose
name has been mentioned before, a neighbor of Professor Austin's,
who had known him since the early 6o's, caught his spirit, and
after reading this sketch insisted that the writer was not nearly
enthusiastic enough in her portrayal of her father's life. While
that may be true, the respect and veneration which the writer has
for her father's memory will not permit her to record a single state-
ment which she does not know to be based upon solid fact. His
unostentatious character has been to a certain extent reproduced
in his daughter, and so she has recorded merely the ground-work.
Let the laudatory trimming be added by some other hand!

RECENT BOTANICAL COLLECTING IN THE
REPUBLIC OF COLOMBIA

By H. H. Rusby

College of Pharmacy, Columbia University

There are three reasons why the flora of Colombia is of excep-
tional interest to students of plant distribution:

A. This country contains what might be called the "elbow"
of the Andes mountains; the region where their northern extension
is exchanged for a broad sweep to the east along the Caribbean Sea.

B. Soon after entering southern Colombia, the Andes divide
like the tines of a fork, into three parallel branches. Since these
three ranges are of very considerable height and a large part of
their intervening valleys is elevated but little above sea level,
there results an extreme range of climatic conditions, with a cor-
responding diversity of flora.

C. The country yields a number of important drugs, besides
many other economic products of great interest and value.

Although the Colombian flora has been much studied, this
study has been rather fragmentary than general and we have yet
a great deal to learn regarding the relations between its different
parts. Among the noted botanists who have studied and collected
there, are Humboldt, Zea, Mutis, Triana, Karsten, and Caldas, in
former times. During the eighties, Lehmann collected very ex-
tensively in the south and west. More recently, many small col-
lections have been made, especially by American botanists, be-
sides quite an extensive one by Mr. Herbert Smith, in the vicinity
of Santa Marta.

My personal interest in the study of this flora is of a rather ex-
ceptional character, because of the great amount of work that I
have done upon the flora of neighboring portions of the Andes.
In 1885, I made very extensive collections from Peru southward to
Chile, later traversing the entire length of the Madeira and Amazon
Valley. Thereafter, I maintained a collector in Bolivia for several

39

40 Semi-centennial of Torrey Botanical Club

years, and more recently still have been receiving collections from
the Bolivian Department of Agriculture. During the nineties, I
spent a season collecting in the lower Orinoco region. All of Mr.
Smith's collections are represented at The New York Botanical
Garden, as well as most of those of Dr. Lehmann. There was thus
left an Andean region, occupying most of Colombia, the flora of
which I had not seen and I have for many years desired greatly to
visit it. This desire was increased by reason of my special interest
in medical botany.

During the past season an opportunity was afforded for grati-
fying this desire, when I .was asked to go to Colombia and investi-
gate certain of its drug supplies. Although the work to be per-
formed was of a commercial character, an opportunity was afforded
for extensive botanical collecting, and I brought back with me some
thirteen hundred collection numbers. This collection was princi-
pally the work of my associate, Dr. Francis W. Pennell, of the
Garden staff, who accompanied me.

We left New York on June 27 and I returned on September 29,
so that the entire journey occupied a period of three months and
two days. Almost immediately after reaching the Port of Colom-
bia, we boarded a river steamer and ascended to the head of
steamer navigation on the Magdalena River, a journey which
occupied more than a week. It thus happened that all but fifty-
one days of our time were spent in sailing, when little or no col-
lecting could be done. Quite a number of plants were collected
along the river shore, when the steamer was making prolonged
stops for taking on fuel, or for discharging and receiving freight.
Since the vessel was close to the shore during most of its sailing,
there was good opportunity, by the use of the field-glass, to ascer-
tain the character of the neighboring vegetation. Because of my
previous familiarity with tropical American plants, I was enabled
to utilize this opportunity to excellent advantage.

On leaving the steamer at Girardot, we traveled by mule
directly southward for seven days, getting pretty well up on the
table-land adjacent to the upper Magdalena Valley. We next
secured fresh mules and proceeded eastward, just crossing the
ridge of the eastern range. About a week was spent in collecting
about the summit of this Cordillera. We then returned north

Rusby: Recent botanical collecting in Colombia 41

to Girardot and took railroad train to Bogota. Circumstances
interfered with collection work during my stay at this city, but
Dr. Pennell, then and subsequently, did a large amount of work
within a radius of two or three days' foot travel from the city.
At Bogota, we separated, I returning north to visit the western
mountains, while he descended to the great plains where the
tributaries of the Orinoco and Negro Rivers take their rise. At
Puerto Berrio I left the river and crossed the central range, where
several days were spent. I had intended descending the Cauca
River and possibly getting over into the valley of the Sinu, but
adverse circumstances forced me to change my plan.

Inasmuch as not even that part of the collection received here
has yet been studied, it will be seen that only the most general
statements regarding the flora can be made. Arrangements have
been made by which Dr. Pennell is to remain for several months
in Colombia, visiting and collecting in districts where little botani-
cal work has been done. It is to be expected that his collections
will be very rich and that their study will add greatly to our present
knowledge of the Colombian flora.

The Magdalena River flows through a broad, low, flat valley
for at least half of its length. From the steamer, the mountains
can be seen in the distance on both sides during most of the time.
As we start our journey from the mouth of the river, we can see,
near Santa Marta, the snowy summit of the highest peak in Colom-
bia, said to have an altitude of more than twenty thousand feet.
The river plain is covered with tall and luxuriant grasses, as well
as sedges, and affords excellent grazing. The cattle industry
here is large but is not a tithe of what is possible with an abun-
dance of labor and economic methods. A great drawback to this
industry is the extent to which young cattle are destroyed by the
annual freshets, which occasionally inundate almost the entire
plain. The most conspicuous features of the flora here are mag-
nificent palms (called "palma real" by the people, and yielding
very important useful products), bamboos, and pampas grass.
Up to the time of this journey I had regarded the tree fern as
being the most beautiful representative of the vegetable kingdom,
but I am now disposed to accord this place to the bamboos of the
Magdalena Valley. I can compare one of them to nothing more

42 Semi-centennial of Torrey Botanical Club

fitting than a very fine ostrich plume. Their shape and their
method of drooping is exactly the same, and when one finds a
cluster of them, with the individual fronds arching over from the
center, the effect is indescribably handsome. The pampas grasses
also present a lovely appearance. The entire flower stalk may
reach a height of twenty feet or more. Its lower two- thirds is
very leafy. Then there is an elongated bare peduncle surmounted
by a panicle several feet in length. The branches of this panicle
are exceedingly long and slender, so that the slightest breeze is
sufficient to blow them out in a horizontal position at one side,
giving a remarkably close imitation of a flag. The color of this
flag ranges from light pink to a rather dark purple. All travelers
are captivated by the beauty of this grass, which grows in patches,
rarely of any great extent, throughout the entire river valley.
The clumps of shrubbery that are scattered over the plain belong
very largely to the Mimosa family, especially in the lower part of
the valley. As we ascend, other classes mingle with them and
they become very much more abundant and larger, at length
giving way to a heavy forest growth which extends quite to the
river's edge. In this region, especially after we reach the hilly
section, the river bank is gay with Heliconias of several species.
The inflorescences are of a brilliant red, largely variegated with
bright yellow, and to a lesser extent with blue. Those of one
group are strictly erect, with slender stiletto-like branches, while
those of another are pendulous, several feet in length, and of a
regularly sinuous form. Throughout the greater length of the
river, the trees near the water are largely Cecropias, of a number
of species, and are of very striking appearance. Some have simple
trunks, their huge digitate leaves on very long petioles, and radiat-
ing directly from the summit to form an umbrella-shaped crown,
while others have a few loose and open branches. The trunks
and branches of all are very light colored, appearing whitish at a
distance when the sun strikes them. Most of them have hollow
stems and branches which are inhabited by colonies of fiercely
stinging ants. Back of these ambaibas'' comes a growth of
Ceibas or silk-cotton trees, which are even more conspicuous, and
are stately in their beauty. These trees have tall, straight trunks,
without branches until after they have surpassed the trees around

Rusby: Recent botanical collecting in Colombia 43

them. Usually there is a conspicuous and graceful enlargement
of the trunk at one or more points. The branches are almost
horizontal and often of great length, while the crown is flattened,
thus giving them a peculiar parasol-like appearance.

There are many Bignoniaceous vines, but they are scattered,
this being one of the most conspicuous differences between this
and the shore flora of the lower Orinoco, where there is often a con-
tinuous curtain, miles in length, of brilliantly blooming vines of
this family.

As we ascend the river, the mountain ranges on both sides
steadily approach the shore. Every now and then the river will
take a wide sweep and impinge against the foothills of the moun-
tains, now upon one side and now upon the other. At such places
we can distinguish no characteristic difference between the com-
position of the flora upon the two sides. This flora is very rich
and varied, so that one can scarcely attempt a description of it.
Ingas, Pithecolobiums and other related plants are freely repre-
sented. There are also many Cassias. Toward the upper part of
the river. Acacias become the most conspicuous trees. When we
go ashore, we find Zanthoxylums quite abundant among the shrub-
bery, together with Muntingias and other shrubs and small trees
belonging to the Malvaceae and Tiliaceae. Clitorias are very
numerous and very handsome. Rubiaceous shrubs, herbs, and
small trees are exceedingly abundant, as are herbaceous Euphor-
bias and shrubby Crotons. Shrubby and arborescent Solanums
are in bewildering variety. Large cactuses are seen occasionally
in the lower river valley and become more and more abundant
toward the highlands. They are, however, never in great variety
and never so abundant as to be a very conspicuous feature of the
landscape. Crotons and Solanums also increase in variety and
abundance as we go southward, up the river.

In the vicinity of Girardot, the land has been mostly cleared
of its forests, and we have an excellent opportunity to study and
collect the flora of these open hills and fields, exposed to the baking
of an extremely hot sun and generally known as " pajinales." Mal-
vaceous and Tiliaceous shrubs, twining herbaceous Leguminosae,
Zanthoxylums, Borrerias and related Rubiaceae, and especially
Crotons and Solanums cover these grounds. There are very many

44 Semi-centennial of Torrey Botanical Club

sedges and in the more barren places, large areas clothed with
Andropogons. Beautiful aquatics are found wherever the soil is
suitable.

Leaving the steamer and traveling by mule, we rapidly climb
to the dry table-lands near the base of the mountains and at places
are obliged to cross projecting mountain spurs. In the lower
places, we are impressed by the beauty of hedges of mata-raton,
a small tree related to Rohinia and handsomely covered with rose-
purple panicles of flowers. Upon this part of the journey, we find
great numbers of shrubby and herbaceous vines belonging to the
milkweed and dogbane families. We crossed during the height
of the dry season so that there was almost no collecting to be done,
but it was quite evident that at certain seasons the flora of this
mesa must be exceedingly rich and wonderfully beautiful. Among
the grasses, Boutelouas are the most conspicuous. Water was
scarce at this season, so that cattle and other domestic animals
were forced to confine themselves to the narrow strips along the
rivers and quebradas. As a result, these places were very much
over-run and their flora largely destroyed. The shrubby and
arborescent vegetation of these ravines and smaller river valleys
consists largely of Acacias, among which are many cactuses, so
that travel among them is very difficult. Large shrubby and
arborescent Crotons and Solanums here continue to maintain a
prominent position.

At length we are so fortunate as to be able to leave the pros-
trating heat, filthy odors, and mosquito- and malaria-infested
valleys of the mesa, and to plunge among the ravines and canyons
of the eastern mountain range. No sooner does one enter one of
these valleys than he finds a rich forest growth, maintained at all
seasons by the streams which flow upon or close to the surface of the
earth. The composition of this flora bears a general resemblance
to that of the remainder of the Andes. Its chief interest will
center in the mixture of genera and species respectively peculiar
to the south and east, and cannot be discussed until our collec-
tions shall have been studied. Crotons maintain the supremacy
at the lower altitudes, while Solanums persist for a great distance
farther up. We see many large areas on the open hillsides that
are covered with a tall and stout Andropogon of a deep rusty-red

Rusby: Recent botanical collecting in Colombia 45

color. Prominent and beautiful are one or more species of Securi-
daca, Polygalaceous woody vines which drape many trees with an
unbroken canopy of pink or rose purple. Orchids and bromeliads
are increasingly abundant as we approach the summit and many
of them are very beautiful. No sooner do we begin to approach
the summit than we encounter blackberry thickets, and these
become more than conspicuous to the very summit. Of these
there are many species and very possibly many hybrids, so that it
is difficult for one to keep track of his collections. The canes are
very tall and heavy, and in many cases assume a half-climbing
condition among the trees. Their panicles of fruit are sometimes
a foot in length and almost as broad at the base, and very dense,
so that they droop heavily over the shrubbery. The individual
fruits are sometimes more than an inch in length and breadth and
their drupelets of surprising size. These larger varieties are
scarcely edible, being sour and somewhat bitter, and reputed as
poisonous. Others are of delicious flavor and are largely marketed.
There is a strawberry with very small, extremely deeply pitted
and rather poorly flavored fruit, which is also considerably mar-
keted. The false strawberry forms large patches, which are bril-
liantly and temptingly fruited. Many of the timber woods are of
great utility and value, especially a species of black walnut which
is largely employed in cabinet work. Among the more beautiful
flowers of the summit are Gesneriaceae, in great variety and of
lovely shades, many terrestrial and arborescent orchids, several
species of Fuchsias of exquisite beauty. Begonias and Oxalis of
numerous species and several Fagelias. The handsomest flower
here, and one of the handsomest that I have ever encountered, is
a species of Bomarea. It climbs to a height of several yards, its
flowering tops and branches then drooping deeply over the banks
of verdure that line the trail. The flower clusters are often large
enough to fill a peck measure and are of a rich maroon color,
sometimes almost as deep as chocolate, at others of a rich crimson.
The bell-shaped flowers are richly mottled in the throat and the
effect is too handsome for description. This species has large
tuberous roots which can be used as food. Dahlias are quite
abundant and we are astonished to find one species forming clumps
fifteen feet or more in height and more like small trees than herbs.

46 Semi-centennial of Torrey Botanical Club

We are more than astonished to spe Physalis peruviana, which
with us is but a few inches in height, there reaching a height of
seven feet and spreading equally in breadth, and capable of yield-
ing a half bushel or more of fruit. Melastomaceae are quite
varied and abundant, and rather beautiful, but cannot compare
in this respect with the representatives of the family in the more
southern countries. There are one or two species of Befaria^
having the same habit of growth as our Azaleas and much re-
sembling them when in full bloom, which are exceptionally
beautiful features of the landscape, their color shading variously
through pink, purple, lavender, and violet. Ferns are in great
variety but not so handsome as in other parts of the Andes which
I have visited. To this statement, we must except the tree ferns,
which are abundant and lovely.

It remains to be stated that Vacciniaceous plants are abundant
and diversified and many of them decidedly showy. For the
most part, they bear scarlet or cherry-red tubular flowers in large
and dense clusters, at the ends of long pendent branches. A num-
ber of them produce delicious edible fruit, one closely filling the
place of our cranberry, but sweet and of better flavor. Erica-
ceae are also quite numerous, especially in swampy regions at the
summit, but not nearly so conspicuous or beautiful as those of the
family last considered.

There is one important feature of the climate here which it
seemed to me might well be taken into consideration by those who
endeavor to grow in temperate houses the plants of such so-called
temperate regions as those under consideration. A "temperate"
climate at these high altitudes is fundamentally different from a
temperate climate at a lower altitude and farther north, even
though the average temperature may be the same in both cases.
In such elevated regions, among tropical mountains, the tempera-
ture regularly falls at night, even in the midst of the summer
season, to a much lower point than is experienced in temperate
latitudes. On the other hand, during the sunny portions of the
day, we find it extremely hot, so that the variation between mid-
day and midnight is extreme. It is not conceivable that plants
which have been developed and which have lived under such con-
ditions should not have acquired a constitution which requires

Rusby: Recent botanical collecting in Colombia 47

such sudden and wide changes for their life and health. It oc-
curred to me while experiencing these conditions that some in-
structive experimental work might be done by removing those
plants to conservatories and subjecting them to various conditions
of temperature, among others the changeable ones under which
they have been accustomed to grow, and to compare the results
upon their life history.

COLLECTING FUNGI AT DELAWARE WATER

GAP

By William A. Murrill

The New York Botanical Garden

The writer's first acquaintance with the region about Delaware
Water Gap was made on the Decoration Day excursion of the
Torrey Botanical Club, May 29-31, 191 7, a brief account of which
appeared in Torreya for August, 191 7. At that time, with the
help of other members of the Club, 81 species of the more conspic-
uous fungi were collected.

Attracted by the variety of soils, exposures, and host plants
which this region affords, I spent also a vacation of two weeks
there, August 1-15, 191 7, and secured a fairly representative col-
lection for that season of the year. I hope some time to be able
to secure the autumnal species. My collections to date include
about 200 species, most of which are to be found in the following
list.

The abundance of a species is indicated by exponents, the
numerals 1-5 denoting a definite number of times collected and
the letters n, nn, and nnn meaning "frequent," "common," and
"very common," respectively.

A. ASCOMYCETES

Daldinia concenirica^
Dothichiza populea^
Endothia parasitica^'^'^
Helotium citrinum^
Helvella lacunosa^
Hypomyces hyalinus^
Hypoxylon coccineum^
Lachnea scutellata^
Leotia lubrica^

Macro podia fusicarpa^

Morchella esculenta^
Orbilia chrysocoma^
Peziza badia^

Phyllactinia suffuUa^
Plasmopora viticola^

Sarcoscypha occidentalism

Xylaria Hypoxylon^
Xylaria polymorpha^

B. UREDINALES

Allodus Podophylli
Gymnoconia inlerstitialis
Gymnosporangium effusum

Gymno sporangium germinale
Gymnosporangium globosum
Gymnosporangium Juniperi-virginianae

48

Murrill: Collecting Fungi at Delaware Water Gap

Gymnosporangium Nidus-avis
Nigredo Caladii
Nigredo Houstoniata
Polylhelis fusca
Puccinia Impatientis

Exidia glandulosd^
Tremella lutescens^

Calocera cornea^

Puccinia Osmorrhizae
Puccinia urlicata
Puccinia Violae
Transchelia punctata

c. hymenomycetes

(a) Tremellales

Tremellodon gelatinosum>

(6) Dacryomycetales

Guepinia spathularia^

Craterellus cornucopioides^
Lachnocladium Michenert^
Lachnocladium Schweinitzii^
Peniophora cinerea^
Stereum complicatum^

(c) Agaricales

I. Thelephoraceae

Stereum frustulosum\
Stereum hirsutum^
Stereum lohatum^
Thelephora multipartita^
Thelephora regularise

Clavaria coronata^

2. Clavariaceae

Clavaria cristata^

Hydnum ochraceum^
Gloeoporus conchoides^

Bjerkandera adusta^
Cerrena unicolor^
Coltricia cinnamomea^
Coriolellus septum^
Coriolus ahietinus^
Coriolus molliusculus^
Coriolus nigromarginalus'"
Coriolus versicolor^
Daedalea confragosa^
Daedalea quercina^^^
Elfvingia foment aria^
Elfvingia megaloma^
Ganoderma Tsiigae^
Gloeophyllum trabeum^
Hapalopilus gilvus^
Hapalopilus rutilans^
Hexagona alveolaris^

3. Hydnaceae

4. Xylophagaceae

5. Polyporaceae

Hydnoporia fuscescens^
Irpiciporus lacteus^^
Irpiciporus mollis^
Lenzites betulina^
Polyporus elegans^
Polyporus Polyporus^
Poria medullapanis^
Poria vapor aria"^
Poronidulus conchifer^
Pycnoporus cinnaharinus^
Pyropolyporus igniarius^
Tyromyces chioneus^
Tyromyces lacteus^
Tyromyces semi pile atiis^
Tyromyces semisupinus^
Tyromyces Spraguei"-

50 Semi-centennial of Torrey Botanical Club

6. Boletaceae

Ceriomyces communis^

Fistulind hepdticd"^

Ceviomyces cvassus scpavcins^

Stvobilomyccs slyobildceus^

Ceviotnyces viscidus^

Tylopilus fcllcus^

7. Agaricaceae

Avmilldyid fncllea}

M^dydsmius dychyyopus^

Chatitevel cintia-bayinus^

^Mdydsmius Cdyyophyllcus^

ChdHtcyel tnitiov^

I^dydsyyiius conjlucns^

Clitocyhe infundibulifoytnis^^

J[^dydsmius clongdtipes^

Clitocybc loctoyiifoyfnis^

JUdydsmius insititius^

Clitocyhe viyctis^

M^dydsmius pcyfoydns^

Copyinus dtyafnetitdyius^

M^dydstnius pydCdcutus^

Copyinus micdceus^

Mdydsmius yesinosus^

Copyinus Spydguei^

Mdydsmius Rotuld^

Coytindyiiis coyyugdtus^

' IMdydsmius siccus^

Coytinellus multifoymis^

Mdydsmius subnudus^

Coytinellus yutildtis^

M eldnoleucd sp.^

Cyifiipellis zotidtd^

Monddelphus illudens^

Entolotnd seyiciceps^

Omphdlopsis cdmpdnelld^

Entolotnd styictius^

Omphdlopsis Fibuld^^

Galeyuld cyispd^

Pdndeolus yetiyugis^

Gdleytild heynisphdcyicd^

Pdnellus stypticus^^

Geopetdluyyi cdfididissiyyiuyyO-

Pdxillus involutus^

Gymnopus dyyophilus^

Pholiotd Johnsonidnd^

Gymnopus Idchnophyllus^

Pholiota mutdbilis^

Gymnopus pldiyphyllus^

Pholiotd pydecox^

Gymnopus yddicdtus^

Plcuyopus unitinctus^

Gymnopus velutipcs^

Pluteus ceyvinus^

Hy pholomd dppcndiculdtum^

Plutcus gydnuldyis^

Hypholomd Cdndollednum^

Pluteus longistyidtus^

Hy pholomd- yugoce phdlufh^

Psdthyyelld dissemindtd^

Inocybe spp.

Russuld JldVd^

Ldccdyid Idccdtd^

Russuld foetens^

Ldccdyid siyidtuld^

Russuld viycscens^

Ldctdyid glyciosmd^

Schizophyllus dlncus^^^

Ldctdyid hygyophoyoidcs^^

Styophdyid setniglobdtd^

LdCtdyid pipcydtd^

Vdgindtd dgglutindtd"^

Ldctdyid vimoselld^

Vaginatd plumbed^"'

Ldctdyid scyobiculdtd

V enendyius cothuyndtus^

Ldctdyid subdulcis^

Venendyius Fyostidnus^

Ldctdyid vdyid^

Venendyius muscdyius^

Lentinus styigosus^

Venendyius phdlloides^

Lepiotd sp.^

Venenayius yubens^"-

Leptonielld sp.^

D. gasteromycetes

Cyucibulum vulgaye^ Lycopeydon gemmdtum^

Cydthus styidtus^ Scleyodeymd duydntium^

Dictyophoyd duplicdtd^

Murrill: Collecting Fungi at Delaware Water Gap 51

Summary

Ascomycetes i8

Uredinales 15

Hymenomycetes :

Lower groups 19

Polyporaceae 33

Boletaceae 6

Agaricaceae 86

Gasteromycetes 5

Total 182

A METHOD OF TEACHING ECONOMIC

BOTANY

By Edward S. Burgess

Hunter College

It may be of interest to put on record a brief synopsis of the
method of work in economic botany which I have worked out for
Hunter College in New York City — aided by assistant teachers.
The course is known as Biology 12, extends through one semester,
and occupies 3 hours a week (or 5 when practicable). Students
taking this course are young ladies, most of whom expect to teach
in the public schools of this city. The conditions under which
we work include the following: from 130 to 200 students to be
provided for, to be met in divisions or laboratory-sections planned
for 20 each, which are supplemented by lectures before a combi-
nation of sections, with some use of lantern, and with exhibition
of specimens additional to those of laboratory or class use.

The students to be considered are city residents; and as usual
with city residents, they have little opportunity for knowledge of
the country or of the details of our flora. The other subjects of
their college course call for about four fifths of their time or more,
and prevent the use of sufficient time in excursion-work to give
much of the desired knowledge of natural habitat. Excursions
and field-work are taken, but necessarily the principal work is in
the class-room.

The relation of this course to others in the college is that it
forms the second among the five half-year courses in botany re-
quired from all students who select the natural science department.
The succession of these required courses is: first, systematic botany
(our Biology 11), February to June, with study of morphology and
classification of Gymnosperms, Monocotyledons, and spring-
flowering Polypetalae; second, economic botany (Biology 12),
uses of plants combined with study of Gamopetalae, and with fall-
flowering Polypetalae and Apetalae; third, plant physiology

52

Burgess: A method of teaching economic botany 53

(Biology 13); fourth, Lower Cryptogams; Algae and Fungi chiefly
(Biology 15); fifth, Higher Cryptogams, with comparison and
review of Spermatophytes (Biology 16).

In planning this sequence, Biology 11-16, it has been my effort
to promote both the knowledge and the love of plants, and to
arrange each course so that it shall provide individual work from
fresh specimens. I also deem it axiomatic that the student's
earlier botanical studies should proceed from the known to the
unknown; and, therefore, that flowering plants should be quite
well understood before beginning detailed work with cryptogams.

In planning this particular portion. Biology 12, our introduction
to economic botany, there are also the following special objects:

First, that the student obtain systematized knowledge of the
relation of the plant-world to man's use — the special province of
economic botany.

Second, that this knowledge be accompanied by distinct con-
ceptions of the plants which furnish economic material; of their
names, appearance, habitat, and structure; also of their relation-
ships. Therefore we study them in a sequence of families.

Third, that, so far as possible, our local plants be used as basis
for study. Therefore our sequence of families is such as will
yield fresh material during the weeks of this course, beginning in
September, and avoiding the use for class purposes of any but
abundant plants (for our rare plants, and any others which are
liable to extermination, should never be gathered for class study.)

Fourth, that foreign plants also should be shown or illustrated,
as supplementary matter.

To secure these objects I have arranged a sequence of topics
which presents in succession the economic relations of families of
plants available in autumn in the vicinity of New York City.
It might also be used in its entirety or with appropriate modifica-
tions, in many other cities.

I have considered the course as forming properly the second
half of a first year in botany; in which year the first half, which
with us begins with February, is an introduction to systematic
botany (our Biology 11), consisting of studies from seeds and
plants, in laboratory and in the field, proceeding from germina-
tion-work and the Gymnosperms, through the Monocotyledons
and many of the Polypetalae.

54

Semi-centennial of Torrey Botanical Club

Therefore I have deemed it essential to this following course
in economic botany that it be based on native fall flowers; that
it must cover gamopetalous families with some additions of fall-
blooming polypetalous and apetalous families; must include mor-
phological characters but devote emphasis to utilities; must not
be confined to succession of families by affinity, but must be in-
fluenced in its succession by blossoming-time and availability of
material.

With these provisions as requisites, the following is an avail-
able approximate order of material used for class work, as now
tested for four or five years.

Utilizing the opportunities given by the fall-flowering Gamo-
petalae, classes take up families somewhat in the following suc-
cession :

1. Labiate families; as Scrophulariaceae, Labiatae, Bignoniaceae;

with references also to Acanthaceae, etc.

2. Kindred non-labiate families; as Boraginaceae, Polemoniaceae,

Convolvulaceae, etc.

3. Orders showing tendency to coalescence in stamens; Cucur-

bitaceae, Campanulaceae, Lobeliaceae.

4. Coalescence in heads (or cymes); from Hamamelidaceae,

Caprifoliaceae, Rubiaceae, to Platanaceae, Valerianaceae,
and Dipsacaceae.

5. Coalescence in both stamens and heads; Compositae, Cichoria-

ceae.

6. Apetalous weedy families; as Ambrosiaceae, Chenopodiaceae,

Amarantaceae, Polygonaceae, Plantaginaceae, Phytolacca-
ceae; with a glance at Euphorbiaceae.

7. Gamopetalous rotate-flowered families; Solanaceae, Apocyna-

ceae, Asclepiadaceae ; noting also Gentianaceae and Oleaceae.

8. Polypetalous families; flowers available in fall; as Cruciferae,

Leguminosae, Cactaceae; and, fibre available, Malvaceae,
Linaceae, Tiliaceae.

9. Apetalous tree-bearing families; Urticaceae, Juglandaceae,

Cupuliferae, Betulaceae.
From point of view of their economic relationships, these
families have meanwhile yielded subjects of study, approxi-
mately in this order:

Burgess: A method of teaching economic botany 55

A. Medicinal plants, sedatives, stimulants, condiments, healing-
agents (many of Scrophularia and Labiate families).

B. Inert related plants, and associated plants chiefly useful as
garden flowers, from the previous, and Bignonia, Verbena,
Phlox families.

C. Foods derived from fleshy roots (Sweet Potato) or fleshy
fruits (Cucurbits) with glance at the reductions in related
parasites (Broom-rape, Dodder) and in submerged plants
(Utricularia) .

D. Beverages and drinks; from Rubiaceae; comparison of tea,
chocolate, etc. ; consideration of caffeine, quinine.

E. Bitters, herb-teas, folk-medicine; Compositae (with compari-
son of Gentianaceae).

7^. Salad-plants; Cichoriaceae, and comparisons.

G. Weedy plants, their values, their control; reasons for their
prevalence ; Compositae and Apetalae.

H. Alkaloids and other drugs and important vegetable poisons;
Euphorbiaceae, Solanaceae, Apocynaceae, Asclepiadaceae.

/. Oils and perfumes; Oleaceae, Linaceae, etc.
/. Dyes; Leguminosae, and families following.
K. Fodder; Leguminosae.

L. Food from seeds; Leguminosae, Buckwheat, Sunflower,

Southwestern Amarantaceae, etc.
ikf. Food from roots, leaves, etc. ; Cruciferae.
N. Mucilage and emollients; Malvaceae, Linaceae, Tiliaceae,

Ulmaceae.

0. Fibre, paper, etc.; the preceding families, Moraceae, Urtica-

ceae, etc. Comparison of tissues, with microscope.
P. Rubber, latex; Ficus, etc.
Q. Tannin and cork; Cupuliferae.
R. Nuts; Juglandaceae, Cupuliferae.

S. Forests, their value, relation to rainfall, and distribution.

T. Forestry methods, their history in Europe, India, United

States, etc.; the present United States Government Forest

Reserves.

I add to the above one further explanation; that cereals, most
fleshy fruits, and berries are not omitted by accident, but are
studied in Biology ii,with the Monocotyledons and Spring Poly-
petalae.

PHILIPPINE MICROMYCETOUS FUNGI

By Paul Weidemeyer Graff

University of Montana

Interest in the field of taxonomic mycology with reference to
tropical localities seems at present to be on the increase. A num-
ber of recent papers have dealt with the situation in the West
Indies in a limited way. The situation in our Pacific possessions
has received, as yet, so little attention that any contribution, how-
ever limited in extent, must be of some interest. It is with this
hope that the following enumeration of species collected in the
Philippine Islands is offered.

The specimens included in the following list were gathered for
the most part in the provinces of Rizal, Laguna, Bataan, and the
vicinity of Manila, on the island of Luzon, by E. D. Merrill and
the writer. A few incidental collections made by others in
various localities are also included. Where several collections
of a species have been made in one locality or vicinity, reference has
been made to only one, making the citations of distributional
value rather than quantitative. All specimens cited are in the
herbarium of the Bureau of Science, Manila.

PHYCOMYCETES
Synchytrium de Bary

Synchytrium Puerariae (P. Henn.) Miyabe, Bot. Mag. Tokyo
19: 199. 1905.

Aecidium Puerariae P. Henn. Bot. Jahrb. 15: 6. 1892.

Uromyces Puerariae Diet. Bot. Jahrb. 27: 282. 1900.

Luzon, vicinity of Manila, Merrill 7424, November 27, 1910,
on leaves of Pueraria Thunhergii. Island of Romblon, Hallier
397 f January, 1904, parasitic on leaves and stems of Pueraria sp.

This species was originally described as an Aecidium from ma-
terial collected on Pueraria sericantha, from New Guinea, and
on Pueraria Thunhergii, from Japan, by Hennings. Later it was

56

Graff: Philippine Micromycetous Fungi 57

recognized as a Synchytrium by Miyabe. It is probable that
Hennings had old and over-mature material as, in this condition,
the fungus pustules might easily be mistaken for the aecidia of
a rust when given a hasty or superficial examination.

Metarrhizium Giard
Metarrhizium Anisopliae (Metsch.) Sor. Zeits. Land. Ges.
Neu.-Russ. 268. 1879.
Entomophthora Anisopliae Metsch. Zeits. Land. Ges. Neu.-
Russ. 21. 1879.
Oospora Destructor Delacr. Bull. Soc. Myc. France 9: 261.

pi. 14. f. 2. 1893.
Isaria Anisopliae Pettit, Bull. Cornell Univ. Exp. Sta. 97: 356.
pi. 6. 1895.

Penicillium Anisopliae Vuill. Bull. Soc. Myc. France 20: 221.
1904.

Septocylindrium suspectum Mass. Kew Bull. Miscel. Inf. i: 4.
1910.

Luzon, vicinity of Manila, Mackey s. n., March, 1914. Para-
sitic on the rhinoceros beetle.

This fungus, commonly known as "Green Muscardine," has
been reported on a number of insect hosts and from a number of
localities. It was first found in the Philippines on the cocoanut
borer, the larva of the rhinoceros beetle, and appears to be fairly
common about the vicinity of Manila. It is also fairly established
in the cocoanut district of Laguna Province.

Reported from tropical America, the Hawaiian Islands, Samoa,
and Europe.

AcHLYA Nees

AcHLYA APicuLATA de Bary, Bot. Zeit. 46: 635. 1888.

Luzon, vicinity of Manila, M. A. Barber, August, 1912, para-
sitic on fish eggs.

The fungus was producing only zoospores when collected.
As the writer desired to use the material for class-room work, pure
cultures were made in distilled water on bits of sterile meat and
also on tubed agar slants. A number of these cultures were kept
growing for some time before they were desired for use, in the hope
of inducing the formation of oospores and antheridia. The results

58 Semi-centennial of Torrey Botanical Club

were entirely negative until some of the common species of water
bacteria were introduced into one of the cultures which had been
growing well but producing only a profusion of zoosporangia.
After the bacteria had been growing in the culture but a few days
the material was examined and a plentiful supply of oogonia and
antheridia was found to have been formed. Bacteria were then
introduced into other pure cultures with a like result. Pure
cultures for zoosporic material and cultures contaminated with
bacteria for the sex-organs were then grown on small bits of meat
and fly larvae which had been sterilized and hardened in alcohol
with the result that extremely satisfactory material was had for
laboratory demonstration.

This species was originally described from material collected
in Germany.

ASCOMYCETES
AscoPHANUS Boudier

Ascophanus verrucosporus sp. nov.

Ascomatibus gregariis vel sparsis, immarginatis, convexis,
lenticularibus, glabris, sessilibus, badiis, minutis, 0.6-1 mm.
diam.; basi filamentis intricatis; ascis maximis, clavatis, apice
rotundatis vel truncatis, operculatis, octosporis, 215-245 ju X 30ju;
sporidiis monostichis, ellipsoidiis, minute verrucosis, 22.8 /x X I5ai»
hyalinis; paraphysibus filiformibus, simplicibus, raro bifidis,
septatis, miniatis, .258 /z X 3.8 /a.

Perithecia clustered to scattered and sparse, at first closed and
lenticular in shape, then expanded, immarginate, convex, fleshy,
smooth, sessile, brown, small, 0.6-1 mm. diam. Base on closely
interwoven hyphal filaments. Hymenium convex. Asci large,
clavate, with rounded to truncate ends which protrude beyond
the surface of the hymenium, 8-spored, 215-245 X 30 /x, discharg-
ing their spores through an operculum. Spores monostichous,
ellipsoidal, minutely verrucose, 22.8 /x X 15 /x, hyaline. Paraphyses
filiform, usually simple but occasionally bifid, septate, reddish-
brown, 258 /X X 3.8 IJL.

Luzon, Province of Rizal, Fort McKinley, Mary S. Clemens
s. n., February 15, 1912, growing on earth in a moist shaded loca-
tion.

Triblidiella Saccardo
Tryblidiella rufula (Spreng.) Sacc. Syll. Fung. 2: 757. 1883.
Hysterium rufula Spreng, Sv. Vet. -Acad. Handl. 1820: 30.
1820.

Graff: Philippine Micromycetous Fungi 59

Tryhlidium guaraniticum Sacc. Syll. Fung. 9: 1103. 1891.
Tryhlidiella Balansae Sacc. Syll. Fung. 9: mo. 1891.
Rhytidhysterium javanicum Penz. & Sacc. Malpighia 11 : 528.
1897.

Rhytidhysterium guaraniticum Sacc. & Syd. in Sacc. Syll.
Fung. 16: 666. 1902.

Luzon, Province of Bataan, Bur. Sci. igogz P. W. Graff, No-
vember 3-19, 1912, on dead twigs in the forest.

Reported previously as being collected in Amboina, New
Zealand, Brazil, Guiana, and Cuba, on a variety of hosts.

Genea Vittadini

Genea Thwaitesii (Berk. & Br.) Fetch, Ann. Myc. 5 : 475. 1907.
Hydnocystis Thwaitesii Berk. & Br. Jour. Linn. Soc. Bot.
14: no. 1875.

Luzon, Province of Laguna, Mount Maquiling, Bur. Sci.
16040 Brown, February 27, 1912, on dead fallen twigs.

This is an interesting fungus which is located systematically
between the Discomycetes and Tuberaceae. When mature, the
round or irregular waxy ball opens very much like a Peziza with
an inrolled margin. The spores, however, are not exposed by this
but are still covered by an inner layer of tissue. As a result, the
mature specimen has the appearance of a cup fungus with an outer
and inner layer of sterile tissue between which is located the unex-
posed hymenial layer. This, species is yellow in color and often
opens at the side or irregularly instead of at the top.

Previously collected in Ceylon.

Meliola Fries
Meliola amphitricha Fr. Elench. Fung. 2: 109. 1828.
Sphaeria amphitricha Fr. Syst. Myc. 2: 513. 1823.
Amphitrichum Hibisci Spreng. Sv. Vet. -Acad. Handl. 1820:
52. 1820.

Amphitrichum Sacchari Spreng. Sv. Vet. -Acad. Handl. 1820:
52. 1820.

Luzon, vicinity of Manila, Bur. Sci. 11002 P. W. Graff, De-
cember 28, 191 1, on Tamarindus indica.

This fungus has been previously collected in the Philippines

60 Semi-centennial of Torrey Botanical Club

on Pithecolohium apoense, Sapindus Saponaria and Viburnum
odoratissimum but not on the tamarind. It has also been reported
from North and South An^erica and Austraha.
Meliola Desmodii Karst. & Roum. Rev. Myc. 12: 77. 1890.

Luzon, Province of Bataan, Mount Mariveles, Bur. Sci.
igo2^ P. W. Graff, on leaves of Desmodium virgatum; Bur. Sci.
igosS P. W. Graff, on Desmodium gangeticum, November, 191 2.

Reported also from Tonkin, Indo-China, on living leaves of
Desmodium sp.

Meliola quadrispina Rac. Parasit. Alg. Pilz. Javas 3 : 33. 1900.

Luzon, Province of Laguna, Mount Maquiling, Merrill 86^5 y
March, 191 3, on leaves of Hewittia suhlohata.

Originally described from material collected at Buitenzorg,
Java, on Ipomoea sp,

Meliola Arundinis Pat. Jour, de Bot. 11 : 348. 1897.

Panay, Province of Iloilo, Bur. Sci. 18024 C. B. Robinson,
December 27-31, 191 2, on leaves of Saccharum sp.

Previously collected at Tonkin, Indo-China, on living leaves
of Arundo Donax.

Meliola substenospora v. Hohn. Sitzb. Akad. Wiss. Wien 118:
317. 1909.

Luzon, Province of Laguna, Mount Maquiling, Merrill 8653 y
March, 191 3, on leaves of Oplismenus compositus.

Collected previously at Buitenzorg, Java, on Phragmites sp.
Meliola Mangiferae Earle, Bull. N. Y. Bot. Card. 3: 307.

1905.

Luzon, Province of Rizal, Bosoboso, ^m/'. Sci. S112 M. Ramos,
October, 191 2, on the leaves of Mangifera indica.

A species quite common on mangoes in the Philippines. Ap-
parently the fungus is not parasitic but, as in the case of the other
members of the genus, appears in connection with aphid attacks,
finding nourishment in the honey-dew excreted by them. Any
injury to the host is probably due to a smothering effect caused by
the heavy incrustation sometimes developed by the fungus.
For the most part this fungus is found on the under surface of the
leaf.

Described from material collected on the same host in Porto
Rico.

Graff: Philippine Micromycetous Fungi

61

Meliola Litseae sp. nov.

Maculis mycelii hypophyllis, rotundato-angustatis, nigres-
centibus, gregariis vel sparsis, subcrustaceis, superficialibus, 3-8
mm. diam., margine hyphis radiantibus; hyphis ramosis usque
ad 5.7 M crassis, atro-fuscis, septatis; hyphopodiis capitatis, 15
longis, capitulo subgloboso, 7.6 ju latis, oppositis vel alternantibus;
peritheciis gregariis vel sparsis, rotundatis, 190-230 fi diam. ; setulis
rigidis, atris, rectis, abrupte basi curvatis, apice acutis, fuscis,
182-437 )LtX 7.6 ju; ascis cylindraceis, bisporis, 26/xX68/x; sporidiis
ellipsoideis, utrinque rotundatis, 4-septatis, constrictis, 13/XX42/X,
atro-fuscis.

Mycelial spots on the under surface of the leaf, of rather
limited area, round, black, sparse to gregarious, becoming con-
fluent, subcrustaceous, superficial, 3-8 mm. in diameter. Mar-
ginal hyphae radiating. Hyphal branches uniformly of about the
same diameter, 5.7 ju, dark brown, septate. Hyphopodia capitate,
about 15 /X long, extremity subglobose, 7.6 fx broad, arranged oppo-
site or alternate on the hyphae. Perithecia gregarious to sparse,
round, 190-230 ju in diameter. Setae rigid, dark-colored, upright,
with an abruptly curved base, apex acute, brown, 182-437 MX7.6/i.
Asci cylindrical, two-spored, 26juX68)U. Spores rounded to ellip-
soid, 4-septate, constricted at the septa, mature spores averaging
i3)uX42/i, dark brown.

Luzon, Province of Laguna, Mount Maquiling, P. W. Graff
s. n., February 28, 1912, on the under side of living leaves of
Lit sea sp.

Phyllactinia Leveille

Phyllactinia guttata (Fr.) Lev. Ann. Sci. Nat. Bot. IIL 15:

144. 7./. ji. 1851.
Erysiphe guttata Fr. Syst. Myc. 3: 245. 1829.
Sclerotium Erysiphe /3 corylea Pers. Syn. Fung. 124. 1801.
Sclerotium suffultum Rebent. Prod. Flor. Neom. 360. 1804.
Erysiphe Coryli Hedw. f. in DC. Fl. Fr. 2: 272. 1805.
Erysiphe Alni DC. Syn. PI. Fl. GalL 57. 1806.
Erysiphe suffulta Nees, Syst. Pilz. Schw. 148. pi. 14. f. IJ4.

1817.

Phyllactinia Candollei Lev. Ann. Sci. Nat. Bot. IIL 15: 150.

pi. 7. f. 12. 1 85 1.
Phyllactinia suffulta Sacc. Michelia 2: 50. 1880.
Phyllactinia antarctica Speg. Boletin Acad. Nac. Cien. Cordoba

11:34. 1887.

62 Semi-centennial of Torrey Botanical Club

Luzon, Manila, Bur. Sci. q6sq C. B. Robinson, January 28,
1910; Bur. Sci. 16/QJ, SjS P. W. Graff, September-October, 1912;
Province of Bataan, Lamao, Bur. Sci. 19123 P. W. Graff, Novem-
ber 6, 1 91 2, on leaves of Morus alba.

Though observed in luxurious growth at the end of the rainy
season and for some time after, in no case was the perfect form
of the fungus found. A yellow spotting of the leaves was caused
which advanced as the fungus developed, keeping somewhat ahead
of the growth until the entire leaf might be involved. Leaf fall
frequently followed.

A fungus of practically universal distribution and found on a
great many hosts.

Parodiella Spegazzini
Parodiella grammodes (Kunze) Cooke, Austr. Fungi 301. 1892.
Sphaeria grammodes Kunze, in Weigelt, PI. Exs. Surin. — .
1828.

Dothidea grammodes Berk. Jour. Linn. Soc. Bot. 10 : 390. 1869.
Dothidea peris porioides Berk. & Curt. Grevillea 4: 103. 1876.
Parodiella peris porioides Speg. Ann. Soc. Cien. Argent. 9: 178.
1880.

Dothidella grammodes Sacc. Syll. Fung. 2: 634. 1893.

Luzon, Province of Rizal, Antipolo, Bur.. Sci. 21878 M. Ramos,
August 19, 1 91 3, on the under side of leaves of Desmodium Scor-
piurus; Manila, Bur. Sci. S148 E. D. Merrill, January 11, 1913,
on leaves of Desmodium trifiorum.

This fungus has previously been reported from the Philippines
on Crotolaria stenophylla and Desmodium capitatum. It has also
been reported from the Islands under the name ''Parodiella pumila
(Cooke) Sacc." on Desmodium trifiorum and Smithia ciliata,
which proves to be an error both in determination and typography.
Parodiella puncta (Cooke) Sacc. is evidently the species intended.

Previously reported from North and South America, Ceylon,
India, Australia, and Natal.

AsTERiNA Leveille
AsTERiNA PEMPHiDioiDES Cookc, GreviUca 5: 16. 1876.

Luzon, Province of Bataan, Mount Mariveles, Bur. Sci.
iQoys P. W. Graff, November, 191 2, on leaves of Eugenia sp.

Graff: Philippine Micromycetous Fungi

63

This species seems to be related to Asterina pelliculosa Berk. &
Br., but has much longer sporidia.

Collected previously in India.
Asterina Elmeri Syd. Philip. Jour. Sci. Bot. 9: 181. 1914.

Luzon, Province of Bataan, Mount Mariveles, Bur. Sci.
IQ060 P. W. Graff, November, 191 2, on leaves of Champeria
manillana.

Found as yet only in the Philippines.
Asterina Lawsoniae P. Henn. in Warb. Monsunia i : 159. 1900.

Luzon, vicinity of Manila, Bur. Sci. S42 P. W. Graff, on Law-
sonia inermis.

This collection is interesting in that it is the first of the species
reported from the Philippines since the collection of the type
material by Warburg in 1888. In the meantime, the species has
been collected in India.

Phyllachora Nitschke
Phyllachora circinata Syd. Ann. Myc. 8: 38. 1910.

Leyte, Jaro, Wenzel 559, February 8, 1914, on Ficus chryso-
lepis in the forest at 500 m. altitude.

This fungus was described from material collected in the
Philippines on an undetermined species of Ficus and has also been
reported on Ficus odorata. This collection adds a new host to the
list.

Not reported outside the Philippines.
Phyllachora Kaernbachii P. Henn. Bot. Jahrb. 18: 39. 1904.
Phyllachora Merrillii Ricker, Philip. Jour. Sci. i : Suppl. 280.
1906.

Phyllachora Fici-fulvae Koord. Bot. Unters. 183. 1907.
Phyllachora Fici-minahassae P. Henn. Philip. Jour. Sci. Bot.
3:45. 1908.

Luzon, Province of Benguet, Merrill 7914, May, 191 1, on
Ficus validicaudata; Manila and vicinity, Merrill 7468, January-
February, 1911, on Ficus ulmifolia; Province of Laguna, San
Pablo, Merrill 7486, February, 1911, on leaves of Ficus odorata;
Mount Maquiling, Bur. Sci. 15988 P. W. Graff, February 23-28,
1 91 2, on leaves of Ficus ulmifolia; Mindoro, Bulalacao, Merrill
927, on Ficus sp.; Mount Halcon, Merrill s 57 9, 5625, on leaves of

64 Semi-centennial of Torrey Botanical Club

Ficus heterophylla. Balut Island, Merrill 5422, October, 1906,
on leaves of Ficus Minahassae.

The only distinction between the species Phyllachora Kaern-
bachii, P. Fici-fulvae and P. Minahassae appears to be age of
infection and hostal in nature. In cases where the leaf infections
are few and scattered we have one species but if they have become
numerous and coalescent with apparently a more even distribution
of the fungus stroma we have, according to report, another. The
spore and stroma characters in these three species show no specific
distinctions.

Collected in Java, New Guinea, and the Philippine Islands.

Chaetomium Kunze
Chaetomium stercoreum Speg. Michelia i: 222. 1877.

Luzon, vicinity of Manila, Bur. Sci. idoyi P. W. Graff, Sep-
tember, 1912, on horse-dung.

The characters of this collection seem to fit in with Spegaz-
zini's species more closely than any other and any differences are
not distinct enough to warrant a separation from it.

This seems to be a fungus of very general distribution.

Astrocystis Berkeley & Broome
AsTROCYSTis MiRABiLis Berk. & Br. Jour. Linn. Soc. Bot. 14: 123.
1875.

Rosellinia Bamhusae P. Henn. Hedwigia 47: 256. 1908.

Luzon, Province of Pampanga, Mount Arayat, Merrill 5030,
1906; Province of Bataan, vicinity of Limay, Bur. Sci. igooy P.
W. Graff, November, 191 2, on bamboo.

Previously collected on the culms of dead Bamhusa sp., at
Peradeniya, Ceylon.

Tryblidiella Saccardo

Tryblidiella rufula (Spreng.) Sacc. Syll. Fung. 2: 757. 1883.
Hysterium rufulum Spreng. Sv. Vet. -Acad. Handl. 1820: 20.

1820; Fr. Syst. Myc. 2 : 584. 1823.
Hysterium confluens Kunze, in Weigelt, PI. Exs. Surin. — .
1828.

Luzon, Province of Rizal, Bur. Sci. 21885a M. Ramos, August,

Graff: Philippine Micromycetous Fungi

65

1 91 3, on Citrus decumana, associated with Amphisphaeria hesper-
idum Penz.

A fungus of very general tropical and subtropical distribution.

Amphisphaeria Cesati & De Notaris
Amphisphaeria Hesperidum Penz. Michelia 2: 414. 1882.

Luzon, Province of Rizal, Antipolo, Bur. Sci. 2188$ M. Ramos,
August 16, 1 91 3, on dead twigs of Citrus decumana.

The perithecia in the mature specimens measure up to 500 in
diameter with an ostiole of 200 /z. The asci are somewhat longer
and more slender than those described by Penzig, measuring on an
average 75 long and 11.5/x at their broadest part. The base of
the ascus is rather attenuated than thick. The spores vary from
3.8 to 4.5 //by 14 to 15 /X, corresponding exactly with those of the
original description. They are dark brown in color, biguttulate,
mono- or distichous and slightly curved. The paraphyses are
slender, hyaline, and simple.

Originally described from material collected in Italy on twigs
of Citrus Aurantium.

Mycosphaerella Johanson
Mycosphaerella Fragariae (Tul.) Lindau, in Engl. & Prantl,
Nat. Pflanzenfam i^: 425. 1897.
Stigmatea Fragariae Tul. Select. Fung. Carp. 2: 286. pi. ji.
1863.

? Sphaeria fragariaecola Wallr. Flor. Crypt. Germ. 2: 767.
1833.

Sphaerella Fragariae Sacc. Syll. Fung, i : 505. 1882.
Sphaeria Fragariae Fuckel, in Frank, Krankh. Pflanz. 607.
1880.

Ramularia Fragariae Peck, N. Y. State Mus. Rep. 34: 29. pi. j.
f. 12-15. ''1881" [1883].

Ramularia Tulasnei Sacc. Syll. Fung. 4: 203. 1886.

Luzon, Province of Benguet, Baguio, Bur. Sci. 20968 Wolseley,
February, 1913, on cultivated strawberry plants.

Of very general occurrence through Europe and America.
Probably introduced into the Philippines with plants for culti-
vation.

66 Semi-centennial of Torrey Botanical Club

Mycosphaerella Musae (Speg.) Syd. Philip. Jour. Sci. Bot. 8:
482. 1913.

Sphaerella Musae Speg. An. Mus. Buenos Aires III. 12: 354.
1909.

Luzon, Province of Laguna, Los Banos, Baker 21, September
10, 1 91 2, on leaves of Musa sapientum.

This fungus causes one of the rather common leaf-spots of the
cultivated banana in the vicinity of Manila and the surrounding
provinces. No great loss is occasioned by its attack, but in severe
cases a proportionate weakening of the host must necessarily
result.

Probably introduced into the Philippines. Originally described
from South America.

Mycosphaerella Pericampyli Syd. Philip. 'Journ. Sci. Bot. 8:
270. 1913.

Luzon, Province of Bataan, Lamao, Merrill 867 g, January,
1 91 3, on leaves of Pericampylus incanus.

Reported as yet only from the Philippines.
Mycosphaerella Alocasiae Syd. Philip. Jour. Sci. Bot. 8: 195.
1913.

Luzon, Manila, P. W. Graff, on living or dying leaves of
Alocasia indica.

DiDYMOSPHAERIA Fuckel

Didymosphaeria striatula Penz. & Sacc. Malpighia 15: 227.
1901.

Luzon, Province of Bataan, Mount Mariveles, Bur. Sci. S156
P. W. Graff, November, 191 2, on dead bamboo.
Previously collected in Java.

Anthostomella Saccardo
Anthostomella mirabilis (Berk. & Br.) v. Hohn. Fragm. Myc.
6:54. 1909.

Astrocystis mirabilis Berk. & Br. Jour. Linn. Soc. Bot. 14: 122.
1873.

Rosellinia Bamhusae P. Henn. Hedwigia 47: 256. 1908.
Luzon, Province of Bataan, Lamao, Bur. Sci. igooy P. W.
Graff, November 3-19, 1912, on dead Bambusa sp.; Mount Mari-

Graff: Philippine Micromycetous Fungi 67

veles, Bur. Sci. Si 25 P. W. Graff, November 10, 191 2, on dead
Schizostachyum sp.

Collected previously in France and Ceylon.

NuMMULARiA Tulasne

NuMMULARiA ANTHRACODES (Fr.) Cooke, Grcvillea 11 : 126. 1882.
Sphaeria anthracodes Fr. Linnaea 5: 544. 1830.
Hypoxylon anthracodes Mont. Ann. Sci. Nat. Bot. II. 13: 359.
1840.

Luzon, Province of Laguna, Mount Maquiling, Bur. Sci.
15955 P' Graff, February, 191 2, on dead unidentified bark;
Province of Bataan, Mount Mariveles, Bur. Sci. IQ026, 19054 P.
W. Graff, November 3-19, 1912, on dead tree bark.

Collected previously in Brazil, Guiana, Argentine, and Borneo.

Hypoxylon Bulliard

Hypoxylon marginatum (Schw.) Berk. Jour. Linn. Soc. Bot.
10: 385. 1869.

Sphaeria marginata Schw. Trans. Am. Phil. Soc. 4: 190. 1832.

Luzon, Province of Bataan, Mount Mariveles, Bur. Sci.
igo55 P' Graff, November, 1912, on dead twigs.

Previously reported from Venezuela, Cuba, United States,
Ceylon, and Borneo.

Hypoxylon rubiginosum (Pers.) Fr. Summa Veg. Scand. 384.
1845.

Sphaeria ruhiginosa Pers. Syn. Fung. 11. 1801.

Luzon, Province of Bataan, Mount Mariveles, ,Bur. Sci.
19063, igoSi P. W. Graff, November, 1912, on a decaying log in
the forest.

Previously collected on a great variety of hosts in Europe,
North America, Cuba, Ceylon, Java, and North Africa.
Hypoxylon effusum Nits. Pyren. Germ. 48. 1867.

Luzon, Province of Bataan, Limay, Bur. Sci. IQ068 P.
Graff, November 3-19, 19 12, on decaying tree branches.

Common in central and southern Europe.

68 Semi-centennial of Torrey Botanical Club

FUNGI IMPERFECTI

Phyllosticta Persoon

Phyllosticta Brideliae sp. nov.

Maculis versiformibus, rubro-fuscis, demum griseo-fuscis vel
griseis; peritheciis epiphyllis, sparsis vel laxe congregatis, puncti-
formibus, semiimmersis, minutis, semiglobosis, 50-57 /x X 64.5 />t;
ostiolis rotundatis 3.5 /x; sporulis simplicibus, hyalinis, cylindraceis,
1.2 M X 9-5M.

Fungus attack causing reddish-brown irregular anastomosing
spots which are evident on both sides of the leaf. Color of spots
more dull on the under surface than the upper due to the texture
of the leaf surfaces. As the fungus becomes mature the spots
become grayish in color and the perithecia break through the
upper surface of the leaf. The perithecia are sparse to somewhat
gregarious, small, partly immersed and semiglobose, measuring
50-57 jjL X 64.5 fji. The ostiole is round and about 3.5 ix in diameter.
The spores are cylindrical, 1.2 /x X 9.5 m» simple and hyaline.

Luzon, Province of Laguna, Mount Maquiling, P. W. Graff s.
n., February 22, 1912, parasitic on leaves of Bridelia sp.
Phyllosticta cocophila Pass. Diag. Fung. Nuov. 3: no. 63.
1888.

Luzon, vicinity of Manila, Bur. Sci. 20644 P' Graff, Jan-
uary, 1 91 3, on leaves of Cocos nucifera.

Previously reported and described from material collected in
the Botanical Garden of Parma on leaves of Cocos flexuosa.

Phoma Fries
Phoma herbarum West, Michelia 2 : 92. 1880.

Luzon, vicinity of Manila, Merrill 846 j, December, 191 2, on
pods of Cassia occidentalis.

Previously reported from the Philippines on branches of
Manihot utilissima. This fungus, with its numerous varieties,
is of very general distribution throughout Europe, Asia, and North
America.

Macrophoma Berlese & Voglino
Macrophoma Musae (Cooke) Berl. & Vogl. Atti Soc. Venet.
1886: 187. 1886.
Sphaeropsis ? Musarum Cooke, Grevillea 8: 93. 1879.
Phoma Musae Sacc. Syll. Fung. 3 : 163. 1894.

Graff: Philippine Micromycetous Fungi

69

Luzon, vicinity of Manila, Bur. Sci. Si 66 P. W. Graff, March
26, 1 91 3, on dead leaves of Musa paradisaca.
Collected previously in India.

CoNiOTHYRiUM Corda

CoNiOTHYRiUM MELASPORUM (Berk.) Sacc. Syll. Fung. 3: 319.
1884.

Darluca melaspora Berk, in Cooke, Nuovo Gior. Bot. Ital.
10:26. 1878.

Luzon, Subprovince of Bontoc, Banco, Vanoverhergh 3710,
August 21, 1 91 3, on dead stalks of Saccharum spontaneum.

Previous descriptions of this fungus have been so meager in
their characterization that the following is appended: Pustules
dark, usually long rather than round, when mature up to I mm.
in length. Perithecia hemispherical with flattened base and
rounded top, 280-330 /x broad and 1 15-160 ix high, at first covered
by the epidermis of the stem, which at maturity breaks along the
longest axis of the pustule and lengthwise the stem. Spores
oblong to oblong-allantoid, 4-4.5 m X 11-15 light to moderately
brown when examined singly, in mass dark brown.

Reported previously from Australia on Saccharum officinarum.

Naemgspora Persoon

Naemospora Fici (Brond.) Sacc. Syll. Fung. 10: 507. 1892.

Libertella Fici Brond. Rec. PI. Crypt. Agen. 31. pi. 8. f. 3-5.
1830.

Luzon, vicinity of Manila, Bur. Sci. iiooi P. W. Graff, De-
cember 27, 191 1, on leaves of Ficus sp.
Reported on Ficus Carica in France.

DiPLODiA Fries

DiPLODiA Agaves Niessl, Hedwigia 17: 176. 1878.

Luzon, Province of Bataan, Lamao, Bur. Sci. 206^7 P. W.
Graff, November, 191 2, on leaves of Agave Cantula {A. americana) ;
M. M. Saleeby s. n., September 18, 1913, on the same host.

Originally described from material collected on Agave Cantula
in the Botanical Garden of Calcutta, India.

70

Semi-centennial of Torrey Botanical Club

BoTRYODiPLODiA Saccardo
BoTRYODiPLODiA Elasticae Petch, Ann. Roy. Bot. Card. Perad.
3:7- 1912.

Luzon, Province of Batangas, Tanauan, Merrill 8j6^, Decem-
ber 19, 191 1, on dead branches of Citrus nohilis.

A comparison of this collection with Petch's species, described
by him as being found in Ceylon on Hevea brasiliensis and Cas-
tilla elastica, seems to indicate their identity.

Aschersonia Montague
Aschersonia sclerotioides I^. Henn. Hedwigia 41: 146. 1902.

Luzon, Province of Bataan, Lamao, Bur. Sci. 1913Q P. W.
Graff y November, 191 2, associated with coccids on dead branches
of Citrus sp.

Reported previously on Lecanium sp., on Castilla elastica
collected at Buitenzorg, Java.

ACTINOTHYRIUM Kuuze

Actinothyrium Hopeae sp. nov.

Peritheciis plus minus dense gregariis, orbiculato-scutiform-
ibus, latissime conicis, 415-460 /z diam., fusco-castaneis, contextu
fibroso, compacto, margine breviter radiato-fimbriatis; ostiolo
manifesto; sporulis cylindraceis aliquantum curvatis, continuis,
II /X X 49-53 hyalinis; basidiis brevis, simplicibus, hyalinis.

Perithecia more or less densely gregarious, round-shieldlike,
center much higher than the sides, making it conical in form, 415-
460 ju in diameter, chestnut-brown. Context compact-fibrous.
Margin typically of a short radiating fringe. Ostiole present.
Spores cylindrical, slightly curved, continuous, 11 juX 49-53 m>
hyaline. Basidia forming a short, simple, hyaline base.

Luzon, Province of Tayabas, Mount San Antonio, For. Bur.
19556 H. M. Curran, December 14, 191 1, on living leaves of
Ho pea Pie'rrei.

Gloeosporium Desmazieres & Montague
Gloeosporium Palmarum Oudem. Contr. Flor. Myc. Pays-bas
14: 48. 1890.

Luzon, vicinity of Manila, Merrill 858^, February 22, 1913, on
leaf sheaths of Areca Catechu.

Reported in Holland on Areca sapida.

Graff: Philippine Micromycetous Fungi 71

Pestalozzia de Notaris
Pestalozzia Palmarum Cooke, Grevillea 4 : 115. 1875.

Luzon, Province of Laguna, Mount Maquiling, Merrill 8646,
March, 1913, on leaves of Pinanga sp.

A very common leaf-spot is also caused by this fungus on Cocos
nucifera, which in severe cases must impoverish the tree. Very
severe attacks of this disease have been observed by the writer in
the cocoanut districts of Laguna and Tayabas Provinces and the
northern portion of the island of Mindoro.

This fungus is of common distribution in the Asiatic tropics.

OiDiuM Link

OiDiUM OxALiDis McAlp. Proc. Roy. Soc. Vict. 6: 219. 1894.

Luzon, Manila, Merrill 8j6q, March 12, 1912, on leaves of
Oxalis repens.

Previously reported and described from material collected on
Oxalis corniculata in Victoria, Australia.

Aspergillus Link
Aspergillus periconioides Sacc. Ann. Myc. 11 : 320. 1913.

Luzon, Province of Bataan, Lamao, Bur. Sci. S136 P. W.
Graff, November, 191 2, on living leaves of Carica Papaya.

This fungus is a parasitic species which has been found to
attack the leaves of the papaya and, in some cases, apparently
resulting in considerable damage to the host. Some of the plants
from which this collection was made were losing their leaves in
serious numbers. The infection spots spread till the entire leaf
becomes involved and premature leaf fall follows.

Not reported as yet outside the Philippines.

Haplographium Berkeley & Broome
Haplographium echinatum (Rivolta) Sacc. Syll. Fung. 4: 307.
1884.

Penicillium echinatum Rivolta, Parass. 451./. 150-151. 1873.

Luzon, Manila, P. W. Graff s. n., March 25, 1912, on moist
paper which was folded together.

Reported previously from northern Italy on a collection of
wheat culms.

72

Semi-centennial of Torrey Botanical Club

Cladosporium Link
Cladosporium subfusoideum McAlp. Fung. Dis. Citrus Austr.
79. pL 15. f. 21, 22. 1899.

Luzon, Province of Bataan, Lamao, Bur. Sci. 21000 P. J.
Wester, June 16, 191 3, on leaves of Citrus Aurantium.

This fungus has been reported from a number of orange-growing
places in the Philippines and is not, in all probability, confined
to this species of Citrus. It appears to be the common scab-
forming fungus affecting citrus trees in the Philippines but is
usually found to have other fungi associated with it.

Previously collected on the lemon in Victoria and New South
Wales, Australia.

Alternaria Nees
Alternaria Brassicae (Berk.) Sacc. Syll. Fung. 4: 546. 1886.
Macrosporium Brassicae Berk, in Smith, Engl. Flora 5: 339.
1836.

Polydesmus exitiosus Kiihn, Krankh. Kult. 165. pi. 6. 1858.

Luzon, vicinity of Manila, Merrill 8463, December, 1912,
associated with Cercospora occidentalis Cooke, on dead pods of
Cassia occidentalis.

A fungus, with its varieties, of very broad distribution.

Cercospora Fresenius
Cercospora Gliricidiae Syd. Philip. Jour. Sci. Bot. 8: 283.
1913.

Luzon, Province of Batangas, Santo Tomas, Bur. Sci. iQi2y
P. W. Graf, November 30, 1912; Province of Laguna, Pagsanjan,
Bur. Sci. Si 6 1 P. W. Graff, February 22, 191 3, both collections on
leaves of Gliricidia sepium.

As yet reported only from the Philippines.
Cercospora Litseae-glutinosae Syd. Philip. Jour. Sci. Bot.
8:284. 1913.

Luzon, Province of Bataan, Mount Mariveles, Bur. Sci.
IQ042, iQO'/s P. W. Graff, November, 191 2, on leaves of Litsea
glutinosa.

Reported only from the Philippines.
Cercospora occidentalis Cooke, Hedwigia 17: 39. 1878.

Graff: Philippine Micromycetous Fungi 73

Luzon, vicinity of Manila, Merrill 8463, December, 191 2,
associated with Alternaria Brassicae (Berk.) Sacc, on dead pods
of Cassia occidentalis.

Originally described from material collected on leaves of the
same host in the southern portion of the United States.
Cercospora personata (Berk. & Curt.) Ellis, Jour. Mycol. i : 63.
1885.

Cladosporium personatum Berk. & Curt. Grevillea 3: 106.
1874.

Luzon, Province of Bataan, Lamao, Merrill 8684, January,
1913; Province of Batangas, Tanauan, Merrill 8364, December,
191 1, on Arachis hypogaea.

This fungus has undoubtedly been introduced from the United
States in connection with the importation of seed for planting in
the Islands.

A fungus of common occurrence in the United States and the
West Indies.

CiLiciOPODiuM Corda
CiLiciOPODiuM GRAYANUM Sacc. & Ell. Michelia 2: 581. 1878.

Luzon, Province of Laguna, Los Banos, Baker J5, September
12, 1 91 2, on an undetermined piece of dead wood.

This species has been previously reported from North America.

»

WEATHER CONDITIONS AND CROP DISEASES

IN TEXAS

By Frederick H. Blodgett

Agricultural and Mechanical College of Texas

The local environment under which plants grow is of interest
to collectors and other students of plant life to a degree varying
with the particular end in view, or group under observation. For
the collector, local soil characters and gross moisture conditions
usually serve sufhciently well to indicate the probable habitat
in which specimens may be expected. More detail is desirable
in the forms commonly regarded as more sensitive to substratum
variations such as mosses and hepatics, but herbarium labels and
field notes are usually inadequate even in these groups. That
the distribution of so readily disseminated forms as parasitic fungi
attacking field crops may also show response to environmental
conditions of distinctly local character was shown by field obser-
vations here recorded.

Following the Gulf storm of August i6 to 20, 1915, in Texas,
the damage to the cotton crop by the anthracnose (Glomerella
Gossypii Edg.) was found to be directly related to the distributions
of rainfall during the time the storm was passing over the affected
area of the state. But in this case the relation of environment
to disease was on a scale of great proportions and was in accordance
with one's expectation, except as to degree of damage, which could
not be anticipated. In 1916 there was no such great disturbance
over the cotton area of the state and local factors were able to act
more nearly as distinct elements of the environment. The season
was rather dry than wet, though the drouth was not so severe as
during the current season. The general condition of the cotton
crop in Hill County in the central portion of the state (Texas) was
approximately 80 per cent of normal, the drouth stunting the
crop to the amount of 20 per cent, considering the general vigor
of plants and yield of crop together. Five inspections of separate

74

Blodgett: Weather and crop diseases in Texas 75

fields were made near Hillsboro, at distances varying from one
half mile to one and one half miles from the starting point; on the
west of town, 8 per cent of the bolls were found to be spotted with
anthracnose or bacterial spot {Bacterium malvacearum E. Sm.) . In
fields to the south, east and north of town successively the counts
rose steadily until the last field inspected showed 26 per cent of
the bolls (in scattered counts of 100 each) to be spotted. The air-
line distance from the 8 per cent field to the 26 per cent one was
perhaps three miles, the general character of the land was level, but
somewhat broken by shallow erosion washes and depressions,
irregularly distributed. It was found, however, that the local
showers, though fewer than usual, commonly were more abundant
and heavier in precipitation on the east and northeast of the town
than elsewhere. The high percentages were in the area of the
more frequent showers, i. e., of greater local humidity. This
factor was found active at other points also.

Not only is the areal distribution of local rainfall important,
but the periodic recurrence of showers is of consequence also.
Near Austin, in a field some five miles east of the city, a field was
inspected on September 4, 1916, and gave a count of 10 spotted
bolls to the ICQ. On the 23d the same field showed 28 per cent
spotted. During the three weeks intervening, several showers
had fallen, but apparently none so heavy as to give a half-inch
precipitation and conditions were more or less cloudy. This
favorable weather lasted for- about one week with September 15
as the mean date for the period. The spots in this case were
almost wholly due to anthracnose. The seasonal distribution of
rainfall is especially important in the case of cotton anthracnose
under Texas conditions, where first-class conditions may exist for
several weeks in the picking season, only to become suddenly
serious by the disturbance in weather conditions due to some trop-
ical storm of more or less severity, as in August, 191 5 (Galveston
storm), and August, 191 6 (Corpus Christi storm). In the eastern
cotton states where the normal humidity is greater the degree of
damage by these storms is probably less through change of humid-
ity than by mechanical effect of wind and driven rain within the
actual path of the storm. These factors are effective in Texas
also, as a matter of course, but then further damage results from

76 Semi-centennial of Torrey Botanical Club

the disturbed meteorological conditions resulting in continued
showers or unsettled weather for some days or even several weeks
after the storm has passed.

The dwarfing or stunting of the plants in the field, combined
with the elevation of various parts of the same field as a factor
influencing disease development, was indicated in a series of counts
near Dallas. In a sandy soil a field of some 15 acres sloped from a
small run upward to the boundary road, rising perhaps 20 or 25
feet in 100 yards. On the lower aide, where moisture was best, and
where damp morning air would tend to linger among the trees, the
first count showed 8 per cent spotted bolls, mainly bacterial. At
intervals of 2 or 3 rods other counts were made in passing to the
higher ground. The plants became smaller as the drier parts of
the field were approached, the most stunted plants being probably
one third smaller than those along the low side of the field. In
these plants the count showed 35 per cent of the bolls affected
mainly by the bacterial infection. The greater exposure of the
individual plants to wind-borne spores was evidently an important
factor in the increased degree of spotting, as the spread of plants
was reduced as well as their height, thus permitting free circulation
of air and germ-laden dust among the plants.

In a field located in a damp spot near a creek, the plants were
large enough to meet between the rows, and more than waist-high.
Under usual rainfall such a field would be expected to show con-
siderable anthracnose injury. At this time (same day as dry field
count) only 5 per cent spotted bolls were found, nearly evenly
distributed between the two spot diseases. This field was about
three fourths mile from the preceding one reported.

Finally, the occurrence of weather conditions especially
favorable for an epidemic development of an infrequent disease
may be responsible for serious injuries to specific crops. In the
Rio Grande Valley region of Texas, in Hidalgo County, a consid-
erable area is devoted to cotton grown under irrigation. In June
of this year many of the fields suddenly showed a marked yellow
color of the leaves. Specimens sent to the Experiment Station
were identified as Aecidium Gossypii (Science II. 46: 268. 14 S
1 91 7). The affected fields were visited by the writer in July and
the aecidial irruptions found to be generally past activity and

Blodgett: Weather and crop diseases in Texas 77

commonly invaded by the rust-parasite Tuherculina in so far as the
leaves still attached to the plants were concerned. These were,
however, far fewer in number than those affected earlier, as shown
by the dry leaves under the plants in the rows, which were shed
by the plants before becoming parasitized.

The disease was noted almost simultaneously by Dr. Morton
of Mercedes and County Agent Miller of Edinburg, from both of
whom specimens were sent to the Experiment Station. Field
inspections and interviews four weeks later showed the epidemic
to be past, and no fresh areas developing. It was learned that
about two weeks previous to the observed outbreak of the disease
about a week of showers and cloudy weather had occurred, this
being distinctly unusual in that section. After making allowance
for possible inaccuracy of statement, it seems probable that the
disease was present in cotton fields on the Mexican side of the Rio
Grande, as Dr. Morton learned of ''yellow leaves" occurring in
that locality. Southwest winds prevail during much of the season
in that part of Texas, and invasion of spores from the alternate
host (as yet unknown) or possibly viable sporidia from germinating
teleutospores might be carried over the half mile of river between
the Mexican and American fields. According to Dr. Morton, the
trouble spread northeasterly from Rio Grande City or Sam For-
dyce to Edinburg and Donna, making a total travel of 25 or 30
miles in a couple of weeks or less, according to his observations and
conversations with farmers. '

An interesting ecological detail was learned during a personal
inspection of the fields, namely, that those fields suffered most
which were nearly ready to show first blooms. Fields either older
or younger were less seriously injured. This was shown in the
fields showing effects of serious damage by the presence of bolls
with the involucral bract and calyx carrying aecidial sori; fallen
leaves with the rust areas in abundance were numerous under such
plants.

There may be some close relation between the date of appli-
cation of irrigation water to the field and the appearance therein
of the rust. This point did not come to mind in time to receive
attention while in the field, but may have an important bearing
through the use of the Rio Grande water as a vehicle of transport

78 Semi-centennial of Torrey Botanical Club

of debris of vegetation in which the tehal phase of the rust might
have been distributed. This possibihty would help to explain the
uniformity in degrees of infection existing over entire field, as
though irrigation water had been in some way related to the epi-
demic, as well as the age of the crop plants.

About a week after my visit, Dr. E. W. Olive examined the
affected area and confirmed the above details from his observa-
tions and interviews. In addition he learned from one or two of
the more observing farmers that the aecidial sori had been seen
scatteringly as early as late April of this year, and apparently the
same trouble noticed, though doing no essential damage in other
years. This would indicate that the aecidial stage (and by in-
ference the other stages of the rust) has been present for some time
as a parasite too insignificant as to damage to come to notice as a
''disease" until the special conditions of weather and crop de-
velopment made the outburst this season possible, as an epidemic.
This would appear to be confirmed by the abundance of infections
of the aecidial sori by the secondary parasite Tuberculina, which
at that season would hardly find other hosts (rust) in abundance
(sunflower, cocklebur and Bermuda-grass leaves with rust pustules
appeared to be free from the Tuberculina) .

The sudden cessation of fresh infection with the passing of
favorable weather conditions is in keeping with similar sensitive-
ness among other rust species, and is one of the natural checks to
the spread of such parasites.

The above examples indicate the intimate relation that evi-
dently exists between the healthy development of crop plants and
the injuries caused by invasion of parasitic fungi producing disease
conditions. The influence of climatological changes over con-
siderable areas or during a number of days or weeks has been recog-
nized for a long time, but the direct relation of small variations
in limited areas has been less evident. In connection with diseases
conveyed from season to season in planting seed {e. g., bean an-
thracnose, cotton anthracnose), the saving of such seed from areas
of lightest local rainfall, and during the most favorable period for
seed-ripening becomes a practice of demonstrable value based on
"crop hygiene."

EARLY HORTICULTURAL JOURNALISM IN
THE UNITED STATES

By James Grimshaw Scott

Germantown, Pennsylvania
"Time consecrates and what is gray with age becomes ReHgion."

On the authority of our most accurate Germantown historian,
Edwin C. Jellett, we have it that the pioneer horticultural journal
was the Florist and Horticultural Journal founded in Philadelphia
in 1852 by R. Robinson Scott.

In presenting to you the facts of the beginning of horticultural
journalism in these United States, I have taken the stated "text"
from the pages of the Philadelphia Florist and Horticultural
Journal, the first issue of which came from the press in April,
1852, and the publication of which was suspended in 1855, having
run through part of that year.

Explaining the suspension of publication, the editor printed
the following: "The only apology we have to make for our sus-
pension, to those of our subscribers who paid us promptly their
subscriptions, is, that a greater number have not paid and some,
perhaps many, do not intend to pay."

This shows that the Journal was an indigenous one — not an
exotic, as the circulation editor of a journal of any kind to-day will
advise you that this condition of the finances is normal throughout
the country. We must not censure the delinquents for the state
of affairs entirely, for the publication is often thrust upon them by
the importunities of the publishers and the charm of the journal
so grips the reader that he is loth to cancel his subscription even
when he has no funds with which to pay for it.

This period, say from 1850 to i860, seems to have been the
golden age of horticulture in the United States and the storm area
extended from Massachusetts to Maryland, where many earnest
workers were engaged in planting the wilderness and encouraging
the remainder of the country to cultivate the soil.

79

80 Semi-centennial of Torrey Botanical Club

To the publishers of this pioneer horticultural magazine it
seemed a necessity to have an organ devoted entirely to botany,
horticulture and pomology and of this necessity the Philadelphia
Florist and Horticultural Journal was born. At the time of its
first appearance there were journals that "dabbled" in the news
of the farm and garden but this was the first strictly horticultural
and agricultural journal to be attempted in the New World.

New York too, at this time, had her horticultural dreams and
ambitions and one of the first secretaries of a New York xlorti-
cultural Society was George William Curtis.

Hon. Marshall P. Wilder of Massachusetts, no doubt a kins-
man of our late beloved Prince of Entertainers, "Little Marsh,"
seems to have held the New England front. The Honorable
Nicholas Longworth, of Ohio, was the active spirit in the horti-
culture of the then far western limit of activity.

On the last page of the last issue of the Philadelphia Florist
and Horticultural Journal, we find notes of three other contem-
porary journals, namely, the Western Agriculturist, published in
Pittsburgh; the Homestead, published in Hartford, Conn.; and the
Pennsylvania Farm Journal, published in Philadelphia by Messrs.
Samuel Emlen & Co. and edited by David Wells and A. M.
Spangler. Samuel Emlen still survives in our Germantown and
he of all others was the most helpful spirit in encouraging these
early garden publications. It was he who, with the late John
Jay Smith, steadied the hand of the editor and proprietor of the
pioneer Florist and Horticultural Journal, applauding him when he
ran, lifting him when he fell, and enabling him honorably to pro-
ceed with the work which made a place for the Journal and blazed
the way for those that followed. From that early time till now
the United States has been benefited by the stimulus of splendid
horticultural journals until at the present time we find that
flowers and fruits, deified by the refined ancients under the titles
of Flora and Pomona, have unseated Jove, who grasped with mailed
hand the thunderbolts of Heaven; and to-day, passing the City
Hall of Philadelphia, we read the inspiring announcement, "Food
will win the war — don't waste it."

Reared then in the atmosphere of the stoke-hole among the
tobacco stems and grafting twine in the caboose at the end of the

Scott: Early horticultural journalism 81

greenhouse, we are pleased to salute the pagan deities Pomona and
Flora and brush impatiently aside all other fabulous personifica-
tions.

This pioneer horticultural journalist was besides the pioneer
fern student of the United States and his memory has been kept
green by those beloved publications, the Fern Bulletin and its
successor, the American Fern Journal, in whose pages may be read
the enchanting history of Asplenium ehenoides R. R. Scott, one of
the most noted of the world's ferns.

It gives me great pleasure to revive the memory of the early
Journal and in the brief time at my disposal, I cannot refer to the
splendid magazines that have succeeded it until to-day we are in
possession of the crowning effort in that distinguished quarterly,
the Addisonia of the New York Botanical Garden.

The memory that endures is the consecrated shrine of the
historian and the Greek statue of antique time is of greater value
than the whole of Manhattan Island!

SISYRINCHIUM BERMUDIANA

By Oliver Atkins Farwell

Parke, Davis and Company, Detroit, Michigan

Many botanists have in the past considered the pale-blue-
flowered Sisyrinchium Bermudiana L., of the Atlantic coast, and
the violet-blue 5. iridioides Curtis, of Bermuda, to be conspecific
and have united them under the Linnaean name. Philip Miller,
who cultivated both, side by side, considered them to be amply
distinct and described them separately in the Gardeners Dictionary
in 1768 but applied the Linnaean name to the Bermuda plant and
renamed the Atlantic coast species as 5". angustifolium. William
Curtis, who, like Miller, knew both plants, also considered them
to be distinct and in the Botanical Magazine, plate 94, named the
Bermuda plant S. iridioides; the date of the title page of volume 3
of the Botanical Magazine is 1790 but the printed date on the
plate itself is September i, 1789; the publication of the binomial
must, therefore, date from that of the plate, 1789. Modern
botanists follow the interpretation of Philip Miller by applying
the name Sisyrinchium Bermudiana L. to the plant that is endemic
in the Bermudas but this is contrary to the laws of priority as
expressed in both the Vienna and American codes. Both of these
species were described and illustrated by Plukenet in the Alma-
gestum under his genus Sisyrinchium; likewise by Dillenius in
Hortus Elthamensis under the Tournefortian genus Bermudiana.
Linnaeus in the Species Plantarum, page 954, 1753, combined both
species under the binomial Sisyrinchium Bermudiana, thus pre-
serving to science both of the old generic names under each of
which the species had previously been known. The specific name
Bermudiana perpetuates an old generic name and cannot be con-
sidered as having been given to the species as a geographical name
to indicate the nativity of the species; had that been the idea
actuating Linnaeus he in all probability would have given it the
name bermudiense adopting it from Plukenet providing he had

82

Farwell: SisYRiNCHiuM Bermudiana Linnaeus 83

intended the Bermudian plant to he the type of the species. But
Hemsley has already shown (Journal of Botany 22 : 108-1 10. 1884)
that Linnaeus in all probability had never seen the plant from Ber-
muda. As a matter of fact he made the Bermuda plant his var. /3
and considered it to be of such small categorical importance that
he did not give to it even a varietal designation. That he intended
the Virginia plant to represent typically his S. Bermudiana is
clearly proved by the fact that all references to it were enumerated
under his specific name and description while those referring to the
Bermuda plant were grouped under his unnamed variety jS and by
the fact which is still more to the point, that the explanatory note
with its fuller description was drawn entirely from his " Planta a,"
i. e., the Virginia plant. A careful study of all the evidence seems
to indicate that:

1. Linnaeus probably never saw the plant from Bermuda.

2. The specific name Bermudiana perpetuates an old generic
name and was not used as a geographical name to indicate the
origin of the species; this view per se would prevent the adoption
of the Bermuda plant as the type of the species.

3. The Linnaean descriptions (diagnosis and footnote) are
based upon the plant from Virginia, which must therefore be taken
to be the type of the species.

4. The plant from Bermuda should be known under the first
name applicable to it, S. iridioides Curtis.

THE EFFECT OF ACIDS AND ALKALIS ON
THE GROWTH OF THE PROTO-
PLASM IN POLLEN TUBES

By Francis E. Lloyd*

McGill University

In 1 91 5 it was shownf that the protoplasm of pollen grains
acts in the presence of acids, alkalis, and salts in general accord-
ance with the behavior of the biocolloid gelatin. The evidence
then secured appears to indicate that the amount of swelling is
greater in acids than in alkalis, and less in salts than in pure water*
from which, in the light of MacDougal's experiments, the prepon-
derating protein component of the complex may be inferred. t It
soon became evident, however, that the amount of swelling for
various concentrations of solution of the reagents used was not
constant, and it was found necessary to determine this relation
for an assumed analog, gelatin, and then to find the material and
method by which the comparative behavior of the protoplast in
the living condition could be studied. The results of the measure-
ments of the swelling rates of gelatin in a number of acids and
alkalis, and these in combination with certain salts have been
reported upon in general form.§ These are in brief as follows:

I . The swelling rates differ in both acids and alkalis for dififerent
concentrations. The maximum rates are found at certain con-
centrations above ca. iV/640, which are higher for organic acids
than for inorganic acids. At higher concentrations " repression " ||
occurs. The higher rates are at first at higher concentrations,
but as time elapses the rates at successively lower concentrations

* The writer acknowledges the assistance of a subvention from the Cooper Fund
for Medical Research, McGill University.

t Lloyd, F. E. Carnegie Inst. Wash. Ann. Rep. for ipiS-
% MacDougal, D. T, Science IL 44: 502. 6 O 1916.

& Spoehr, H. A. Science IL 45: 484. 18 My 1917.

§ Lloyd, F. E. Trans. Roy. Soc. Canada. 1917 (In press).

II Procter, H. R., & Wilson, J. A. Jour. Chem. Soc. London 109-110: 307-319-
1916.

84

Lloyd: Growth of protoplasm in pollen tubes 85

overtake them. At certain concentrations below ca. N/64.0, the
rates are lower than those for pure witer, but much more markedly
so for acids than for alkalis.

2. The swelling rates and total swelling in acids is greater than
in alkalis.

3. The swelling rates and total swelling is greater in some
organic acids (citric, malic) than in inorganic acids.

Acetic and tartaric acids appear to be excepted. Malic acid
has a far-reaching effect on gelatin in causing it to fragment at
N/160 and above after one to two hours. This fact, together with
the lower swelling rates at the higher concentrations, suggests
that at these there is a coagulation effect which sets in to repress
swelling.

It thus appears that in trying to establish any analogy between
gelatin (or other emulsoid) and protoplasm, the concentrations of
the reagents to which they are subjected must be considered. For
example, during the increase or decrease of acidity which may take
place in the living tissues, the swelling effects may be alternately
repressed and increased, aside from alterations in the relative
composition of the body fluids due to change in salt, protein, or
other content, such as MacDougal has indicated.*

The determination of growth rates and accompanying phe-
nomena in pollen tubes confirms the expectation that their pro-
toplasm behaves toward the above reagents in many important
respects as does gelatin rather than agar.f The method employed
consists in sowing pollen of Phaseolus odoratus in hanging drops of
the various reagents at different concentrations, associated with
cane sugar in constant concentration, it having been foundt that
the rate of growth of pollen tubes is inversely as the concentration
of cane sugar, the maximum accomplished growth occurring in ca.
20 per cent solution. It has been shown that this is explainable
on the assumption that imbibition by the protoplasm rather than
osmotic pressure is the dominant growth factor. In weaker solu-
tions of cane sugar the pollen tubes burst, the lower the concen-

* MacDougal, D. T. Science IL 46: 269. 14 S 1917.

t The contrary has been found to hold for complex tissues such as cactus stems
by MacDougal and Spoehr. . Proc. Amer. Phil. Soc. 56: 289. 1917 (and other al-
ready cited papers).

X Lloyd, F. E. Carnegie Inst. Wash. Ann. Rep. for 19 16.

86 Semi-centennial of Torrey Botanical Club

tration the more quickly, and for this reason Httle total growth can
be attained, although the initial rates are higher than that at
higher concentrations. The cultures were run in a double series,
and the concentrations of acid and alkali were varied between
iV/25,600 and iV/ioo.

During the course of experimentation it was found that posi-
tive results in terms of growth could be obtained with alkali when
associated with 20 per cent cane sugar, but that acids so associated
caused the pollen grains or tubes to burst. This again indicated
the greater swelling effect of acid over alkali. It was then found
that by increasing the concentration of cane sugar to 40 per cent,
the effect of acids was held in check and that growth proceeded,
bursting taking place only at certain concentrations of acid, but
in the lower of these only after a certain amount of growth had
been attained. With this difference understood, it was shown
that the behavior of the growing protoplasm was otherwise and
in general the same toward both acids and alkalis. Summarily
stated it is as follows:

At certain concentrations of the reagent, growth proceeds more
rapidly than in the control, namely, the pure cane-sugar solution.
The maximum growth occurs for acetic acid at iV/3200, for malic
acid at iV/12,800, and for citric acid at iV/12,800, or perhaps less.
Hydrochloric, formic, and oxalic acids did not afford positive
results in terms of growth, and indeed the evidence for citric acid
was not unequivocal. This was not because they did not produce
increased imbibition in the protoplasm, but probably because of
pathological results which militated against the attainment of
growth. It is important to note that for those acids which gave
the data sought, there was less growth for concentrations above
and below the ones just indicated, and in this we may see a cor-
respondence with gelatin in its maximum swelling response to
certain concentrations already mentioned. The correspondence
is heightened in the growth rates which in low concentrations of
the reagents are lower than in the control. It was previously
shown that essentially the same behavior occurs in alkali, sodium
hydrate having been used.

That a higher concentration of cane sugar must be used with
acids may be due to the already acid condition of the protoplast.

Lloyd: Growth of protoplasm in pollen tubes 87

Growth took place in alkali also in a greater range of concentra-
tions, namely, from iV/400 to iV/25,600, and it was determined
that the Na-ions penetrated the protoplast. In iV/400 the growth
was less than in the control, this, it is possible, being related to an
increase in salts formed. At all other concentrations the amount
of growth was greater than in the control, increasing from the
lowest concentration used to iV/3200 and falling for those still
higher.

The failure to obtain positive results with certain acids in
terms of growth, as above stated, need not, indeed should not, be
interpreted except as indicating that other effects antagonistic to
normal behavior intervene.

We may note especially the bursting of the protoplast beyond
the confines of the cell wall. The weakest point in the pollen-tube
wall is at the apex, and it is here that bursting takes place if it has
not already occurred before growth begins. Bursting is due to the
imbibition of the protoplast beyond the strength of the wall to
confine it, and not, as might be expected, to any change in the wall
itself, such as hydrolysis, since the bursting takes place more
rapidly at concentrations of the reagent which would cause less
hydrolysis.

In acids the bursting takes place within a certain range of con-
centrations, namely, those above that at which maximum growth
takes place and below those at which syneresis of the protoplasm
is caused. Syneresis is quite evident in all the acids studied at
concentrations at or above iV/3200, and it is of more than passing
importance that syneresis occurs in formic, oxalic, and hydro-
chloric acids at lower concentrations (iV/3200 to iV/i6oo) than in
citric, malic, and acetic acids, and was not observed at all in
alkali. It is evident in the course of a short time in the highest
concentrations (iV/8oo to iV/400) but ensues more slowly in the
lower effective concentrations. It should be stated that in all
these the protoplast swells fully when first subjected to them,
completely distending the pollen-grain walls. It then slowly
shrinks.

At the close of shrinkage it can be shown that the protoplasm
is coagulated, for on pressure it breaks out as a cheesy mass. In
this connection it is important to note that at the higher concen-

88 Semi-centennial of Torrey Botanical Club

trations at which bursting occurs, the protoplasm oozes out of
the broken pollen tube in strings, in such a manner as to show that
it has a much higher viscosity than has the protoplasm exposfed
to the same reagent at lower concentrations but which neverthe-
less cause bursting.

It will thus be seen that the maximum swelling of the proto-
plasm is at the concentration of the reagent which causes the most
rapid bursting, and this is higher than the concentration which
causes swelling which can be utilized in growth. The former is
chiefly a physical result, the latter physiological. A glance more
particularly at the behavior of pollen protoplasts toward malic
acid may be taken, this serving as a typical example.

At concentration There occurs:

iV/400 Coagulation and complete syneresis in the course of an hour of all

the pollen grains (100 per cent.).

N/Soo The same but more slowly and less completely (95 per cent.).

A''/i6oo 50 per cent, of the pollen grain shrinks in the course of 3 hours, the

remainder having burst; the protoplasm highly viscous, bursting
in strings after some growth (0.3 unit).

Nls200 Bursting of 90 per cent, after 1-1.5 units growth attained; viscosity

of protoplasm lower than above but still showing coagulation
("clots").

NI6400 3.5 units growth in 5 hours. Some bursting but no coagulation.

22 hours later: alive, no further growth.

A'^/12,800 8 units growth in 5 hours, no bursting; 22 hours: 16 units growth,

alive.

iV/25,600 4 units growth in 5 hours; 22 hours: 5 units growth, alive.

Control: 4.5 units growth in 5 hours; 22 hours: 5 units growth, alive.

In formic acid, bursting unaccompanied by coagulation oc-
curred in iV/25,600 (with 20 per cent cane sugar) after one unit of
growth was attained. Partial coagulation occurred in iV/6400,
more in iV/'3200 and complete in 7V/i6oo.

The above results indicate that the protoplasni of pollen grains
is affected by acids and alkalis in the same fashion as geicitin, and
that the increased swelling caused by such reagents can actually
be used in growth. The extreme sensitiveness of this protoplasm
to low concentrations of acids and alkalis, as evidenced in coagu-
lation and syneresis in the higher, and in the swelling and growth
in the lower, is to be noted. It has become patent that the mech-
anism of growth in more complex plants includes emulsoids which
exhibit swellings at much higher concentrations of acids and

Lloyd: Growth of protoplasm in pollen tubes 89

alkalis* and a final analysis of their relations to growth must, as
it will, include the behaviors of these emulsoids. Similarly in the
animal body, so far as studied in these relations, it is impossible
to analyze the phenomena, and to separate that which occurs in
substances extraneous to the protoplasm {e. g., sarcolemma), and
that which occurs in the protoplasm itself.

* Long, E. R. Bot. Gaz. 59: 491. 1915; MacDougal & Spoehr, as above cited;
Fischer, M. Oedema, 1910.

THE ORIGIN OF THE HAWAIIAN FLORA

By Douglas Houghton Campbell

Leland Stanford Junior University

The Hawaiian Islands afford perhaps the most important
problem in plant distribution that exists anywhere. The most
isolated land area of equal size on the globe, the origin of their
extremely peculiar and interesting flora opens a wide field for
research and speculation.

There is much difference of opinion as to whether or not the
Islands have had at any time connection with any of the great
continental areas. Hillebrand,* whose flora of Hawaii is well
known, and has been followed in the tables given in this paper,
believed that the Islands had always been isolated, having been
thrown up from great ocean depths through volcanic action.
This view has been recently advocated by Muirf as the result of
his studies on the insect fauna of the islands. On the other hand,
Wallace,}: on the basis of the occurrence of certain north temperate
genera in the high mountains of Hawaii, believed that there had
been a land connection with west North America. Recently,
Pilsbry§ has brought forward evidence which he thinks proves
conclusively some ancient connection of the Islands with the
Malaysian region. The peculiar land-snails, so largely developed
in the Islands, are, according to Pilsbry, ancient forms, whose sur-
vival outside of the Pacific Islands is known only in the Malaysian
region.

The writer, up to the present time, has taken it for granted
(largely on Hillebrand's evidence) that the Islands always had
been completely isolated; but the evidence offered by Pilsbry for
an ancient land connection seems very strong, and, moreover, is

* Hillebrand, W. Flora of the Hawaiian Islands. 1888.
t Muir, F. Proc. Haw. Ent. Soc. 3: 198-200. 1916.

X Wallace, A. R. The geographical distribution of animals, i: 447. 1876.
§ Pilsbry, H. A. Mid-Pacific land-snail faunas. Proc. Nat. Acad. Sci. 2:
429-433. 19 16.

90

Campbell: The origin of the Hawaiian flora 91

quite in line with certain facts of plant distribution which appear
to have been overlooked.

The occurrence in the Islands of many hygrophilous liverworts
and the filmy ferns (Hymenophyllaceae) seems to make it practi-
cally certain that, as in the case of the snails cited by Pilsbry, the
presence of these in the Islands can be explained only by considering
them as remnants of the flora of a formerly much more extensive
area connecting the Islands with some ancient continent. These
plants are peculiarly unfitted for transportation over long dis-
tances and it is difficult to see how they could possibly have sur-
vived the exposure to heat and dryness to which they must have
been subjected, assuming that they have come directly from either
the American tropics or the remote tropical regions to the south.
These plants inhabit, for the most part, the cool dark rain-forests
and are quickly destroyed by exposure to the heat and sunshine
of the lower levels.

During the past summer the writer made a brief visit to the
Islands, with a special view to examining tke hepatic flora; and
although the collections made were not as comprehensive as it
was hoped to make them, owing to the remoteness of the col-
lecting grounds, the results tend to confirm Pilsbry's view of a
connection with the Malaysian and Australasian region.

The most conspicuous of the liverworts in the lower forests
are two species of Dumortiera, a genus peculiarly adapted to wet
dark conditions. According to Stephani's* list of the liverworts
of Hawaii, these species are D. trichocephala (Hook.) N. ab E., a
species widely distributed through the eastern tropics, and D.
hirsiita (Sw.) R. Bl. & N., an even more widely spread species.
It is probable, however, that a critical examination of the Hawaiian
plants will show that they are not identical with those species.
The so-called D. trichocephala is certainly quite different from
material of the same species collected in the Malayan region, and
the form attributed to D. hirsuta resembles very closely the Java-
nese D. velutina Schiffn. Stephani states also that the monotypic
genus Wiesnerella, which is closely related to Dumortiera, occurs
in Hawaii, and immature material collected by the writer perhaps
belongs here. This species occurs also in Java, the Himalayas,
and Japan.

* Stephani, F. Hepaticae Sandvicenses. Bull. Herb. Boiss. 5: 840-849. 1897.

92 Semi-centennial of Torrey Botanical Club

The writer found repeatedly a species of Megaceros closely re-
sembling the species first described by him from Java* but after-
wards found abundantly throughout the Malayan region. This
genus, like Dumortiera, is characteristic of very wet, shady local-
ities, and its thin-walled, green spores are certainly not fitted to
being dried up and transported over long distances by the wind.
The same may be said of the related genus, Dendroceros, which
inhabits the dripping rain-forest of the higher altitudes. The
latter was found by the writer only at an elevation of about 4000
feet, in regions of almost constant rain.

Other characteristic thallose liverworts of the upper rain-
forests were species of Pallavicinia, Symphyogyna, and Aneura
(Riccardia). One of the last-named genus appears to be very
close to A. maxima Schiffn. of Java.

It may be safely asserted that there is a marked resemblance
between the liverwort floras of Hawaii and the Malaysian region,
but further material is necessary before the exact degree of re-
lationship can be established.

The Islands at present consist almost solely of volcanic masses,
and it is very evident that the volcanic activity has proceeded from
the northwest to the southeast.

The oldest formations in the north island, Kauai, and part of
Oahu, show much weathering and disintegration, while in the
newest and largest island, Hawaii, volcanic activity is still in
progress.

Hillebrand made a careful study of the distribution of the
vascular plants of the Islands and found that there is a marked
increase in the number of species, especially endemic species, in
the older islands, this being specially marked in Kauai, where pre-
sumably the evolutionary forces have been at work for the longest
time.

The preponderance of the Australasian-Malaysian elements
in the Hawaiian flora, indicated by a study of the liverworts, is
amply confirmed by a comparison with the vascular plants. This
will be sufficiently evident from an examination of the tables
appended, based upon Hillebrand's Flora of the Hawaiian Islands.

It is evident at a glance that the Australasian, Polynesian, and

* Campbell, D. H. Ann. Bot. 21: 469. 1907.

Campbell: The origin of the Hawaiian flora 93

Malaysian genera are much more numerous than the American.
There are forty-five genera of Phanerogams belonging to the
former regions which are entirely absent from the New World;
while only eight genera are exclusively Hawaiian-American. One
genus, Gynandropsis (Capparidaceae) belongs to South America
and South Africa, while three endemic Hawaiian genera of Com-
positae — viz., Argyroxiphium, Wilkesia, and Raillardia are closely
related to certain Californian types.

The Pteridophytes emphasize even more strongly the intimate
relation between the floras of the Australasian and Malaysian
regions and Hawaii. No less than thirty-eight species, absent
from America, are common to the two areas, while only two
species are confined to Hawaii and the American continent.

While it is not unlikely that certain species of Phanerogams
common to Hawaii and the southern Polynesian region nfay
have been introduced in recent times, in most cases the Ha-
waiian species are distinct and peculiar to the Islands. The
cocoanut, taro (Colocasia), sugar-cane, bread-fruit, and some
other cultivated plants were undoubtedly introduced by man, and
it is not unlikely that such useful trees as the kukui (Aleurites
moluccana) and the mountain apple {Eugenia malaccensis) were
also introduced, although now they form almost the entire forest
of the lower elevations.

The American-Hawaiian genera are mostly found in the An-
dean region and as there is considerable evidence of a former con-
nection of South America with the Australasian region, it is pos-
sible that some of these forms may have reached Hawaii from the
south and have survived in the. two extremes of their range, dis-
appearing in the intermediate regions.

While it is extremely probable that some species reached the
Islands since their complete isolation, either by means of ocean
currents, wind, or the agency of migratory birds, this, as condi-
tions are at present, could have taken place only under very excep-
tional circumstances. It is difiicult to see how any of these
agencies would account for the introduction of many plants of the
cool rain-forest, which could hardly survive any such means of
transportation.

As to the' line of connection between Hawaii and some former

94 Semi-centennial of Torrey Botanical Club

continental or sub-continental area, we can only conjecture. An
examination of the water areas existing at present (see map 2,
Century Atlas) shows that between the Islands and North America
there is an enormous and perfectly continuous area of very deep
water which extends to the north and to the south of the Islands,
but leaves an opening on the west which is continuous with a large
area of less depth comprising pretty much the whole of Polynesia.
To the southwest are two very large shallow areas including re-
spectively the Marshall and Caroline Islands, presumably the
remains of large sunken land masses. A chain of similar but
smaller shallows extends to the Malay Archipelago, and it is pos-
sible, at least, that this indicates approximately the line of connec-
tion between Hawaii and some ancient great southern continent —
in short, that Polynesia comprises merely the remnants of a larger
cofttinent, or group of continental islands like Australia.

As to the period at which Hawaii became completely isolated,
this of course can only be guessed. It could hardly have been
earlier than the later Cretaceous or early Tertiary since few of the
modern Angiosperms existed prior to the upper Cretaceous, so far
as we know.

Hillebrand, reasoning from the absence of Conifers, thinks
that the Islands must have been formed "subsequent to the age in
which these were universally distributed." It is quite conceivable
that Conifers may have existed formerly and become extinct as the
result of the extensive volcanic activities subsequent to the isola-
tion of the Islands. As the soils of the Islands at present are
practically exclusively volcanic and are said to be strongly acid,
this might well account for the absence of many plants which may
have been found at an earlier period, but which require different
soil conditions from those now existing.

The subsidence of the assumed ancient Pacific continent per-
haps coincided with the great uplift during the late Cretaceous
when most of the existing mountain systems of western America
came into existence.

Australasian-Malaysian genera occurring in Hawaii, but not in America

Pittosporum Banks.

Pittosporaceae
Ternstroemiaceae
Tiliaceae
Ilicaceae

Eurya Thunb.
Elaeocarpus L.
Byronia Endl.

Campbell: The origin of the Hawaiian flora

Alphitonia Reissek Rhamnaceae

Slrongylodon Vogel 1

„ ^ Y Leguminosae

Mezoneuron Uesf. J

Metrosideros Banks Myrtaceae

Tetraplasandra A. Gray 1 ...

„ ,,. ^ ^ ^....c Araliaceae

Reynoldsta A. Gray J

Gardenia L. >

PleUronia L. >■ Rubiaceae

Coprosma Forst. ^

Scaevola L Goodeniaceae

Cyathodes R. Br Epacridaceae

Emhelia Burm Myrsinaceae

Ochrosia Juss. 1

AlyxiaR. Br. | Apocynaceae

Cyrtandra Forst Gesneriaceae

Myoporum Banks & Sol Myoporaceae

Plectranthus L' Herit. ^

. ^ r Labiatae

Phyllostegia Bent, J

Achyranthes L. . . . Amarantaceae

Wikstroemia Endl Thymeliaceae

Santalum L. 1 ^ ,

^ ^ >• Santalaceae

Exo carpus Labill. J

Vis cum L Loranthaceae

Claoxylon A. Juss. >i

^/gwrz/e^ Forst. r Euphorbiaceae

Antidesma L.
Pseudomorus Bureau 1

P/^iMrM5 Wedd. V Urticaceae

Cypholophus Wedd. J

Anoectochilus Blume .Orchidaceae

Cordyline Commers.
Dracaena Vand.
Astelia Banks & Sol.
Dianella Lam.

JoinviUea Gaud Flagellariaceae

Pritchardia Seem. & Wendl Palmaceae

Pandanus L.

Liliaceae

}

Pandanaceae

Freycinetia Gaud.

Baumea Gaud. 1 <^ ^

\ Cyperaceae

Gahnia Forst. J

Garnotia Brogn Graminaceae

Hawaiian-American genera, not found in the Australasian region
Genus Family

Perroitetia U.B.K Celastraceae

Vallesia Ruiz & Pav Apocynaceae

Nama L Hydrophyllaceae

Jacquemontia Chois Convolvulaceae

Sphacele Benth Labiatae

Hesperocnide Torr. & Gray Urticaceae

Sisyrynchium L Iridaceae

96 Semi-centennial of Torrey Botanical Club

Pteridophytes* common to Hawaii and the Australasian-Malaysian region,
but absent from america

Ophioglossum pendulum 'L.
Marattia Douglasii Baker
Schizaea australis Gaud.
Gleichenia longissima Blume
Acrostichum gorgoneum Kaulf.
Gymnogramme javanica Blume
Vittaria elongata Sw.
Polypodium Hookeri Brack.
P. samoense Baker
P. tamariscinum Kaulf.
P. lineare Thunb.
P. Spectrum Kaulf.
Phegopteris punctata Hillebr.
Aspidium aristatum Sw.
A. caryotideum Wall.
A. truncatum Gaud.
A. terminans Wall.
A. squamigerum Mann
Doodya media R. Br.

Asplenium Nidus L.
Asplenium normale Don
'Asplenium varians Hook. & Grev.
Asplenium contiguum Kaulf.
A. caudatum Forst.
A. horridum Kaulf.
A. spathulinum Hook.
A. Adiantum-nigrum L.
A. polyphyllum Presl
Odontoloma repens Desv.
Microlepia strigosa Presl
M. tenuifolia Metten.
Pteris ex eels a Gaud.
Trichomanes parvulum Poir.
T. meifolium Bory
Lycopodium serralum Thunb.
L. Phlegmaria L.
L. volubile Forst.

Pteridophytes common to Hawaii and America, but not found
elsewhere

Asplenium fragile Presl Pellaea ternifolia Fee

* Nomenclature according to Hillebrand, Flora of the Hawaiian Islands.

UREDINALES OF CUBA

By J. C. Arthur and J. R. Johnston

Purdue University Office of Plant

Lafayette, Indiana Sanitation, Havana

Shoot of Rivina octandra distorted by aecia of Puccinia Rivinae. Photo by

Johnston, 191 7.

97

98 Semi-centennial of Torrey Botanical Club

Introduction

In a letter recently written by Professor F. S. Earle, the well-
known botanist of Cuba, he says: "There is a very varied and
interesting fungous flora here, but 'collecting' is not nearly as good
as in the States. Field work is difficult, and many species are
rare and local. I am curious to see how many rusts you have
from here. There are really a great many, but it will take years
to find them all." Professor Earle speaks with the knowledge of a
mycologist, as he has to his credit some forty published papers
dealing with mycological subjects, issued largely in the score of
years between 1884 and 1904. Since becoming a resident of Cuba
in 1904 he has occasionally gathered specimens of rusts, and he
took a prominent part in the rediscovery of the rare Prospodium
plagiopus at San Marcos in 1909-1910, as stated under that
species.

The study of the Cuban Uredinales, or rusts, resulting in the
present paper, has abundantly confirmed Professor Earle's state-
ment. Although the first collections of rusts were made in Cuba,
doubtless about 1840-45, by Ramon de la Sagra, a few added by
Charles Wright in 1855-7, and by others from time to time, yet
prior to 1 91 5 the known Cuban rust flora would not have much
exceeded half a hundred species. During 191 5 the junior author,
who had come to Cuba the previous year from a residence of three
years in Porto Rico, began to gather material with the intention of
publishing in Spanish a pamphlet on the rusts of Cuba for the
use of students of the native flora. On February 22, 1916, he
wrote to the senior author that "I have apparently so much new
material, or at least new hosts, that I can not handle the subject
properly under conditions here." He then proposed a joint paper
on the rusts of Cuba, in English of course, and the presentation
here made is the result.

A study of available material, together with all information
which the authors have been able to collect, brings the present list
of Cuban rusts up to 112 species, with 28 additional names of rusts
belonging to the form-genera Aecidium and Uredo, which doubt-
less largely represent additional species whose life histories are too

Arthur and Johnston: Uredinales of Cuba 99

imperfectly known to permit of their reference to a true genus,
Even many of the 112 true species have one or more spore forms
yet to be discovered in order to make the Hfe cycle fully known,
and to permit a full technical description.

This first published list of Cuban rusts contains a fairly respec-
table and representative number of species, as is evident by com-
paring with the Porto Rican list of no true species and 42 others
belonging to the form-genera Aecidium and Uredo. The rust
flora of Porto Rico is the best known of all the West Indian islands.
But Cuba is an island of thirteen times the area of Porto Rico,
and with a more varied topography. It is evident, therefore, that
the present showing must be accepted as only a beginning to the
study of the Uredinales of Cuba.

There have been available for the present study about 470
collections, of which 57 were taken from specimens deposited in
phanerogamic herbaria. These collections are represented in the
Arthur herbarium. Only one species has been introduced into
the list from published records, with no specimen available for
examination, and that is the stem rust of wheat (no. 63), the best
known and most cosmopolitan of all rusts.

The largest contributors toward material for a list of Cuban
rusts have been the men employed since 1904 at the Cuban Agri-
cultural Experiment Station (Estacion Experimental Agronomi-
ca), located at Santiago de las Vegas, some score of miles from
Havana. A large portion of the material has also been contrib-
uted by members of the expeditions sent to Cuba by the New
York Botanical Garden, beginning in 1903. In addition to these
some material has come from individual collectors, mostly while
engaged in securing phanerogamic specimens.

Activity in the field of cryptogamic botany at the Cuban sta-
tion began with the accession of Prof. F. S. Earle to the director-
ship of the station in 1904, and the contemporaneous and sub-
sequent appointment of able botanists to other positions.

Professor Earle (i 904-1 906) gave chief attention to the fleshy
fungi. Five collections of rusts are credited to him, representing
as many species, and eight other numbers in association with
other collectors. He retired from the station to his farm at Her-
radura in the Province of Pinar del Rio, eighty miles to the west-

100 Semi-centennial of Torrey Botanical Club

ward, where he still resides and maintains an interest in mycolo-
gical studies.

Professor Charles F. Baker was chief of the department of
botany in the Cuban station (i 904-1 907), and has nineteen col-
lections of rusts credited to him, representing sixteen species, with
three others as associate collector. He has always been an enthu-
siastic and tireless botanical explorer. From Cuba he went to the
Museo Goeldi at Para, S. A., then to the College of Agriculture
at Los Banos, Philippines, of which he is now Dsan.

Mr. Percy Wilson was assistant botanist from July to Decem-
ber, 1904, and collected rusts in connection with Messrs. Earle
and Baker. His more extensive association with the rust studies
on the island is spoken of later.

Sr. Miguel Zarragoitia y O'Donovan, credited in part with one
collection, and with material from phanerogamic collections, was
assistant to Professor Baker, and is now employed at Havana in
clerical work in the Department of Agriculture.

Sr. Manuel Abarca y Vazquez, credited with two collections,
was also assistant to Professor Baker. He is deceased.

Professor Mel. T. Cook, pathologist of the station (1904-1906),
had his chief interest in galls produced by insects, but has six col-
lections to his credit, representing five species. He is author of a
work on Diseases of Tropical Plants (1913), and has for some years
been plant pathologist of the New Jersey College of Agriculture
and of the Experiment Station.

Professor Wm. T. Horne was pathologist at the Cuban Station
( 1 904-1 909), the first two years being assistant to Professor Cook,
and has twenty-three collections of rusts, representing an equal
number of species, credited to him. A few of these collections
were made during a visit to Cuba in the year 191 7. He is now
professor of plant pathology in the University of California.

Mr. H. A. Van Hermann, assistant in horticulture (1904-1906),
and later chief in horticulture (1914-1916), has the credit of two
rust collections, and one other in association with Professor Baker.
He was until recently chief in the Office of Vulgarization, and is
owner of one of the largest nurseries in Cuba, and well acquainted
with the Cuban flora.

Sr. P. Cardin, having one rust collection to his credit, has

Arthur and Johnston: Uredinales of Cuba 101

been Chief of the Department of Entomology at the Station since
1909.

Mr. S. C. Bruner is assistant pathologist at the station, having
been appointed in 1916. He has one rust collection to his credit.

Mr. J. R. Johnston, the associate author of this paper, was
pathologist to the station (1914-1917), and is now Chief, and
especially in charge of pathology and microbiology, in the Office
of Plant Sanitation at Havana. He is credited with 233 collec-
tions, representing 96 species, about five sevenths of the total
number. He has also supplied much general information, espe-
cially in the way of field observations.

The above showing is an interesting indication of the botanical
enthusiasm and devotion of the scientific men that have been
employed by the Cuban Experiment Station. It becomes all the
more noteworthy when it is understood that botanical collecting,
and especially collecting of fungi, is not considered at all necessary
at this station, and that no systematic effort is made to maintain
a cryptogamic herbarium, although incidentally much material
has accumulated.

The second group of contributors toward material for the
present study has consisted of the men forming the expeditions
sent out by the New York Botanical Garden, together with those
joining these expeditions for a part of the time. Beside supplying
specimens, much assistance has, been rendered also by members of
the garden staff in determining the hosts and in other services.
There have also been much general good will and interest dis-
played toward this work.

The first expedition from the garden, in which rusts were
secured, was in 1903. It was composed of Prof. F. S. Earle and
Prof. L. M. Underwood, and was joined for a part of the time by
Prof. E. W. D. Holway. Professors Earle and Holway sailed
from New York on Feb. 26, and reached Santiago de Cuba on
March 5, being joined there by Professor Underwood. After
three days a coasting steamer was taken for Baracoa on the north
shore, a region where Charles Wright collected in the fifties.
Most of the collecting here was done on the slopes of El Yunque.
After three days Professor Holway left the party (cf. Jour. N. Y.
Bot. Card. 4: 81-84. I903)» and took steamer for Gibara, and

102 Semi-centennial of Torrey Botanical Club

staging to the railway at Holquin, proceeded to Havana. On the
way a few hours' stop was made at Santa Clara. In a few days
he left for the north, arriving at Miami, Fla., on March 2. As a
result of this trip Professor Holway contributed 42 specimens,
representing 34 species of rusts. Messrs. Earle and Underwood
left Baracoa on March 19, returning to Santiago de Cuba. After
a few days at Alto Cedro, Professor Underwood sailed for Jamaica
on March 27, and Prof. Earle for New York on the following day
(cf. Jour. N. Y. Bot. Card. 4: 81-85. 1903). Together they con-
tributed four specimens. The results of the expedition were
disappointing, as the time fell in the dry period of an unusually dry
season.

An expedition from the New York Garden spent the time be-
tween Feb. 21 and April 3, 1910, in Cuba. It was composed of
Dr. and Mrs. N. L. Britton and Mr. Percy Wilson, and was joined
at Havana by Prof. F. S. Earle. Most of the time was given to
the Province of Santa Clara (cf. Jour. N. Y. Bot. Card. 11 : 109-
117. 1 910), and incidentally eight specimens of rusts were
secured.

A second expedition in 1910 spent Aug. 24 to Sept. 23, in west-
ern Cuba, exploring the Province of Pinar del Rio. It consisted
of Dr. and Mrs. Britton and Dr. C. Stuart Gager, and was joined
by Professor Earle (cf. Jour. N. Y. Bot. Gard. 11 : 226-236. 1910).
Two specimens of rusts were secured.

In 1911 an expedition consisting of Dr. and Mrs. Britton and
Mr. J. F. Cowell gave the time between Feb. 22 and the end of
March, to an exploration of western and central Cuba (cf. Jour.
N. Y. Bot. Gard. 12: 89-95. 191 1)» securing one rust collection.

The most fruitful expedition in its bearing on the present rust
study was in 1916, when Dr. and Mrs. Britton and Mr. Wilson
devoted the time between Jan. 29 and March 28, largely to the
Isle of Pines. They were joined for a few days by Brother Leon
of the Colegio de la Salle, Havana (cf. Jour. N. Y. Bot. Gard.
18: 64-71. 1916). Heretofore Mr. Wilson had taken an occa-
sional specimen of rust, beginning when connected for a time in
1904 with the Cuban Experiment Station, and continuing to do so
on his many subsequent trips to Cuba. But, having become espe-
cially interested in the rusts of the island while assisting in the

Arthur and Johnston: Uredinales of Cuba 103

critical determination of hosts, he gave considerable attention to
these fungi during the present expedition. The result was that
the two months' exploration yielded some 60 specimens of rusts,
largely from the Isle of Pines, and largely collected by Mr. Wilson,
part of which represent species, as well as hosts, not otherwise re-
ported. This number is second only in size and importance to the
contributions of Mr. Johnston, and considerably larger than the
number secured during the first expedition to the island in 1903,
when Professor Holway devoted special attention to the rusts,
although under unfavorable conditions. Mr. Wilson's intimate
knowledge of the phanerogamic flora of Cuba gave him unusual
advantage as a collector of parasitic species. The whole insular
list, as here presented, has also been made more valuable and
accurate by Mr. Wilson's critical examination of the hosts of many
collections. While engaged in this helpful study from time to
time he detected rusts on the phanerogamic specimens in the her-
barium of the New York Botanical Garden, and in this way added
to the completeness of the presentation.

Other expeditions to Cuba from the New York Garden are not
mentioned here, as they yielded no collections of rusts, although
rusts have later been found on the phanerogamic specimens se-
cured by some of them.

The names of a few persons who are credited with collections,
and who were not connected at any time with the Cuban Experi-
ment Station or the expeditions of the New York Botanical Gar-
den, should be mentioned.

Ramon de la Sagra* came to Cuba from Spain in 1822 and be-
came professor of botany in the university and director of the
botanical garden at Havana. He took great interest in the flora
of the island, collecting extensively all kinds of plants, and securing
the assistance of many specialists in Europe and especially of P. de
Candolle in the determination of the species. By his adminis-
trative ability and his numerous important writings he became
famous throughout the island, and was assisted in his work by
many local collectors. He projected and edited a monumental
folio work in twelve volumes on the physical, political and natural

* For brief biographical account by Ignatio Urban, see Symbolae Antillanae 3 :
117-118. 1902.

f

104 Semi-centennial of Torrey Botanical Club

history of Cuba, himself writing the general introduction and the
part on climate and agriculture. In 1835 he went to Paris, taking
with him his numerous collections. The cryptogams were placed
in the hands of M. Montague, and were described in the ninth
volume of the folio work. The specimens are now in the Museum
of Natural History at Paris. Only three collections of rusts are
accredited to Sagra. Two of these, Pros podium plagiopus (Puc-
cinia plagiopus), and Puccinia poculiformis (P. graminis) are
cited in the volume by Montague, and the third, P. Anthephorae,
is said to have been collected by him. Sagra did not return to
Cuba, and in 1871 died in Switzerland.

Mr. Charles Wright* spent nearly ten years in Cuba, between
November, 1856, and July, 1867, collecting plants, chiefly phanero-
gams. The first expedition was confined to the province of Oriente
and extended from Nov. 25, 1856, to about Sept. i, 1857. Most
of his fifteen numbers of rusts known to the authors were obtained
during this period. Of the later ones only one has- been seen by
the authors, that on Limnanthemum, which was obtained in the
province of Pinar del Rio in December, 1858, and is a form which
has not been collected by any one else in America. The specimens
bear little data, the date of collecting being confined to the years
covering the expedition, rarely to the exact year, and the locality
to ''Cuba," or "in Cuba orientale," if any at all. The rusts form
parts of the sets of fungi to be found in the Kew herbarium in
London, and in the Herb. Curtis and also the Herb. Gray at
Harvard University. The following is a list of the numbers known
to the writers. t

275. ''Puccinia Asteris Schw. on some unknown leaf," in
Fungi Cubenses. The species is very rare in the tropics, and must
be considered a doubtful determination for Cuba. Specimen has
not been seen.

276. Puccinia solida B. & C. on "leaves of Compositae," in
Fungi Cubenses, = P. Synedrellae P. Henn., on Eleuther anther a
ruderalis. Type in Kew has been examined. See no. 109.

* For brief biographical account, by Asa Gray, see Am. Jour. Sci. III. 31:
12-17. 1886; and for an account of Wright's itinerary in Cuba, by L. M. Under-
wood, see Bull. Torrey Club 32: 291-300. ipoS-

t Most of the numbers are cited in the Fungi Cubenses, by M. J. Berkeley and
M. A. Curtis, Jour. Linn. Soc. 10: 280-391. 1869,

Arthur and Johnston: Uredinales of Cuba 105

278. Uromyces gemmatus B. & C, on "the underside of leaves
of Convolvulus,'" in Fungi Cubenses. Host is Jacquemontia nodi-
flora. Type in Kew has not been seen, but apparently the same
collection in Herb. Curtis was studied. See no. 51.

279. Uromyces appendiculatus Lev., on "leaves of Legumi-
nosae," in Fungi Cubenses. The collection has not been seen,
but as the species is common in Cuba, it is doubtless correctly
referred. See no. 44.

281. Puccinia obliqua B. & C, on "leaves of some plant re-
sembling chickweed," in Fungi Cubenses. Fragment from Kew
sent without number, which may be this one, has been examined.
The host is reported from Kew as probably Metastelma penicil-
latum. See no. 89.

282. Cited as the second number under Uromyces gemmatus
in Fungi Cubenses. Fragment of the collection from Kew, and
also part of specimen in Herb. Curtis, have been examined.
It is Puccinia Gouaniae Holw., H, on Gouania polygama. See
no. 79.

283. Puccinia heterospora B. & C, on "the leaves apparently
of some malvaceous plant," in Fungi Cubenses. The collection
has not been seen, but it is probably correctly referred. See
no. 81.

284. Puccinia deformata B. & C, on " Olyra latifolia, January,"
in Flora Cubenses. The type at Kew has not been examined, but
there is no doubt of the correctness of the naming. See no. 58.

288. Puccinia obliqua B. & C. The number is not cited in
Fungi Cubenses. A small leaf from the Kew herbarium has been
examined and the collection appears to be the same as the one
given above as 281. See no. 89.

480. The number is not cited in Fungi Cubenses. A specimen
in Herb. Curtis, without data other than the number, has been seen
and determined as Ravenelia portoricensis Arth., on Cassia emar-
ginata. See no. 23.

596. Puccinia deformata B. & C, on Olyra latifolia. The
number is not cited in Fungi Cubenses. A specimen in Herb.
Curtis has been examined. See no. 58.

720. Trichobasis euphorbiaecola B. & C, on "leaves of some
Euphorbia," in Fungi Cubenses. The fragment of this collection

106 Semi-centennial of Torrey Botanical Club

from the Herb. Curtis, which has been seen, shows uredinia, and
it is doubtless to be refererd to Uromyces proeminens. See no. 48.

. 727. Trichobasis labiatarum Lev., on "leaves of Labiatae," in
Fungi Cubenses. A portion of the collection from the Herb.
Curt, has been seen, but the species represented remains uncertain.
It may be some species on LeonoHs, Hyptis, or Salvia.

730. Puccinia Cynanchi Schw., on "the leaves and stem of
some Asclepiad," in Fungi Cubenses. The collection has not been
seen, although the type of P. Cynanchi from Surinam has been
examined, and the Wright collection may well be the same species.
It is now referred to P. Gonolobi. See no. 88.

756. Aecidiiim Rivinae B. & C, on "racemes of Rivina octandra,^*
in Fungi Cubenses. The collection has not been seen, but there
is no doubt regarding the names. The species is now referred to
Puccinia Rivinae (B. & C.) Speg. See no. 74.

929. A collection in the Herb. Curtis, which has been examined,
is labelled Aecidium Nymphaearum DC, on Limnanthemum
Grayanum Griseb., and doubtless correctly so. The species is
now referred to Puccinia Scirpi. See no. 69.

Mr. Otto E. Jennings, accompanying a natural history expedi-
tion* from the Carnegie Museum of Pittsburgh, Pa., collected in
the Isle of Pines from May 5 to May 26, 1910. One rust collection
(cf. no. 89), and one phanerogamic specimen bearing a rust (cf. no.
60), were obtained at this time.

Beside the above collectors of fungi, who have enriched our
knowledge of the rust flora of Cuba, there are some fourteen botan-
ists, whose names appear in the following list in connection with
phanerogamic specimens from the island, found to bear rusts.
Altogether half a hundred botanists are represented as field col-
lectors in the present account of the rusts of Cuba, to whom is due
the credit of making material available as a basis for this first
account of the Cuban rusts.

Comparison of the list as it now stands with the list of rusts
for Porto Rico, bringing the last published account of the latter
up to date in order to make the two more accurately comparable,
shows a close agreement in the number and kind of genera and in

* For some account of the expedition and of the topography and fioristic condi-
tions on the Isle of Pines, see Am. Fern Jour, i: 129-136. 191 1; and Ann. Car-
negie Mus. 11: 19-290. 1917.

Arthur and Johnston: Uredinales of Cuba 107

the number of species in each genus, as well as in the comparative
numbers of long- and short-cycle forms. The differences are only
such as might be expected from incomplete exploration. The
greater total number of species now known for Porto Rico is
chiefly accounted for by the greater number of unconnected
Uredo-iorms, which have been found in that island. Whether
this array of forms still under the genus Uredo indicates that the
flora of Porto Rico is more tropical than that of Cuba, or whether
it is better known for such forms, is doubtless debatable.

A comparison of the Cuban with a continental area presents
marked differences. The most instructive comparison at present
possible is that with Guatemala. A recent study of the Uredi-
nales of Guatemala, not yet published, gives a list of species about
as complete for that country, as that here presented for Cuba.
Nearly a third of the area of Guatemala, especially the northern
part in the Department of Peten, is unrepresented by collections.
The remainder of the country is of about the area of Cuba, but of a
more varied topography, having many high mountains. The
climate is doubtless somewhat more tropical that that of Cuba,
although northern species find congenial conditions in the high
altitudes. Thus the genera Melampsora, Melampsoridiuniy Puc-
ciniastrum, Uropyxis, and Phragmidium, not represented in Cuba
or Porto Rico, are to be found there. Also the forms under Uredo
are only about one half those in Cuba and one third those in Porto
Rico. Both classes of facts indicate certain less tropical aspects
of the flora, but nevertheless, they are aspects that may have to do
with the boreal features of the mountainous part of the flora.
There are, however, three genera, that is, treating the correlated
Uromyces and Puccinia as a single genus, which can be taken as
comparable factors to indicate differences between the insular and
continental floras.

Number of species of rusts

Cuba

Porto Rico

Guatemala

Coleosporium

5

3

8

Ravenelia

9

10

20

76

74

90

87

178

108 Semi-centennial of Torrey Botanical Club

The table indicates the presence in Cuba or in Porto Rico of
only half as many species in each of the three genera, Coleosporium^
Ravenelia, and Puccinia-Uromyces, or in all combined, as occur in
Guatemala. The difference in abundance may be ascribed to the
differences between insular and continental conditions, or to the
diversity of topography, or to both factors combined.

The only genera represented in Cuba, not found in the other
West Indian islands, or the nearby continent, are Sphaerophrag-
mium and Uromycladium, both based somewhat doubtfully upon
material needing further field observations and study. The
seemingly greater similarity to the rust flora of southern Florida
and the regions bordering the Gulf of Mexico, than is shown by the
rust flora of Porto Rico, has been previously pointed out,* as well
as the surprisingly large number of short-cycle species. To go
into a more detailed comparison of the Cuban rust flora with that
of other regions is not likely to be particularly profitable at the
present time owing to the imperfect data available, not only for
Cuba, but even more so for most regions with which it might be
compared.

In the following enumeration twelve species are described as
new, and five species are transferred to other genera, making new
combinations. Most of the changes in genera are due to finding
additional spore forms, but some are in the nature of quite new
discoveries, as in the case of Uromyces cristatus, which was em-
balmed under the name of Uredo, although not itself uredinial,
or belonging to a species having a uredinial stage.

The list introduces 15 species new to the North American flora,
of which 10 species are exclusively Cuban, so far as present knowl-
edge extends, the others being mostly South American forms.

In conclusion it may be said that the present enumeration of
140 species of Cuban rusts must be considered only the basis for a
thoroughly scientific and economic exploration of the island for
this group of obligate parasites. When sufficient taxonomic data
are finally accumulated the still more interesting task of studying
the species in relation to their distribution, the abundance from
year to year, their origin on the island, and their relation to eco-
nomic problems, can be taken up with interest and profit. The

* Arthur, Rusts of the West Indies. Torreya 17: 24-27. 1917.

Arthur and Johnston: Uredinales of Cuba 109

study of the rusts opens a field that is Hkely to prove very attrac-
tive to Cuban scholars, as it has been to others not so familiar with
insular conditions.

Map of Cuba, with scale of miles, showing the provinces: A, Pinar del Rio; B,
Habana; C, Matanzas; D, Santa Clara; E, Camagiiey; F, Oriente; G, Isle of Pines.
Some of the principal cities and towns are also indicated.

The microscopical study of the material on which this paper is
based was done in the laboratory of the botanical d-epartment of
the Purdue University Agricultural Experiment Station at
Lafayette, Indiana, as part of the preliminary work on the rust
portion of the North American Flora. Thanks are due to Pro-

H. S. Jackson, chief of the department, and to his assistants, for
their assistance.

Family : Coleosporiaceae

I. CoLEOSPORiUM Elephantopodis (Schw.) Thiim. Myc. Univ.

953. 1878.
On Carduaceae:

Elephantopus mollis H. B. K., El Yunque, Baracoa (Prov.
Oriente), March 12, 1903, Holway; La Cunagua, Isle of
Pines, Feb. 19, 191 6, II, Britton, Britton & Wilson
14554; San Pedro, Isle of Pines, Feb. 12-March 22, 1916,
II, Britton, Britton & Wilson 15808; Baracoa (Prov.
Oriente), April 14, 15, 1916, Johnston 504, 507.
The species is heteroecious, having aecia on leaves of pine.
In tropical regions it is doubtless maintained by the repeating
urediniospores. It occurs in Porto Rico, Jamaica, and St. Vin-
cent, but is more abundant on the continents, both north and
south.

110 Semi-centennial of Torrey Botanical Club

2. CoLEOSPORiUM Vernoniae Berk. & Curt. Grevillea 3: 57.

1874-
On Carduaceae :

Lachnorhiza piloselloides A. Rich., San Pedro, Isle of Pines,
Feb. I2-March 22, 191 6, II, Britton & Wilson 147 13.
The first record of the species for the West Indies. The host is.
also new for the species. Northward aecia occur on pine leaves.

3. COLEOSPORIUM Ipomoeae (Schw.) Burr. Bull. 111. Lab. Nat.

Hist. 2: 217. 1885.
On Convolvulaceae :
Ipomoea cathartica Poir. (/. acuminata R. & S., Pharhitis
cathartica Choisy), Rincon (Prov. Habana), Sept. 26,

1 91 5, Johnston 156.

Ipomoea mutahilis Lindl. (/. Learii Meissn. not Paxton)
Vedado (Prov. Habana), Dec. 6, 191 6, Johnston Q26;
Consolacion del Sur (Prov. Pinar del Rio), March, 191 7,
II, Home.

Ipomoea stolonifera (Cyrill.) Poir., Siguanea, Isle of Pines,
Feb. 26, 1916, II, Britton, Britton & Wilson 14936.

Also found in the phanerogamic herbarium at the N. Y. Bot.
Garden on Jacquemontia tamnifolia (L.) Griseb., collected by
Eugenio Cuesta 329, at Pinar del Rio, December, 1911.

The species is heteroecious, with aecia on leaves of pines, but
in tropical regions is probably maintained by the repeating uredin-
iospores. In the West Indies it is also known from Porto Rico
and St. Croix, but is more common on the continents to the north
and south.

4. CoLEOSPORiUM Plumierae Pat. Bull. Soc. Myc. Fr. 18: 178.

1902.
On Apocynaceae:

Plumiera emarginata Griseb., Limones Cienfuegos (Prov.
Santa Clara), Nov. 4, 191 5, II, Johnston 216; Marianao
(Prov. Habana), Oct. 31, 191 5, II, Johnston 236, Feb. 6,

1 91 6, Johnston 441; Caleta Cocodrilos, Isle of Pines,
March 8, 191 6, II, Britton, Wilson ^ Leon 1S300.

Plumiera obtusa L., Santiago de las Vegas, June 21, 1906,
Cook.

Plumiera rubra L., Santiago . de las Vegas, Sept. 4, 1904,

Arthur and Johnston: Uredinales of Cuba 111

Baker 1378, May 9, 1906, Baker (Barth. Fungi Columb.

2217), April 4, 19,06, Home 18, June 21, 1906, Cook;

Limones Cienfuegos (Prov. Santa Clara), Nov. 4, 191 5,

II, Johnston 227,
This West Indian rust is undoubtedly heteroecious, but with
aecia yet unknown. The telia are rarely produced and the con-
tinuance of the species is probably by urediniospores. It is also
known from Porto Rico and Guadeloupe.

5. CoLEOSPORiUM EuPATORii Arth. Bull. Torrey Club 33: 31.

1906.
On Carduaceae:

Eupatorium macrophyllum L., El Yunque, Baracoa (Prov.
Oriente), March 12, 1903, II, Holway; Taco Taco (Prov.
Pinar del Rio), Sept. 17, 191 6, II, Johnston 87^.
This species is undoubtedly heteroecious, like other species of
the genus. It is at present known only in the uredinial stage, and
there is much likelihood that when more fully studied may be
united with earlier named species on other hosts. It was also
detected in the phanerogamic herbarium of the N. Y. Bot. Garden,
on same host from Trinidad Mountains (Prov. Santa Clara),
March 6, 1910, II, Britton & Wilson 5134. It is also known from
Guatemala, Nicaragua, and from South America.

Family: Uredinaceae (Melampsoraceae)

6. Phakopsora Vitis (Thiim.) Syd. Hedwigia Beibl. 38: 141.

1899.

Physopella Vitis Arth. Result. Sci. Congr. Bot. Vienne 338.
1906.
On Vitaceae :

Vitis vinifera L., Havana, March 21, 1903, Holway; San-
tiago de las Vegas, May 13, 1916, Johnston 701.
This imperfectly known rust has not been found on any native
grape in America. The collections are treated under the name
Vitis vinifera, but no adequate examination of the cultivated hosts
has been made to substantiate this assignment. The species is
abundant in Porto Rico and Jamaica, as well as in the warmer
parts of North and South America and Japan, and is often quite
injurious to cultivated grape vines.

112 Semi-centennial of Torrey Botanical Club

7. Phakopsora Vignae (Bres.) Arth. Bull. Torrey Club 44: 509.

1917.

Uredo Vignae Bres. Rev. Myc. 13: 66. 1891.
Uredo concors Arth. Mycologia 7: 330. 1915.
Physopella concors Arth. Mycologia 9: 60. 1917.
On Fabaceae:

Genus and species undet., Herradura (Prov. Pinar del Rio),
March, 1917, II, Home.
Only uredinia of this species are yet known. The type of
Uredo Vignae came from St. Thomas and of U. concors from Porto
Rico. The species will doubtless be found eventually on other
West Indian islands.

8. Phakopsora Aeschynomenis Arth. Bull. Torrey Club 44:

509. 1917.

Uredo Aeschynomenis Arth. Bot. Gaz. 39: 392. 1905.
Physopella (?) Aeschynomenis Arth. N. Am. Flora 7: 104. 1907.
On Fabaceae:

Aeschynomene americana L., Santiago de las Vegas, Nov. 6,
191 6, II, Johnston gzo.
Only uredinia of this species are yet known. It occurs also in
Porto Rico, as well as in Mexico and South America.

9. Cerotelium Fici (Cast.) Arth. Bull. Torrey Club 44: 509.

1917.

Uredo Fici Cast.; Desmaz. PI. Crypt. 1662. 1848.

Uredo ficicola Speg. Anal. Soc. Ci. Argent. 17: 120. 1884.

Uredo ficina Juel, Bih. K. Sv. Vet.-Akad. Handl. 231°: 25. 1897.

Uredo moricola P. Henn. Hedwigia 41 : 140. 1902.

Physopella ficina Arth. N. Am. Flora 7: 103. 1907.

Physopella Fici Arth. N. Am. Flora 7: 103. 1907.

Kuehneola Fici Butler, Ann. Myc. 12: 76. 1914.
On Artocarpaceae :

Ficus Carica L., Santiago de las Vegas, March 13, 1906,
Home 75, July 21, 1906, Cook; Limones Cienfuegos (Prov.
Santa Clara), Nov. 4, 1915, II, Johnston 215; Omaja
(Prov. Oriente), March 30, 191 6, II, Johnston 547; Paso
Estancia (Prov. Oriente), May 3, 191 6, II, Johnston 685.
Ficus Combsii Warb., San Diego de los Banos (Prov. Pinar
del Rio), Feb. 7, 191 5, II, Johnston 175; Minas (Prov.

Arthur and Johnston: Uredinales of Cuba 113

Camagiiey), Dec. 2, 1915, Johnston 371; Marianao
(Prov. Habana), Feb. 6, 1916, II, Johnston 443.
The full life history of this rust is not known. No telia have
been found in America, and no pycnia anywhere. It is cosmo-
politan throughout the tropics, but in the West Indies is only re-
corded from Cuba and Porto Rico.

10. Cerotelium Gossypii (Lagerh.) Arth. Bull. Torrey Club 44:
510. 1917.

Aecidium desmium Berk. & Br. Jour. Linn. Soc. 14: 95. 1873.
Uredo Gossypii Lagerh. Jour. Myc. 7: 48. 1891.
Kuehneola Gossypii Arth. N. Am. Flora 7: 187. 1912.
On Malvaceae:
Gossypium acuminatum Roxb., Santiago de las Vegas,

March, 1903, II, Underwood & Earle i^g.
Gossypium sp., Santiago de las Vegas, August, 1904, II,
Baker (Barth. Fungi Columb. 2480).
This rust is still imperfectly known, as the pycnia have not
been found. It is rather common upon wild and cultivated cot-
tons in both hemispheres, but in the West Indies is reported only
from Cuba and Porto Rico.

11. Kuehneola malvicola (Speg.) Arth. N. Am. Flora 7: 187.
1912.

Uredo malvicola Speg. Anal. Soc. Ci. Argent. 17: 124. 1884.
Uredo Hibisci Syd. Hedwigia Beibl. 40: 128. 1901.
On Malvaceae:

Hibiscus syriacus L., Santiago de las Vegas, March 14,

1906, II, Home 14.
Malvaviscus Sagreanus A. Rich., Santiago de las Vegas,
Oct. 23, 1906, II, Johnston Q04.
An imperfectly known species, the pycnia not having been
seen. It is not an uncommon species in the southern United
States, Central America, and parts of South America, but from
the West Indies only one other collection is known, which was
obtained by Whetzel & Olive in Porto Rico, on Malache scabra
(Mycol. 9: 63. 1917).

114 Semi-centennial of Torrey Botanical Club

12. Cronartium notatum (Arth.) comb, no v.
Uredo notata Arth. Mycologia 9: 89. 191 7.

On Malpighiaceae :

Byrsonima crassifolia (L.) H. B. K., San Marcos (Prov.
Santa Clara), Nov. 18, 1915, II, Johnston 186; Las Tunas
(Prov. Oriente), March 29, 1916, ii,' III, Johnston 545.
The rust has previously only been known from Porto Rico,
and in the uredinial stage. The fine specimen obtained by Mr.
Johnston at Las Tunas provides the following characters for the
telial stage,

Telial columns hypophyllous, somewhat grouped, rather
numerous, cylindrical, moderately stout and short, 1-1.5 mm. long,
90-145 /X broad, chestnut-brown; teliospores oblong, 17-26 by
45-64 ju, obtuse or truncate at both ends; wall yellow, 2-3 thick,
smooth.

The species differs from C. Byrsonimatis P. Henn., known from
South America on B. coccolobifolia, by the somewhat larger uredin-
iospores, with their walls less thickened at apex, the conspicuous
development of imbricated paraphyses, and by the thicker-walled
teliospores.

It might be well to call attention to the possibility that this
rust, instead of being heteroecious, as in species of temperate
regions, may be autoecious. There is an aecial form in Mexico
on this host, first described as Endophyllum singulare D. & H.,
which from its morphology may well belong here.

13. Cronartium Wilsonianum sp. nov.

On Vitaceae:

Cissus rhomhifolia Vahl, San Juan, Isle of Pines, March
15, 17, 1916, II, III, Britton, Britton & Wilson 15552.
Uredinia chiefly hypophyllous, numerous, scattered, sometimes
crowded in small groups, round, small, 0.2 mm. or less in diameter,
dehiscent by a central rupture, soon wide open, pulverulent, very
pale straw-color; paraphyses apparently free, peripheral, incurved,
clavate, 13-15 by 35-40 ju, the wall very pale yellow or colorless,
thin, about i^, smooth ; urediniospores broadly ellipsoid or obovoid,
18-21 by 24-27/1; wall pale yellow or colorless, thin, 1.5 /x or less,
sparsely and very sharply and prominently echinulate, the pores
obscure.

Telial columns hypophyllous, 2-3 mm. in length, about ti
thick, filiform, dark chestnut-brown; teliospores terete or fusiform,

Arthur and Johnston: Uredinales of Cuba

115

10-13 by 48-61 fjLy usually obtuse at both ends; wall pale cinnamon-
brown, thin, about i ju, smooth.

A distinctive species, appearing in abundance over the large
leaves of the host. The form of the uredinial envelope is especially
interesting, free paraphyses being a novelty in the genus.

A uredinial specimen on the same host was collected by Lager-
heim near Quito, Ecuador, June, 1890, which was given a her-
barium name by the collector not established by description or
publication.

It is doubtful if this is a heteroecious rust, like the members
of the genus occurring in northern regions. Information to com-
plete the life history will be of special interest.

The species is named in recognition of the botanical services of
Mr. Percy Wilson of the N. Y. Botanical Garden in making known
the flora of Cuba, and especially the rust flora. Mr. Wilson's
numerous collections of Cuban rusts, made largely in 191 6, added
much to previous knowledge, and his patient and critical examina-
tion of the hosts for most of the collections in this list has added
immensely to the accuracy and value of their citation.

14. Cionothrix Cupaniae sp. nov.

On Sapindaceae:

Cupania americana L., Ceballos (Prov. Camagiiey),

March 24, 191 6, Johnston 668.
Cupania glabra Sw., Paso Estancia (Prov. Oriente), May 3,
1916, Johnston 6'/g, 6q4 (type), 6g6.

Telial columns hypophyllous, in groups on discolored and some-
what thickened areas, 3-5 mm. across, a mammiform swelling
forming the base from which each column arises, cylindrical,
short, rigid, 0.3-0.5 mm. long, 19-35M wide, colorless; teliospores
fusiform, 7-15 by 40-65/x, tapering at both ends; wall colorless,
thin, ifi or less, colorless.

No indication of uredinia were found on the type, no. 6g4,
or other collections, and for this reason the species is issued under
the short-cycle genus, Cionothrix, even though pycnia w^ere not
seen. Sections of young telia show the young catenulate spores
arising in a layer, like cylindrical hyphae with transverse septa,
and with no indication of peridium or paraphyses.

116 Semi-centennial of Torrey Botanical Club

Family : Aecidiaceae (Pucciniaceae)

15. Ravenelia Indigoferae Tranz. Hedwigia 33: 369. 1894.
On Fabaceae:

Indigofera suffruticosa Mill. (/. Anil L.), Baracoa (Prov.
Oriente), March 9, 1903,' II, Holway; Santiago de las
Vegas, March 15, 1905, II, III, and April 5, 1906, II, iii,
Home, Jan. 15, 1907, II, III, Baker 3055 (Barth. Fungi
Columb. 24^5), July 15, 1915, II, Johnston J57; Conso-
lacion del Sur (Prov. Pinar del Rio), Feb. 2, 191 5, II,
Johnston 141; Santiago de las Vegas, Nov. 29, 191 7, II,
III, Johnston p5j.
The species, like all other rusts referred to the genus Ravenelia
in this paper, is autoecious, the life cycle including pycnia, uredinia,
and telia, although no pycnia have yet been seen in this species.
The other West Indian islands where it has been found are Jamaica,
Porto Rico, and Bermuda. It also occurs in Mexico and South
America.

16. Ravenelia Piscidiae Long, Jour. Myc. 12: 234. 1906.
On Fabaceae:

Ichthyomethia Piscipula (L.) A. S. Hitchc, Ensenada de
Siguanea, Isle of Pines, Feb. 18, 1916, II, III, Britton,
Wilson & Selhy 14533, Feb. 25, 191 6, II, III, Britton
&' Wilson 1484Q; San Juan, Isle of Pines, March 15, 17,
191 6, II, III, Britton, Britton & Wilson 15449.
This rust heretofore has been known only from southern
Florida. No pycnia have yet been discovered for the species.

17. Ravenelia Lonchocarpi Lagerh. & Diet. Hedwigia 33: 46.

1894.
On Fabaceae:
Lonchocarpus latif alius H. B. K., Ceballos (Prov. Cama-
giiey), March 24, 1916, II, 520, 530, April 6, 1916, II,
62^, 665, 666; Baracoa (Prov. Oriente), April 14, 1916, II,
638, all by Johnston.
Heretofore the species has been known only from Brazil, and
on Lonchocarpus campestris. The ample material supplied by
Mr. Johnston shows an abundance of uredinia, agreeing closely
with the uredinial part of the original description, no Brazilian
specimen having been seen. No pycnia or telia could be detected.

Arthur and Johnston: Uredinales of Cuba

117

1 8. Ravenelia siliquae Long, Bot. Gaz. 35: 118. 1903.

On Mimosaceae :

Vachellia Farnesiana (L.) W. & A. {Acacia Farnesiana
Willd.), Santiago de Cuba, April 30, 191 6, II, Johnston
684.

This collection, the first one seen from the West Indies, shows
the usual abundance of uredinia on the pods. The species also
occurs in Central Mexico. No pycnia or telia are known for it,
and it has until recently been taken only on the fruit of the host.
Long (Bot. Gaz. 64: 64. 191 7) reports it on branches and leaves.

19. Ravenelia Pithecolobii Arth. Bot. Gaz. 39: 394. 1905.

On MlMOSAQEAE :

Pithecolobium tortum Mart., Sierra de los Ceballos, Isle of
Pines, March 2, 1916, III, Britton & Wilson 15358.
This is the first record of the species for the West Indies, being
previously known from southern Florida and Mexico, and is also
the first record for this species of host. The pycnia of the species
are yet unknown.

20. Ravenelia Lysilomae Arth. Bot. Gaz. 39: 392. 1905.
Dendroecia Lysilomae Arth. Result. Sci. Congr. Bot. Vienne

340. 1906.
On Mimosaceae:

Lysiloma bahamensis Benth., Punta Sabanilla, Cienfuegos

Bay (Prov. Santa Clara), Feb. 24, 1910, III, Britton,

Earle <Sf Wilson 4586.
This collection shows telia but no uredinia, although a few
urediniospores are present. The species was originally placed in
the genus Dendroecia on evidence similar to this. Dietel states
(Beih. Bot. Centr. 20: 375. 1906) that well-developed uredinia
were found by him upon the type collection, made by E. W. D.
Holway at Iguala, Mexico, on L. tergemina. A more ample por-
tion of this collection than was originally available has enabled the
senior author to confirm the statement made by Dietel. On this
evidence, which completes the life history, the species is returned
to the genus Ravenelia, where it evidently belongs. The above col-
lection gives the first station known outside of Mexico, and adds
another species of host.

118 Semi-centennial of Torrey Botanical Club

21. Ravenelia HuMPHREYANA P. Henn. Hedwigia37:278. 1898. •
Ravenelia pulcherrima Arth. Bot. Gaz. 39: 395. 1905.

On Caesalpiniaceae (Cassiaceae) :

Caesalpinia bahamensis Lam. (C Rugeliana Urban), Santa
Clara to Loma Cruz (Prov. Santa Clara), March 22,
1911, II, Britton, Britton & Cowell 10225 (host no. 10224).
Poinciana pulcherrima L. {Caesalpinia pulcherrima Sw.),
Holguin (Prov. Oriente), March 17, 1903, II, III, Hol-
way; Soledad (Prov. Santa Clara), Nov. 4, 191 5, II,
Johnston 221; Santiago de las Vegas, Nov. 3, 191 7, II,
Johnston g^o.

The pycnia for this species have not yet been found. Type
collection for the species came from Mexico. The species is
known also from Jamaica and Guatemala. The first host named
is a new one.

22. Ravenelia cubensis sp. nov.

On Caesalpiniaceae:

Cassia robinaefolia Benth., Cacocum (Prov. Oriente),
April 6, 1916, Johnston 531.

Uredinia amphigenous, scattered or somewhat grouped, round,
0.2-1 mm. across, subepidermal, rather tardily naked, cinnamon-
brown, pulverulent, ruptured epidermis conspicuous; paraphyses
none; urediniospores ellipsoid or obovoid, 16-19 by 23-26/z; wall
golden- or light cinnamon-brown, rather thin, 1-2^1, thicker at
apex, 3-6/^, moderately and finely echinulate, the pores 4, equa--
torial.

Telia unknown.

This rust is so evidently a Ravenelia^ that it is unhesitatingly
described as such, although no telia are yet known.

23. Ravenelia portoricensis Arth. Bull. Torrey Club 31: 5.

1904.

On Caesalpiniaceae:

Cassia emarginata L., without locality or date, II, Charles
Wright 480; Santiago de Cuba (Prov. Oriente), March 6,
1903, II, Holway; same, April 30, 191 6, II, Johnston 68 j.
Heretofore this rust has been known chiefly from Porto Rico.
One collection is recorded from Jamaica, and Mr. Percy Wilson
has communicated two specimens found in the phanerogamic her-

Arthur and Johnston: Uredinales of Cuba 119

barium of the New York Botanical Garden, one from Barahona,
St. Domingo, May, 1910, II, Pater Fuertes igz, and the other from
San Michel, Hayti, Aug. 5, 1905, ii, Nash & Taylor IJQJ. A part
of a collection made by Charles Wright in Cuba, bearing no data
except the number 480, has been sent by Dr. W. G. Farlow from
the Curtis collection, which proves to be the uredinial stage of this
species. All collections of the species so far seen have been found
to be on Cassia emarginata.

24. Uromycladium (?) cubense sp. nov.
On Mimosaceae:

Mimosa pigra L. (ikf. asperata L.), Soledad Cienfuegos
(Prov. Santa Clara), Nov. 6, 1915, Johnston igi.

Telia amphigenous, emaculate, scattered, oval or oblong,
0.2-0.4 by 0.5-1 mm., subcuticular, soon naked, pulverulent,
chestnut-brown, ruptured cuticle conspicuous, teliospores triangu-
lar-obovate, or short-pyriform, length and breadth the same,
15-18 jjLy remaining attached to pedicel in groups of 1-4 (usually 2) ;
wall at sides colorless, smooth, and thin, i ^t, above chestnut-
brown, finely and closely verrucose, 1.5-2 thick; pedicel in upper
part . firm, remaining attached to spores, chestnut-brown, 7-9 ju
broad, 7-10 At long, the wall about 1.5 ^ thick, part below colorless,
thin-walled, collapsing and readily falling away.

This unique rust does not drop readily into any genuo. It is
placed tentatively in Uromycladium, although the spores are not
clearly arranged in series on the pedicel, and no cyst is present.
There are other variations, as for instance, the lack of an evident
germ pore, as well as the absence of uredinia and pycnia. More-
over, Uromycladium has heretofore been accounted a peculiarly
Australian genus.

25. Calliospora Farlowii Arth. Bot. Gaz. 39: 391. 1905.
On Fabaceae:

Parosela domingensis (DC.) Heller {Dalea domingensis
DC), Vento (Prov. Habana), Aug. 26, 1905, Baker 3312.
This short-cycle rust, having pycnia and telia, is known only
on the one species of host, and only from Cuba and Mexico.

26. Tranzschelia punctata (Pers.) Arth. Result. Sci. Congr.

Bot. Vienne 340. 1906.
Aecidium punctatum Pers. Ann. Bot. Usteri 20: 135. 1796.
Puccinia Pruni-spinosae Pers. Syn. Fung. 226. 1801.

120 Semi-centennial of Torrey Botanical Club

On Amygdalaceae :

Amygdalus Persica L. (Prunus Persica Sieb. & Zucc),
Santiago de las Vegas, Sept. 27, 1915, II, Johnston 166.
The collection shows only uredinia, and these not in much
abundance. The species is heteroecious, having its aecia on
Anemone, Hepatica, and Thalictrum. It is a common rust through-
out tropical and temperate regions of both hemispheres, but only
three collections have been seen from the West Indies, one each
from Cuba, Porto Rico, and Bermuda, and all being uredinia and
on same host.

27. Prospodium appendiculatum (Wint.) Arth. Jour. Myc. 13:

31. 1907.

Puccinia appendiculata Wint. Flora 67: 262. 1884.
On Bignoniaceae :

Stenolohium stans (L.) D. Don {Tecoma stans Juss.),
Holguin (Prov. Oriente), March 17, 1903, Holway,
Havana, March 22, 1903, Holway (Barth. N. Am. Ured.
724); Santiago de las Vegas, Feb. 28, 1916, Johnston 46Q,
The species is common in tropical regions, and is autoecious,
possessing pycnia, uredinia, and telia. The only other West
Indian island represented is Porto Rico.

28. Prospodium plagiopus (Mont.) Arth. N. Am. Flora 7: 162.

1912.

Puccinia plagiopus Mont. PI. Cell. Cuba 294. 1842.
On Bignoniaceae:

Tecoma lepidota (H. B. K.) DC, San Marcos (Prov. Santa
Clara), without date, II, iii {Ramon de la Sagra?); same
locality, Sept. 26, 1910, II, F. S. Earle; Palm barrens
east of Guanabacoa (Prov. Habana), April 2, 1910, II, iii,
Britton, Earle &' Wilson 6264 (host no. 6257).
Tecoma pentaphylla (L.) Juss., Santo Domingo (Prov.
Santa Clara), Nov. 7, 191 5, II, Johnston 18 j; Havana,
Feb. 26, 1916, Johnston 4Q1; San Juan, Isle of Pines,
March 15, 17, 1916. II, Britton, Britton & Wilson 15471.
The species is an autoecious one with uredinia and telia, pre-
ceded undoubtedly by pycnia, which have not yet been seen. It
was described originally by C. Montague, who prepared the volume
of cryptogams which was part of the encyclopedic work on the

Arthur and Johnston: Uredinales of Cuba 121

physical, political, and natural history of Cuba, issued irregularly
in parts about 1840-65, both from Paris and Havana. The
volumes 9-12, treating of natural history, were edited and in part
written by Ramon de la Sagra, volume nine being Montagne's
treatment of the cryptogams, the material for which is generally
understood to have been collected by de la Sagra. An excellent
description of this species was given by Montague under the name
Puccinia plagiopus, together with figures of both kinds of spores,
and also of the coriaceous leaflet and peltate hair of the host.
The reason for figuring the unusual form of hairs was not to help
in the identification of the host, which was said to be wholly un-
known (mihi prorsus ignota), but because the writer thought he had
discovered a remarkable metamorphosis, by which the peltate hair
became the envelope surrounding each group of spores. The
circle of paraphyses bounding each sorus forms a unique fimbri-
ated structure supported on a pedicel-like base, and easily de-
tachable as a whole, having about the size and color of the plant-
hairs, so that the error of interpretation was a natural one.

No further information regarding this rust became available
until the original collection was restudied by Monsieur P. Hariot
of the Paris Museum (Bull. Soc. Myc. Fr. 7: 196. 1891) half a
century later. After an extended discussion he expresses the
opinion that it doubtless constitutes the basis of a new genus
related to Puccinia, Phragmidium, and Uropyxis, having uredinio-
spores surrounded by a membrane, teliospores with a Uropyxis-
like, gelatinous covering, and "no paraphyses." However, with
only one specimen in the herbarium he hesitates to establish the
genus {il serait temeraire de se prononcer). The paraphysate
structure, which Montague thought might be a transformed hair
of the host, is described by Hariot as a hyaline halo in form of a
collar with wavy edge, but he does not associate the term para-
physes with it, hence his nascent genus is said to be without para-
physes.

Hariot attempted to advance the identity of the host by sug-
gesting that the nature and form of the hypophyllous hairs indi-
cated a member of the Oleaceae, a statement followed in Sydow's
Monographia Uredinearum (i: 345. 1902).

When working on the second part of the rust portion of the

122 Semi-centennial of Torrey Botanical Club

North American Flora the senior author observed the pecuHarity
of appendaged pedicels described by Montagne, and thought the
species might be a member of the newly erected genus Prospodium.
An appeal was made to M. Hariot in March, 1909, for a fragment
of the original Cuban collection, which was most kin-dly sent, and
which confirmed the suggestion. Some months previous to this
the Pflanzenfamilien of Engler & Prantl had been searched for
some illustration which would give the kind of leaf or leaflet
figured by Montagne, without much success, although similar
hairs were illustrated under Bignoniaceae. In January, 1909, the
herbarium of the N. Y. Bot. Garden was searched for a similar
purpose, and again with no better success. Going from New
York to Washington, the National Museum yielded two sheets,
possibly two collections, from Cuba by Charles Wright, made in
1865, locality not given, labelled Tecoma lepidota DC, which
showed compound leaves with leaflets that quite well answered to
the requirements in Montague's record. With this aid the assis-
tance of Prof. F. S. Earle, a sometime student of the rusts, was
sought. Prof. Earle was at that time stationed at Jovellanos,
about 80 or 100 miles west of San Marcos, the type locality. A
statement of the situation, with a drawing of the leaf as given by
Montagne, was sent in February, 1909. A hasty visit to the
locality made by Prof. Earle in April, 1909, yielded no results, but
in September, 1910, he transmitted two leaflets from "a Catalpa-
like shrub with 5-parted leaves," and the statement that "at last
I am able to send you what I take to be your long desired rust
from the type locality; I chanced to stop with friends at San
Marcos today [Sept. 26] and found one shrub with a few infected
leaves." This collection gave uredinia only, but there was little
doubt about its being the Montagne species. The host was still
unnamed. In only a few weeks, however, an ample collection
with telia and uredinia was received from Dr. N. L. Britton, col-
lected in another part of the island, which supplied details for
both host and rust, and the long search for definite knowledge of
this curious rust and its obscure host was successfully ended. The
result shows that the rust is curious^ as stated by Montagne, that
it is worthy of being placed in a separate genus related to Puccinia^
Phragmidium, and Uropyxis, as believed by M. Hariot. Finally,

Arthur and Johnston: Uredinales of Cuba 123

another species of host for it was discovered by the junior author
(1912), within the same Hmited region, and later by Whetzel and
Olive (191 6) in Porto Rico.

29. Prospodium Amphilophii (Diet. & Holw.) Arth. Jour. Myc.

13:31. 1907.

Puccinia Amphilophii Diet. & Holw. Bot. Gaz. 24: 30. 1897.
On Bignoniaceae:

Pithecoctenium echinatum (Aubl.) K. Schum., Caleta Coco-
drilos, Isle of Pines, March 8, 1916, 11, III, Britton,
Wilson & Leon 15274.
This long-cycle rust has heretofore been known only from
Mexico.

30. Prospodium Lippiae (Speg.) Arth. N. Am. Flora 7: 161.

1912.

Puccinia Lippiae Speg. Anal. Mus. Nac. Buenos Aires 6: 224.
1898.
On Verbenaceae:

Lippia dulcis Trev., Cacocum (Prov. Oriente), April 6,
1916, II, Johnston 550.
The species has not before been reported from the West
Indies. It is known from Mexico, Central America, and South
America.

The urediniospores are somewhat smaller than those of other
collections examined, with pores less evident, but other characters
of spores and sori, even to the inconspicuous paraphyses, are the
same.

31. Prospodium tuberculatum (Speg.) Arth. N. Am. Flora 7:
161. 1912.

Puccinia tuberculata Speg. Anal. Soc. Ci. Argent. 10: 6. 1880.
On Verbenaceae:

Lantana involucrata L., Santiago de las Vegas, June 4, 1916,
II, iii, Johnston 770.
This long-cycle rust has heretofore been known only from
Mexico.

32. Nephlyctis transformans (Ellis & Ev.) Arth. Jour. Myc.

13:32. 1907.
Puccinia transformans Ellis & Ev. Erythea 5: 6. 1897.

124 Semi-centennial of Torrey Botanical Club

Puccinia exitiosa Syd. & Holw. ; Sydow, Monog. Ured. i : 245.
1902.

On Bignoniaceae :

Stenolohium starts (L.) D. Don {Tecoma stans Juss.),
Havana, March 24, 1903, Holway; Santiago de las Vegas,
May 3, 1905, Home, Feb. 28, 1916, II, III, Johnston
46Q; Cojimar (Prov. Habana), May, 1905, Baker 260S;
Villa Real near Guanabacoa (Prov. Habana), May 20,
1912, Bro. Leon 2Q4j; Cienfuegos (Prov. Santa Clara),
Nov. 3, 1 91 5, II, Johnston 21 g; San Antonio de los Banos
(Prov. Habana), June 11, 191 6, Johnston '/81.
The species is a short-cycle rust possessing pycnia and telia, as
proven by cultures (Arthur, Jour. Myc. 12: 22. 1906; 13: 198.
1907), made in 1905 and 1906. The teliospores for the cultures
came from the material sent by Mr. Horne, as listed above. The
plants were from seed received from the Bahama islands and sown
in the greenhouse the year previous. The greenhouse plants thus
infected continued to produce the rust from the same galls for a
number of years, under greenhouse conditions. Material fixed
from the cultures was utilized by Dr. E. W. Olive in his cytological
studies (Sexual cell fusions and vegetative nuclear divisions in the
rusts. Ann. Bot. 22: 331-360. pi. 22. 1908). The same author
has further studied this material and reported it under the title
"The nuclear conditions in certain short-cycled rusts," but only
an abstract of the paper (Science II. 33: 194. 1911) has yet been
printed.

The species is known from various parts of Mexico, from New
Providence in the Bahama Islands, and from Cuba. Puccinia
elegans Schrot. of Argentina is a similar short-cycle rust, but with
larger and more coarsely sculptured spores, and has not yet been
found in North America. Type material for all the names cited
has been examined by the senior author. The names and de-
scriptions of the two species involved are unfortunately confused
in Sydow's Monographia Uredinearum i: 244-245.

33. Sphaerophragmium Dalbergiae Diet. Hedwigia 32: 30.
1893.

Uredo Sissoo Syd. & Butl. Ann. Myc. 4: 442. 1906.

Arthur and Johnston: Uredinales of Cuba 125

On Fabaceae:

Dalhergia Amerimnum Benth., Baracoa (Prov. Oriente),
April 14, 1916, II, Johnston 6jQ.
The collection shows an abundance of uredinia, but no telia.
The characters agree well with the description of the species as
given in Sydow, Monog. Ured. 3: 186, but no original specimen
has been seen. Although heretofore recognized only from Natal
in southern Africa, yet upon comparing the Cuban collection with
the one distributed in Sydow, Uredineen 21^0, which was collected
by E. J. Butler at Pusa, India, and issued as Uredo Sissoo, the
agreement seems perfect, and the wide tropical distribution be-
comes apparent. The description by Dietel in Hedwigia states
that the paraphyses are septate, having a cross wall at the middle.
That appears to be the case both in the West and East Indian
material, but a more critical examination shows that the appear-
ance is due to the arrangement of the cell contents, and not to a
cellulose wall. The spores are inclined to be bent to one side,
almost kidneyform at times. The two pores of the urediniospores
are easily seen.

The Uredo Dalhergiae P. Henn., on some species of Dalhergia
from Brazil, has much smaller, paler and thinner-walled spores.
The gross appearance of the sorus is also quite distinct, being
smaller and more nearly round.

The reference of this species to the genus Sphaerophragmium
is based upon the statement by Dietel that he found a single sorus
with some half dozen teliospores. The character of the life cycle
is problematical.

34. Uromyces leptodermus Sydow; Sydow & Butler, Ann. Myc.
4:430. 1906.

On Poaceae:

Panicum harhinode Trin., Santiago de Cuba (Prov. Oriente),
March 6, 1903, II, Holway; Santiago de las Vegas, March
I, 1907, II, III, Baker (Barth. Fungi Columb. 26/1),
Jan. 29, 1916, II, Johnston 42^; San Pedro, Isle of Pines,
Feb. I2-March 22, 1 916, II, Britton & Wilson 14715, Brit-
ton, Britton &• Wilson 15357; Herradura (Prov. Pinar
del Rio), March, 1917, II, Home.
The alternate form and host for this heteroecious species are

126 Semi-centennial of Torrey Botanical Club

yet unknown. The rust occurs on a number of species of Panicum
and Lasiacis in tropical regions from southern Florida and Mexico
into South America, and in the West Indies on the islands of
Porto Rico and Jamaica as well as Cuba. I J occurs also in India.

35. Uromyces Eragrostidis Tracy, Jour. Myc. 7: 281. 1893.
On Poaceae:

Eragrostis tephrosanthos Schult., Santiago de las Vegas,
June 5, 1916, II, Johnston 755.
The species has not before been collected in Cuba, and in the
West Indies has been reported only from Bayamon and Rio
Piedras in Porto Rico. It is not uncommon in the southern
United States, Mexico, and Central America on various species
of Eragrostis. The aecial form and its host are not known.

36. Uromyces ignobilis (Sydow) Arth. Mycologia 7: 181. 191 5.
Uredo ignobilis Sydow, Ann. Myc. 4: 444. 1906.

Uromyces major Arth. Bull. Torrey Club 38: 377. 191 1.
On Poaceae:

Sporoholus indicus (L.) R. Br., Santiago de las Vegas, June
5, 1916, II, Johnston 754.

This is the second collection for the West Indies, the other
being from Porto Rico, although it is probably a common tropical
rust. It is known also from Mexico and India. Aecia are not
known, and telia are not common.

Through the kindness of Mrs. Agnes Chase, material has been
examined, taken from the phanerogamic collection of the U. S.
Department of Agriculture, as follows, all on S. indicus: El Guana,
March 24, 1900, II, 404; Consolacion del Sur, April 3, 1900, II,
III, 47j; San Diego de los Bafios, April 26, 1900, II, 627, all from
Prov. Pinar del Rio, and collected by Palmer &' Riley; also Isle
of Pines, Jan. 31, 1904, \l, A. H. Curtiss 323.

37. Uromyces Rhyncosporae Ellis, Jour. Myc. 7: 274. 1893.
On Cyperaceae:

Rynchospora distans (Michx.) Vahl.
The rust was detected on this host in the phanerogamic her-
barium at the N. Y. Bot. Garden, on a specimen collected at
Pinar del Rio, no date but probably about 1857, Charles Wright
3399. This heteroecious rust, whose aecia are unknown, occurs

Arthur and Johnston : Uredinales of Cuba 127

on many species of Rynchospora throughout the eastern United
States and the West Indies, the record for the islands being Ja-
maica, Porto Rico, Martinique, Bermuda, and the Bahamas.

38. Uromyces Scleriae p. Henn. Hedwigia Beibl. 38: 67. 1899.
On Cyperaceae:

Scleria lithosperma (L.) Sw.
This heteroecigus species was detected in the phanerogamic
herbarium of the N. Y. Bot. Garden, on this host, collected six
miles from northwestern end Cayo Coco (Prov. Camagiiey), Oct.
23-24, 1909, J. A. Shafer 2y2Q, showing both uredinia and telia.
Previously it has been known in North America only from Porto
Rico. The aecial form is yet to be discovered.

39. Uromyces Commelinae (Speg.) Cooke, Trans. Roy. Soc.

Edinb. 31: 342. 1888.
On Commelinaceae :

Commelina longicaulis Jacq., Cienfuegos (Prov. Santa
Clara), Nov. 5, 191 5, Johnston 2oy.
This species is imperfectly known. Only uredinia have been
found in the western hemisphere, but telia have been collected on
the African coast of the Red Sea, the adjacent island of Socotra,
and the Malabar coast of India. It is assumed, without direct
evidence, that the species also includes pycnia and aecia in its
life cycle, but whether it is to be considered as autoecious or heter-
oecious is highly problematical. In America it is recorded also
from Jamiaca, Porto Rico, St. Croix, and South America.

The host is usually given as C. nudifiora L., a name, as reported
at the New York Botanical Garden, belonging to a very different
plant, not a member of the true genus Commelina.

40. Uromyces Celosiae Diet. & Holw. Bot. Gaz. 31: 326. 1901.
On Amaranthaceae :

Iresine Celosia L. (I. paniculaia Poir.), Santiago de las
Vegas, Feb. 28, 1916, II, Johnston 468.
This is the first record for the West Indies, although known for
Mexico and Guatemala. The species is imperfectly understood.
Beside uredinia and telia there is a possibility that p^xnia and
aecia appear to complete the life cycle, and there is some likelihood
of heteroecism.

128 Semi-centennial of Torrey Botanical Club

41. Uromyces jamaicensis Vesterg. Ark. Bot. Stockh. 4^^: 33.

1905.
On Fabaceae:

Bauhinia divaricata L., Caleta Grande, Isle of Pines, March
9, 1 91 6, Britton, Wilson & Leon 15335.
An autoecious species, apparently short-cycled although no
pycnia have yet been detected. It occurs in Mexico on the same
host, and in Porto Rico and Jamaica on other hosts.

42. Uromyces Medicaginis Pass.; Thiim. Herb. Myc. Oecon.
156. 1874.

Uromyces Medicaginis-falcatae Wint. in Rab. Krypt.-Fl. i: 159.
1881.
On Fabaceae:

Medicago sativa L., Santiago de las Vegas, May 13, 1916,
II, Johnston yoo.
The alfalfa rust has not before been reported from Cuba, and
only one other West Indian station is known for it, Bermuda.
When it does occur the rust appears to attack its host with much
virulence, causing the leaves to roll up and shrivel.

The species is considered to be heteroecious, with aecia on
upright-growing euphorbias of the genus Tithymalus, but no aecia
have been found in America, and no cultures have been made with
American material.

43. Uromyces Neurocarpi Dietel, Hedwigia 34: 292. 1895.
Uromyces insularis Arth. Bull. Torrey Club 33: 515. 1906.

On Fabaceae:

Clitoria rubiginosa Juss., Baracoa (Prov. Oriente), March,
1903, II, III, Underwood & Earle 1386, April 14, 191 6, II,
Johnston 615; Rio de la Casas, Isle of Pines, March 20,
1916, II, III, Britton &' Wilson J5(55J / Sabanilla (Prov.
Oriente), April 22, 191 6, II, Johnston 631.
This species was detected also in the phanerogamic herbarium
of the N. Y. Bot. Garden on the same host from Tabajo at base of
El Yunque, Dec. 12, 1910, /. A. Shafer 7719. The species has
been taken also in Porto Rico and Jamaica, but outside of the
West Indies only in Vera Cruz, Mexico, and Brazil. The species
is doubtless autoecious, requiring pycnia and aecia to complete
the life cycle, although they have not yet been discovered.

Arthur and Johnston: Uredinales of Cuba 129

44. Uromyces appendiculatus (Pers.) Fries, Summa Veg.

Scand. 514. 1849.
On Fabaceae:

Phaseolus vulgaris L., Havana, March 21, 1903, II, iii,
Holway; Candelaria (Prov. Pinar del Rio), Feb. 17, 191 7,
II, Johnston qj8.
Vigna vexillata (L.) A. Rich., Toa, Baracoa (Prov. Oriente),
April 17, 191 6, II, Johnston 568.
The species has been detected also on phanerogamic speci"
mens in the herbarium of the N. Y. Bot. Garden, on Dolichos
Lahlah L. {Lablab vulgaris Savi) from Regla (Prov. Habana),
April 8, 1903, II, /. A. Shafer g8; on Vigna repens (L.) Kuntze,
Punta de Afuera, Bahia Honda (Prov. Pinar del Rio), Dec. 14,
1910, II, Percy Wilson g26i; Vigna vexillata (L.) A. Rich., Campo
Florido (Prov. Habana), July 18, 1912, II, Bro. Leon 3352.

The collectipn on Vigna vexillata, made by Mr. Johnston, shows
urediniospores with two or three pores, varying from equatorial to
markedly superequatorial, a condition also noted in the Porto
Rican rusts on this host genus (Mycologia 7: 185. 191 5).

The species is autoecious, but pycnia and aecia have not been
reported from the tropics, and even telia are somewhat rare. The
rust is cosmopolitan on Dolichos, Phaseolus, Stropho styles, Vigna,
and probably other genera.

45. Uromyces Dolicholi Arth. Bull. Torrey Club 33: 27. 1906.
On Fabaceae:

Cajan Cajan (L.) Millsp. {Cajanus indicus Spreng.), San-
tiago de las Vegas, Jan. 30, 1916, II, Johnston 461.
Although no pycnia or aecia have been found, yet the affinities
ot this rust lead one to believe that it is autoecious. It ranges from
Texas to Colombia and probably to Argentina, the collections
usually showing only uredinia. Various notes regarding the
species are given in the Uredinales of Porto Rico (Mycologia 7:
186-188. 1 91 5). Other West Indian islands represented are
Porto Rico and St. Domingo.

46. Uromyces Hedysari-paniculati (Schw.) Farl.; Ellis, N. Am.
Fungi 246. 1879.

Uromyces, solidus Berk. & Curt. Grevillea 3: 57. 1874.
Uredo Desmodii-tortuosi P. Henn. Hedwigia 35: 252. 1896.

130 Semi-centennial of Torrey Botanical Club

On Fabaceae:

Meihomia Scorpiurus (Sw.) Kuntze {Desmodium Scorpiurus
Desv.), Soledad, Cienfuegos (Prov. Santa Clara), Nov.
5, 1 91 5, II, Johnston IQ4; Matanzas (Prov. Matanzas),
Feb. 7, 9, 1916, II, Britton, Britton & Wilson 14087; Rio
de las Casas, Isle of Pines, March 20, 191 6, II, Britton &'
Wilson 15652; Santiago de las Vegas, Jan. 30, 1916, II,
Johnston 460; Omaja (Prov. Oriente), March 31, 1916,
II, Johnston 5qq.

Meihomia tortuosum (Sw.) Kuntze {Desmodium tortuosum
DC), Santiago de las Vegas, Feb. 26, 1916, II, Johnston
149.

The rust is probably not rare throughout the island in the
uredinial stage. It has been detected in the phanerogamic her-
barium of the N. Y. Bot. Garden on Meihomia Scorpiurus, Rio
San Miguel (Prov. Pinar del Rio), Dec. 17, 1910, Percy Wilson
pj8o, and on M. tortuosum near Vento (Prov. Habana), Aug. 13,
1907, Baker, Tracy & Hasselbring joyg, and near Herradura (Prov.
Pinar del Rio), Aug. 28-31, 1910, Britton & Earle 65QI.

Through the kindness of Dr. Lindau of the Berlin Museum,
the senior author has recently been enabled to examine the spores
from the collection made by Sintenis in Porto Rico, named by
Hennings Uredo Desmodii-tortuosi, and finds that they are finely
echinulate, and not truly smooth as described by Hennings, and
in every way agree with those of the common form.

The species is only known elsewhere in the West Indies from
Porto Rico, but has a continental range both north and south.
It is autoecious, but pycnia and aecia are rarely seen, the latter
being pale and inconspicuous.

47. Uromyces Janiphae (Wint.) Arth. Mycologia 7: 190. 1915.
On Euphorbiaceae :

Manihot Manihot (L.) Cockerell {M. utilissima Pohl,
Jatropha Manihot L.), Santiago de las Vegas, Feb. 4,
1 91 6, II, Johnston; same, Oct. 13, 191 6, Johnston Q03.
The material collected by Mr. Johnston affected the fruit with
a strong development of uredinia; all other material seen has been
on the leaves or stems of the plant.

The species is autoecious, with all spore forms, and is known

Arthur and Johnston: Uredinales of Cuba 131

from Porto Rico, Mexico, and South America. Aecia and telia
have been found only in Mexico.

48. Uromyces proeminens (DC.) Pass. Rab. Fungi Eur. 1795.
1873.

Uromyces Euphorbiae Cooke & Peck; Peck, Ann. Rep. N. Y.
State Mus. 25: 90. 1873.
On Euphorbiaceae :

Chamaesyce hirta (L.) Millsp. {Euphorbia hirta L., E.
pilulifera L.), Soledad, Cienfuegos (Prov. Santa Clara),
Nov. 4, 1 915, II, Johnston 22j; Herradura (Prov. Pinar
del Rio), March, 191 7, I, II, Home.
Chamaesyce hypericifolia (L.) Small {Euphorbia hyperici-
folia L.), Madruga (Prov. Habana), Aug. 14, 1916, II,
III, Johnston 8yi.
Poinsettia heterophylla (L.) Kl. & Garcke {Euphorbia hetero-
phylla L.), Santiago de las Vegas, May 19, 1905, II,
Home, Feb. 26, 1916, II, Johnston 487; Saetia (Prov.
Oriente), April 8, 191 6, II, III, Johnston ^46.
This autoecious species possessing pycnia, aecia, uredinia, and
telia is probably rather common throughout the West Indies, as
well as northward and southward. It has been reported from
Jamaica, Porto Rico, St. Croix, and the Bahamas. The aecia,
which are not uncommon, usually occupy the under surface of all
the leaves of erect and somewhat drawn and etiolated shoots.
The uredinia of Charles Wright's collection no. 720, reported in the
Cuban Fungi under the name Trichobasis euphorbiaecola B. & C,
appear to be on Chamaesyce hirta, judging from the hairs and
serration of the fragment of leaf seen from the Herb. Curtis,
kindly transmitted by Dr. W. G. Farlow.

49. Uromyces Cupaniae (Speg.) nom. nov.

Uredo cristata Speg. Anal. Soc. Ci. Argent. 17: 119. 1884.

On Sapindaceae:

Cupania macrophylla A. Rich., Santiago de las Vegas, June
6, 1905, Baker 88, Dec. 3, 191 6, Johnston 927; San An-
tonio de las Banos (Prov. Habana), June 11, 1916,
Johnston yyg; Taco Taco (Prov. Pinar del Rio), Sept. 17,
1 91 6, Johnston 8y6.

132 Semi-centennial of Torrey Botanical Club

Pycnia amphigenous, thickly scattered among the telia over
dark-colored, hypertrophied and bullate areas 2-8 mm. across,
noticeable, subepidermal, dark brown, globoid or flattened-
globose in section, 80-130 ju in diameter by 65-125/1 in height;
ostiolar filaments compact, 30-40 ju long.

Telia amphigenous, scattered over dark-colored, bullate areas,
2-8 mm. across, strongly punctiform, main part of the scfrus very
deep-seated within the mesophyll, opening by a pore; paraphyses
and peridium none; teliospores obovate or fusiform, 16-29 by
40-58 />t, narrowed below, rounded or somewhat narrowed above;
wall colorless or nearly so, the inner portion firm, 1-2 /x thick, the
outer portion hygroscopic, swelling to 5-15 /a thick, strongly tuber-
culate along prominent ridges or wings, especially toward the
summit; pedicel completely fugacious.

This rust is most unusual in gross appearance. The swollen
areas are prominent, both from the chocolate-brown color and
from being well raised above the leaf surface. The teliospores are
ejected from the narrow openings of the sori in colorless masses.
In vertical section the sori are found to be flask-shaped, with two
or more layers of host cells above them, and having the spores
developed from a hymenium at the base. The crested appearance
of the teliospores is highly distinctive.

The specimens collected by Baker in 1905 were on old and
bleached leaves, and had little appearance of a rust. A packet in
the Arthur herbarium had been labelled Gymnosporangium guara-
niticum, a synonym for a Hyphomycetous fungus now called
Patouillardiella guaranitica. The identity of the fungus was not
ascertained, although much study was given to it, until the excel-
lent material from the junior author was available. The charac-
ters of the rust were then easily obtained, and soon identified with
those of Uredo cristata Speg., a species founded on an unidentified
species of Sapindaceae, collected by B. Balansa in Paraguay,
January, 1882, no. 3474. A new specific name is now given, as
the one applied by Spegazzini is already in use in the genus Uromy-
ces.

In his comments Spegazzini calls it "species pulcherrima dis-
tinctissima," which it truly is. It is a short-cycled rust, so very
distinctive that it is impossible to state its systematic position.
Although described as a Uromyces, yet it has affinities with both
the Uropyxidatae and Phragmidiatae. There is considerable

Arthur and Johnston: Uredinales of Cuba 133

similarity to the telia of Skierka, and it may represent a correlated
short-cycle genus. Until the spores are germinated, it can not be
known whether the pores are apical or lateral, a character having
considerable importance.

50. Uromyces Howei Peck, Ann. Rep. N. Y. State Mus. 30: 75.

1879.

On Asclepiadaceae :

Asclepias curassavica L., Santiago de las Vegas, Oct. 24,
1915, II, Johnston 132; San Pedro, Isle of Pines, Feb. 12-
March 22, 191 6, II, III, Britton & Wilson 14809; Baracoa
(Prov. Oriente), April 15, 191 6, II, Johnston ^ig.
Asclepias nivea L., San Antonio de los Banos (Prov. Haba-
na), June 11, 191 6, II, Johnston jyS.
The species is common in temperate regions northward, but
rare in the tropics. The only other West Indian island where it
has been taken is Porto Rico.

The life cycle is uncertain. Only uredinia and telia are known,
and whether the rust possesses both pycnia and aecia, and whether
it is autoecious or heteroecious, are yet wholly open questions.

51. Uromyces gemmatus Berk. & Curt.; Berkeley, Jour. Linn.

Soc. 10: 357. 1869.
On Convolvulaceae :

Jacquemontia nodiflora (Desv.) G. Don {Convolvulus nodi-
florus Desv.), Prov. Oriente, 1856, II, Charles Wright.
The species is autoecious, having pycnia, uredinia, and telia,
but no aecia. The collection made by Charles Wright, here listed,
is in the Curtis collection at Harvard University. It is labelled
" Uredo gemmata Berk. & Curt, var.," and possesses only uredinia.
The type of the species in the Kew herbarium, Wright's no. 278,
has not been seen. The species was not again collected until
found by F. L. Stevens in a number of localities in Porto Rico,
1913. Specimens, now first reported, were secured by E. W. D.
Holway in Porto Rico in 191 1, and in Jamaica in 191 5. The only
other known locality is St. Croix. A full description of the species
with notes is given in Mycologia 7: 192-193. 1915.

134 Semi-centennial of Torrey Botanical Club

52. Uromyces dolichosporus Diet. & Holw. Bot. Gaz. 31: 327.

1901.

On Boraginaceae :

Tournefortia voluhilis L., Punta Sabanilla, Cienfuegos Bay
(Prov. Santa Clara), Feb. 24, 1910, II, Britton, Earle &
Wilson 4575.

The species is autoecious, having large subepidermal pycnia,
chestnut-brown uredinia, and colorless telia. The fungus dis-
torts and etiolates young and growing organs. It is known else-
where only from Oaxaca, Mexico.

53. Uromyces Hellerianus Arth. Bull. Torrey Club 31: 2.

1904.

On Cucurbitaceae :

Cayaponia racemosa (Sw.) Cogn., Soledad, Cienfuegos

(Prov. Santa Clara), Nov. 6, 1915, Johnston ig2.
Melothria guadalupensis (Spreng.) Cogn., Riverside (Prov.
Camagiiey), Nov. 30, 191 5, Johnston 28^.
Uredinia and a few telia were detected also on a phanerogamic
specimen in the herbarium of the N. Y. Bot. Garden, on Melothria
guadalupensis, from Buenaventura (Prov. Pinar del Rio), Dec. 13,
1 910, Percy Wilson 9237. The species occurs in Porto Rico and
Guatemala. It is considered autoecious and to have pycnia and
aecia, which are yet undetected, however.

54. Uromyces bidenticola (P. Henn.) Arth. Mycologia 9: 71.

1917.

Klehahnia Bidentis Arth. Mycologia 7: 196. 1915.
On Carduaceae:

Bidens leucantha Willd., Santiago de las Vegas, March 11,
1 91 6, Johnston 729; Baracoa (Prov. Oriente), April 17,
191 6, II, III, Johnston 569; Holguin (Prov. Oriente),
April 4, 1 91 6, II, Johnston S77', Paso Estancia (Prov.
Oriente), May 3, 191 6, II, Johnston 702; San Antonio de
los Banos (Prov. Habana), June 11, 1916, II, III, Johnston
765-

Bidens pilosa L., Havana, Feb. 5, 191 6, II, Britton, Britton
&" Wilson 14138; Caleta Cocodrilos, Isle of Pines, March
8, 1 91 6, II, Britton, Britton &' Wilson 15268.

Arthur and Johnston: Uredinales of Cuba 135

This autoecious species, having pycnia, uredinia, and teHa, is
now first reported from Cuba, but has been known from Porto
Rico, Jamaica, and Martinique, as well as from the continents of
North and South America. It has usually been listed under U.
Bidentis Lagerh., a name belonging to a similar short-cycle species
on the same hosts, not yet reported for Cuba.

55. Uromyces columbianus Mayor, Mem. Soc. Neuch. Sci. Nat.

5:467. 1913.
Nigredo columbiana Arth. Mycologia 7: 194. 1915.
On Carduaceae:

Melanthera brevifolia O. E. Schultz, Santiago de las Vegas,
July 26, 1915, II, III, Johnston 14J.
Uredinia have been found on phanerogamic specimens in the
herbarium of the N. Y. Bot. Garden, on M. brevifolia, Vedado
(Prov. Habana), May, 1909, Bro. Leon 1371, and on M. hastata
cubensis O. E. Schultz, near Sta. Fe (Prov. Habana), Oct. 7, 1915,
Bro. Leon.

This autoecious rust has heretofore been known only from
Porto Rico, and from the type locality in Colombia, S. A. The
life cycle embraces pycnia, aecia, uredinia, and telia.

56. PucciNiA PURPUREA Cooke, Grevillea 5: 15. 1876.
On Poaceae:

Holcus halepensis L. {Sorghum halepense Pers., Andropogon
halepensis Brot.), Santa Clara (Prov. Santa Clara),
March 19, 1903, II, Holway (Barth. Fungi Columb. 4670);
Havana, March 21, 1903, II, Holway (Barth. N. Am.
Ured. Q63) and March 24, 1903, Holway; Aguacate
(Prov. Habana), March 23, 1903, Holway; Santiago de las
Vegas, July 19, 1904, Home, and Sept. 27, 1915, Johnston
169; Herradura (Prov. Pinar del Rio), March, 1917, II,
HI, Home.

Holcus Sorghum L. {Sorghum vulgare Pers.), Santiago de las
Vegas, Nov. 17, 1905 (Broom, Kaffir, and Milo), Horne^
March i, 1907, Baker 1716, 1717 (Barth. N. Am. Ured.
1072, Fungi Columb. 2464, Sydow Ured. 2328), March
10, 1916 (Egyptian millet), Johnston 730.
The species is undoubtedly heteroecious, but the aecial host is
yet undiscovered. A very common rust in warm regions, usually

136 Semi CENTENNIAL of Torrey Botanical Club

producing both uredinia and telia. The other West Indian islands
represented are Jamaica, Porto Rico, and Bermuda.

57. PucciNiA Cenchri Diet. & Holw. Bot. Gaz. 24: 28. 1897.

On Poaceae:

Cenchrus echinatus L., Santiago de Cuba (Prov. Oriente),
March 7, 1903, II, Holway; Itabo (Prov. Matanzas),
Nov. 12, 1915, II, Johnston 181; Baracoa (Prov. Oriente),
April 18, 1916, II, Johnston 664; Santiago de las Vegas,
June 5, 1916, II, III, Johnston 764.

Cenchrus viridis Spr., Santiago de las Vegas, Sept. 27, 1915,
II, Johnston 165; Saetia (Prov. Oriente), April 8, 1916, ii,
Johnston 5J7.

A heteroecious species whose alternate host is not known. It
occurs in the southern United States, Mexico, Porto Rico, and
the Bahamas.

58. PucciNiA DEFORMATA Berk. & Curt. Jour. Linn. Soc. 10: 357.

1869.

On Poaceae:

Olyra latifolia L., without locality, January, 1857, II, III,
Charles Wright sg6; Soledad, Cienfuegos (Prov. Santa
Clara), Nov. 7, 191 5, II, Johnston 247; Ceballos (Prov.
Camagiiey), March 24, 1916, II, Johnston sog.
The type collection was made by Charles Wright in eastern
Cuba, January, 1857, locality not given. A number of collections
have been made in Porto Rico, and one in Nicaragua. The species
is considered autoecious, but no clue to the alternate host has yet
been obtained.

59. PucciNiA HuBERi P. Henn. Hedwigia Beibl. 39: 76. 1900.
On Poaceae:

Panicum fasciculatum Sw., Toa, Baracoa (Prov. Oriente),

April 17, 1916, II, Johnston 641.
Paspalum virgatum L., Ceballos (Prov. Camagiiey), Nov.
23, 1915, II, III, Johnston joy.
This imperfectly known, heteroecious species has been re-
ported from Porto Rico on Panicum fasciculatum and P. trichoides,
but not from other West Indian islands. It was first described
from Brazil.

Arthur and Johnston: Uredinales of Cuba 137

60. PucciNiA SUBSTRIATA Ellis & Barth. Erythea 5: 47. 1897.
Puccinia Chaetochloae Arth. Bull. Torrey Club 34: 585. 1907.

On Poaceae:

Chaetochloa geniculata (Lam.) Millsp. & Chase (C imberbis
Scribn., C. purpurascens S. & M.), Santiago de las Vegas,
Feb. 26, 1916, II, Johnston 483; same, June 5, 1916, II,
Johnston '^62; San Pedro, Isle of Pines, Feb. 12-Mar. 22,
1916, II, Britton &' Wilson 15439; Baracoa (Prov.
Oriente), April 17, 1916, II, Johnston 558.
Chaetochloa onurus (Willd.) S. & M., Ceballos (Prov. Cama-

giiey), Nov. 25, 1915, II, Johnston 301.
Chaetochloa verticillata (L.) Scribn., Botanic Garden,

Havana, March 21, 1903, Holway.
Syntherisma sanguinalis (L.) Dulac (Panicum sanguinale
L.), Santiago de las Vegas, Feb. 26, 1916, Johnston 482.
This heteroecious species, with the alternate host unknown, is a
common cosmopolitan rust occurring on many species of hosts,
especially in warmer regions. It ranges from Nebraska south-
ward into South America. Of the other West Indian islands it is
known from Jamaica, Porto Rico, and Bermuda.

In addition to the above localities, a number have been found in
connection with phanerogamic specimens in the herbarium of the
N. Y. Bot. Garden: on C. geniculata, Isle of Pines, May, 1910,
II, 0. £. Jennings 154; Los Palacios (Prov. Pinar del Rio), Jan. 15,
1912, II, /. A. Shafer 11795; on C. onurus (Willd.) S. & M., near
the mouth of Bueyvaca (Prov. Matanzas), Aug. 28, 1903, II,
Britton & Wilson 2g; near Santiago, Sept. 6, 1906, II, Norman
Taylor 232; Sierra Nipe (Prov. Oriente), Dec. 5, 1909, /. A. Shafer
3020; on C. setosa (Sw.) Scribn. {Setaria setosa Beauv.), El Yumuri
(Prov. Matanzas), April, 1849, Rugel 880.

61. Puccinia Anthephorae (Syd.) comb, no v.

Uredo Anthephorae Sydow, Ann. Myc. i: 22. 1903.
On Poaceae:

Anthephora hermaphrodita (L.) Kuntze {A. elegans Schreb.).
Through the kindness of Dr. H. Sydow, some of the original
collection of this species has been examined. It differs from the
uredinia of Puccinia Chaseana Arth., occurring on the same host,
in having thick-wajled urediniospores. The collection, as stated

138 Semi-centennial of Torre y Botanical Club

by the Sydows (I.e.), was made in Cuba by Ramon de la Sagra.
No date or locality is given.

Telia were found associated with uredinia on a phanerogamic
specimen of the same host, at the New York Botanical Garden,
collected between Portland Point and Rocky Point, Jamaica,
March 5, 1908, N. L. Britton 1917. A few teliospores were found
on another similar phanerogamic specimen in the same herbarium,
collected in St. Croix, Feb. 6, 1896, Alfred E. Ricksecker 253.
From these two collections the following characters for the telia
have been secured.

Telia amphigenous, scattered, linear or oblong, 0.2-2 mm. long,
early naked, opening by a longitudinal slit of the epidermis, com-
pact, blackish-brown, ruptured epidermis noticeable; teliospores
broadly ellipsoid, 23-26 by 31-35 rounded at both ends, only
slightly constricted at septum; wall chestnut-brown, 3-4 m thick,
somewhat thicker above, 5-7 ju, with a broad, low, concolorous
umbo, smooth; pedicel hyaline, 9^^ broad, not tapering, up to 85 /x
long, the wall thin, i ju.

The species is only known from the three collections cited, but
is doubtless widespread, although it may not be abundant,
throughout the West Indies, as the host is a wayside weed.

62. PucciNiA Rhamni (Pers.) Wettst. Verhl. Zool.-Bot. Ges.
Wien 35:545. 1886.

Puccinia coronata Corda, Icones i: 6. 1837.
On Poaceae:

Avena sativa L., Santiago de las Vegas, March 14, 1917, II,
III, Johnston pj/.
This is the only record of the oat rust for the West Indies.
The species is represented in Mexico by two collections on Bromus
from the federal district made by E. W. D. Holway, and is not
reported from Central America.

63. Puccinia poculiformis (Jacq.) Wettst. Verhl. Zool.-Bot.
Ges. Wien 35: 544. 1886.

Puccinia graminis Pers. Neues Mag. Bot. i : 119. 1794.
On Poaceae:

Triticum vidgare Vill. {T. sativum Lam.), without locality
or date {Ramon de la Sagra?).
The only record of the black stem-rust of wheat for any of the
West Indian islands is in the Flora Cubana (Montague, PI. Cell.

Arthur and Johnston: Uredinales of Cuba 139

Cuba 293. 1842), where the species is described in full and the
statement is made that it occurs on culms, leaves, and glumes, and
even awns, of wheat. The record is presumably based on a col-
lection by Ramon de la Sagra.

64. PucciNiA SoRGHi Schw. Trans. Am. Phil. Soc. 11. 4: 295.

1832.
On Poaceae:

Zea Mays L., Santiago de las Vegas, Aug. 26, 1904, Baker
126s, Aug. 4, 1905, Home, Oct. 1, 191 5, Johnston i/i;
Guanajay (Prov. Pinar del Rio), Jan. 27, 1915, II,
Johnston 142; Artemisa (Prov. Pinar del Rio), Jan. 29,
1915, II, III, Johnston 148.
The rust is heteroecious, with aecia on various species of
Oxalis. The rust is rarely collected in tropical regions, this being
the only record for the West Indies.

65. PucciNiA CANALICULATA (Schw.) Lagerh. Tromso Mus.
Aarsh. 17: 51. 1894,

Puccinia Cyperi Arth. Bot. Gaz. 16: 266. 1891.
On Cyperaceae:

Cyperus ferax L. C. Rich., El Yunque, Baracoa (Prov.
Oriente), March 10, 1903, II, III, Holway; Baracoa (Prov.
Oriente), April 15, 191 6, II, Johnston 632.
Cyperus sp., Herradura (Prov. Pinar del Rio), Jan. 191 7,
II, III, Home.

A cosmopolitan, heteroecious rust, common on many species of
Cyperus and Kyllinga. It has been proven by cultures to possess
aecia on Xanthium and Ambrosia in the northern United States,
but it doubtless has other aecial hosts. Of the other West Indian
islands it is known from Jamaica, Porto Rico, and Martinique.

The collection from Herradura has most of the urediniospores
with three pores, instead of the usual 2-pored condition with occa-
sionally three pores.

66. Puccinia Eleocharidis Arth. Bull. Iowa State Coll. 156.

1884.
On Cyperaceae:

Eleocharis geniculata (L.) R. Br., Baracoa (Prov. Oriente),
April 15, 1916, II, Johnston SOI.

140 Semi-centennial of Torrey Botanical Club

The uredinia were also found on E. capitata (L.) R. Br. {Scirpus
capitatus L.), collected by Charles Wright in Cuba, locality and
date not given, the specimen being in the phanerogamic herbarium
of Purdue University at Lafayette, Ind., having been received
from G. W. Clinton.

The species is heteroecious, it having been shown by culture
that the aecia occur on Eupatorium. In the tropics the uredinial
stage is the one usually collected, and it may be that the rust is
there propagated solely by the repeating spores.

Porto Rico is the only other West Indian island from which it
is reported.

67. PucciNiA scleriicola Arth. Mycologia 7: 232. 1915.
On Cyperaceae:

Scleria verticillata Muhl.
This heteroecious species, whose aecia are unknown, occurs on a
number of species of hosts from Florida and Georgia. The record
for Cuba is based upon a phanerogamic collection in the herbarium
of the N. Y. Bot. Garden, from Nuevo Gerona, Isle of Pines,
December, 1903, A. H. Curtiss.

68. PucciNiA FuiRENAE Cooke, Grevillea 6: 137. 1878.
On Cyperaceae:

Fuirena simplex Vahl.
This heteroecious species, whose aecia are unknown, was de-
tected in its uredinial stage on a phanerogamic collection in the
National Herbarium and communicated by Mrs. Agnes Chase.
It was collected near Nueva Gerona, Isle of Pines, Dec. 17, 1903,
by ^. H. Curtiss 2jy. The species has been known from South
Carolina, Florida, Alabama, and Texas, and is now first reported
for the West Indies.

69. PucciNiA SciRPi DC. Fl. Fr. 2: 223. 1805.
Aecidium Nymphoidis DC. Fl. Fr. 2: 597. 1805.

On Menyanthaceae:

Limnanthemum Grayanum Griseb., Pinar del Rio, Decem-
ber, 1858, O, I, Charles Wright gzg.
This heteroecious species is represented in America by only
two collections. The uredinia were detected in 1902 on Scirpus
lacustris L. from Guanica, Porto Rico, A. A. Heller 62Q1. The

Arthur and Johnston: Uredinales of Cuba 141

aecia, as listed above, have been examined in a collection in the
herbarium of Dr. W. G. Farlow at Cambridge, Mass., and are
characteristic of the species in both gross and microscopic char-
acters. They are accompanied by pycnia. The packet is marked
1857-8, but the year is undoubtedly 1858, as Wright was in Cuba
December, 1858, but not there in December, 1857 (cf. Underwood,
Wright's explorations in Cuba, Bull. Torrey Club 32: 293. 1905).
This is the only collection on this host genus yet made in America.

70. PucciNiA Smilacis Schw. Nat. Ges. Leipzig i : 72. 1822.
On Smilaceae:

Smilax havanensis Jacq., San Antonio de los Bafios (Prov.
Habana), June 11, 191 6, II, Johnston ySj.
A seemingly rare rust in the West Indies, having been col-
lected only once before, in Porto Rico.

71. Puccinia Cannae (Wint.) P. Henn. Hedwigia 41 : 105. 1902.
On Cannaceae:

Canna indica L., Santiago de las Vegas, July 2, 1906, Baker
(Barth. Fungi Columb. 2387 j Sydow Ured. 21 14), July

10, 1906, Cook; Los Indios, Isle of Pines, Feb. 13, 1916,

11, Britton, Britton & Wilson 15350; Baracoa (Prov.
Oriente), April 18, 191 6, II, Johnston 662.

Canna sp., Soledad, Cienfuegos (Prov. Santa Clara), Nov.
4, 1915, Johnston 220.
Only uredinia and telia are known for the species, and the
character of the remaining part of the life cycle can not be pre-
dicted. It is also reported from Jamaica and Porto Rico.

72. Puccinia Polygoni-amphibii Pers. Syn. Fung. 227. 1801.
On Polygonaceae :

Persicaria punctata (Ell.) Small {Polygonum punctatum
Ell., P. acre H.B.K.), Paso Estancia (Prov. Oriente),
May 2, 1916, II, Johnston 517; San Antonio de los Banos
(Prov. Habana), June 11, 1916, II, Johnston 777.
These collections show an abundance of uredinia, but no telia,
being the usual condition on this host. The first record for the
West Indies was from Haiti in 1910, and it was lately taken in
Porto Rico. It is a common rust in temperate regions both north
and south, as well as in the Old World. Cultures have been con-

142 Semi-centennial of Torrey Botanical Club

ducted both in America and Europe, showing the aecia to occur on
species of Geranium.

73. Puccinia striolata (Speg.) comb. nov.

Uredo striolata Speg. An. Soc. Ci. Arg. 9: 173. 1880.
Puccinia macropoda Speg. An. Soc. Ci. Arg. 10: 8. 1880.
On Amaranthaceae :

Iresine angustifolia Euphr. (/. elatior Rich.), Cacocum

(Prov. Oriente), April 6, 1916, II, Johnston ^34.
Iresine Celosia L. (/. paniculata Kuntze), Aguacate (Prov.
Habana), March 23, 1903, Holway.
This rust is probably autoecious, but the life cycle is not fully
known. No pycnia or aecia have yet been found. Only one col-
lection made by Spegazzini in April, 1880, at Chacarita, Argentina,
on Iresine Celosia L. (/. celosioides L.), is so far known bearing
telia. All other collections bear only uredinia.

The species has also been reported from the islands of St.
Thomas and Porto Rico, and from Ecuador and Argentina.

74. Puccinia Rivinae (Berk. & Curt.) Speg. An. Mus. Buenos
Aires 19: 304. 1909.

Aecidium Rivinae Berk. & Curt. Jour. Lin. Soc. 10: 358. 1869.
Endophyllum Rivinae Arth. N. Am. Flora 7: 126. 1907.
Puccinia Raunkiaerii Ferd. & Winge, Bot. Tiddskr. 29 : 8. 1908.
On Petiveriaceae (Phytolaccaceae) :

Rivina humilis L., Managua near Havana, June 11, 1906,
Cook; Sabanilla (Prov. Oriente), April 22, 191 6, Johnston

516.

Rivina octandra L., San Antonio de los Baflos (Prov. Haba-
na), April 5, 1905, Baker & Van Hermann 4775; Cacocum
(Prov. Oriente), April 6, 191 6, Johnston 515; Santiago de
las Vegas, Feb. 27, 1916, Johnston 477, May 28, 1916,
I, Johnston 760, June 4, 1916, II, Johnston 771; Paso
Estancia (Prov. Oriente), May 3, 1916, I, Johnston 68^,
70s; Ceballos (Prov. Camagiiey), March 24, 191 6, I,
Johnston S 11; Antilla (Prov. Oriente), April 8, 1916, II,
III, Johnston 512.
An autoecious species with all spore forms, occurring also in
Porto Rico and St. Thomas. The rust greatly distorts the young
shoots, as shown in the cut (p. 97) . Germination of the aeciospores

Arthur and Johnston: Uredinales of Cuba 143

was first undertaken by the junior author in March, 1916, by sowing
them on the surface of hardened beef agar in a Petri dish. Spores
from the small groups of aecia on the leaves germinated readily
over night, producing long hyphal tubes, characteristic of aecio-
spores. The spores from aecia on the greatly hypertrophied
shoots, however, could not be made to germinate. Repeated
attempts again in 191 7 resulted in the same way. Specimens were
sent to Lafayette, Ind., which gave no better results. A collection
on hypertrophied shoots, made at Santiago de las Vegas on June
14, 1917, and received in Lafayette one week later and quite dry,
was sown on the surface of water. The next day, June 22, long
unseptated and unbranched germ tubes appeared sparingly. It is
proven, therefore, that the hypertrophied form, long known as En-
dophyllum Rivinae, is an aecial form belonging to Puccinia Rivinae.

75. Puccinia Zorniae (Diet.) McAlpine, Rusts of Australia 172.
1906.

Uredo Zorniae Diet. Hedwigia 38 : 257. 1899.
On Fabaceae:

Zornia diphylla (L.) Pers., Herradura (Prov. Pinar del
Rio), Sept. 30, 1904, II, III, Baker 2143.
This imperfectly known rust, usually collected in the uredinial
form, occurs also in Florida and Mississippi, as well as in Africa
and Australia. This is the first record for the West Indies*
Heretofore teliospores have been reported from Australia only^
but both of the collections here listed show telia interspersed with
the uredinia. The telia are small and inconspicuous, and were
first found when sectioning to ascertain if pycnia might be present.
Some of the sori contained only teliospores, no urediniospores
being intermixed. The spores are somewhat more slender than
those described by McAlpine, and seemingly paler. They appear
to be capable of germination upon maturity.

The rust was detected also by Mr. Percy Wilson in the phan-
erogamic herbarium of the N. Y. Bot. Garden, on the same host^
collected at Pinar del Rio, April 23, 1903, /. A. Shafer 2g2.

76. Puccinia inflata Arth. Bull. Torrey Club 33: 516. 1906.
On Malpighiaceae :

Stigmaphyllon periplocifolium (Desf.) Juss., Baracoa (Prov.
Oriente), March 13, 1903, II, lll,Holway (Barth. N. Am.

144 Semi-centennial of Torrey Botanical Club

Ured. 42); near Cayamas (Prov. Santa Clara), Oct. 13,
1904, II, Baker 3538.
Stigmaphyllon Sagraeanum A. Juss. {S. reticulatum A. Juss.),
Rio San Juan (Prov. Santa Clara), March 24-25, 1910,
III, Britton, Earle & Wilson 5906; palm barren east of
Guanabacoa (Prov. Habana), April 2, 191 o, II, III,
Britton, Earle &' Wilson 626Q (host no. 6261; Itabo
(Prov. Matanzas), Nov. 12, 1915, II, Johnston lyg;
Cienfuegos (Prov. Santa Clara), Nov. 3, 191 5, II, John-
ston I go; Ceballos (Prov. Camagiiey), March 23, 191 6,
II, III, Johnston 510; Las Tunas (Prov. Oriente), March
29, 191 6, Johnston 543; Santiago de Cuba (Prov. Oriente),
April 30, 1 91 6, Johnston 686.
The species is autoecious, possessing pycnia, uredinia, and
telia. It occurs also in Porto Rico on S. lingulatum, from which
the type was obtained, but has not before been reported elsewhere.

77. Puccinia barbatula sp. nov.
On Malpighiaceae :

Banisteria laurifolia L., Paso Estancia (Prov. Oriente),
May 3, 1916, O, II, III, Johnston 6/8.

Pycnia amphigenous, in small groups on brownish areas 1-3
mm. across, dark brown, noticeable, subepidermal, globoid, 70-
125 [jL in diameter.

Uredinia hypophyllous, few, circinating about the pycnia,
oval or oblong, 0.2-0.8 mm. long, originating deep within the tis-
sues and remaining partly covered by them, dehiscent by slit or
pore, somewhat pulverulent, dark cinnamon-brown; uredinio-
spores broadly ellipsoid or obovoid, 25-32 by 39-45 m; wall dark cin-
namon-brown, 2-2.5 M thick, strongly and very sparsely echinulate,
the echinulations colorless, 2-2.5/1 long, the pores 2, equatorial.

Telia hypophyllous, crowded about the uredinia, oval or
oblong, 0.3-1 mm. long, long covered by the overarching tissues,
whitish or pale brown, inconspicuous; teliospores oblong or clavate-
oblong, 18-26 by 35-48 /z, rounded at both ends or slightly nar-
rowed below, somewhat constricted at septum; wall colorless,
uniformly 1.5-2.5 fx, smooth; pedicel colorless, very broad, 13-15 m»
half length of spore.

78. Puccinia Arechavelatae Speg. An. Soc. Ci. Arg. 12: 67.

1881.

On Sapindaceae:

Cardiospermum microcarpum H.B.K.

Arthur and Johnston: Uredinales of Cuba 145

This short-cycle rust, common in tropical America, was de-
tected on a phanerogamic collection in the herbarium of the N. Y.
Bot. Garden, obtained along the railroad near Cerro (Pro v.
Habana), April 13, 1903, /. A. Shafer 183.

Other West Indian stations are in Jamaica, Porto Rico, Anti-
gua, and the Bahamas.

79. PucciNiA GouANiAE Holw. Ann. Myc. 3: 21. 1905.
On Frangulaceae (Rhamnaceae) :

Gouania lupuloides (L.) Urban {G. domingensis L.), Ceballos

(Prov. Camagiiey), Nov. 25, 1915, II, Johnston 305 .
Gouania polygama (Jacq.) Urban {G. tomentosa Jacq.),
without locality (Prov. Oriente), December, 1856, II,
Charles Wright 282; Gibara (Prov. Oriente), March 15,
1903, II, III, Holway (Barth. N. Am. Ured. 544)] San
Diego de los Banos (Prov. Pinar del Rio), Feb. 7, 1915,
II, Johnston lyd; Matanzas (Prov. Matanzas), Feb. 17,
1916, II, III, Britton, Britton & Wilson 13999; Los Indios,
Isle of Pines, Feb. 13, II, Britton, Britton & Wilson
14242; San Pedro, Isle of Pines, Feb. 12-March 22, 1916,
II, Britton, Britton & Wilson 15804; Santiago de las
Vegas, Feb. 29, 19i6,'ll, III, Johnston 4^0; Baracoa (Prov.
Oriente), April 15, 191 6, II, Johnston 594, 618; Ceiba
Mocha (Prov. Matanzas), July 25, 1916, II, Johnston 868.
This autoecious species, having pycnia, uredinia, and telia, is
discussed in the Stevens' list of Porto Rican Uredinales (Mycolo-
gia 7: 237-238. 1915). The collection by Wright, no. 282, is the
second one of the two numbers cited under the original description
of Uromyces gemmatus B. & C. (Jour. Linn. Soc. 10: 356. 1869),
a species said to be on " the underside of leaves of Convolvulus, &c."
The first number there cited (278) is on a Convolvulaceous host
(see 51), and the second number must, therefore, represent the
"&c." Wright's collection was ample, and the host was deter-
mined from material in the Curtis herbarium by Percy Wilson of
the New York Garden, Jan. 13, 191 5, as Gouania polygama. The
authorities of the Kew herbarium kindly sent a fragment of 282
to the senior author, which agrees perfectly with that in the
Curtis herbarium.

The species occurs on a phanerogamic specimen of G. poly-

146 Semi-centennial of Torrey Botanical Club

gdma in the herbarium of the N. Y. Bot. Garden, collected at
Herradura, March 17, 1907, II, F. S. Earle 606 (not ''806" as
erroneously printed in the Porto Rican list, /. c, p. 237). The
species occurs also in Porto Rico and in Panama.

80. Puccinia invaginata nom. nov.

JJredo Gouaniae Ellis & Kelsey, Bull. Torrey Club 24: 209.
1897.

On Frangulaceae (Rhamnaceae) :

Gouania lupuloides (L.) Urban (G. domingensis L.), Caleta
Cocodrilos, Isle of Pines, March 8, 191 6, II, III, Britton,
Wilson & Leon 15275.

Uredinia hypophyllous, scattered, early naked, pulverulent,
cinnamon-brown; paraphyses peripheral, terete to clavate, some-
what incurved, often from a branching base 9-16 by 29-45 /x, the
wall colorless, thin, smooth; urediniospores introverted from the
side, and appearing (with pore in optical section) arcuate to obo-
void-arcuate, 16-19 by 26-31 or (with pore in surface view)
obovoid, 19-23 by 26-31 /x; wall cinnamon-brown, 1.5-2 thick,
sparingly to moderately echinulate, only one pore, equatorial, on
indented or flattened side.

Telia amphigenous, scattered, early naked, pulverulent, dark
chocolate to blackish-brown, ruptured epidermis inconspicuous;
teliospores broadly ellipsoid or broadly obovoid, 26-29 by 29-37 M>
rounded at both ends, slightly or not constricted at septum; wall
dark chestnut-brown, uniformly 3-4 11 thick, moderately verrucose;
pedicel colorless, 19-50 /x long, fragile.

This is the first collection of the species for Cuba and the first
collection showing telia from any locality. The first uredinial
collection was made on the island of St. Croix by Ricksecker in
1896 on the same host, and numerous collections have been made in
Porto Rico on this host and on G. polygama.

81. Puccinia heterospora Berk. & Curt. Jour. Linn. Soc. 10:

356. 1869.
On Malvaceae:

Anoda hastata Cav., Santiago de las Vegas, July 13, 1904,
Earle 235, July 15, 1904, Home, April 25, 1906, Baker
(Barth. Fungi Columb. 2358) ; Soledad Cienfuegos (Prov.
Santa Clara), Nov. 5, 191 5, Johnston IQ7; Ceiba Mocha
(Prov. Matanzas), July 25, 1916, Johnston 867.

Abutilon hirtum (Lam.) Sweet, Herradura (Prov. Pinar del

Arthur and Johnston: Uredinales of Cuba

147

Rio), Sept. 28, 1906, Van Hermann 2ggi (host no. 2gQ6)
(Barth. Fungi Columb. 2453) ; Guanajay (Prov. Pinar
del Rio), Sept. 13, 1904, Earle &' Wilson 1508; Vedado
(Prov. Habana), Dec. 6, 191 6, Johnston Q2^.
Ahutilon* indicum Sweet, Botanic Garden, Havana, March

21, 1903, Holway (Barth. N. Am. Ured. 240).
Ahutilon permolle (Willd.) Sweet, Gibara (Prov. Oriente),

March 15, 1903, Holway (Barth. N. Am. Ured. jg).
Gaya occidentalis (L.) Sweet, Santiago de las Vegas, Sept.
19, 1915, Johnston 162, June 25, 1916, Johnston 851;
Tetas de Camarioca (Prov. Matanzas), Feb. 7, 9, 191 6,
Britton, Britton & Wilson 14088.
Sida glutinosa Cav., Santiago de las Vegas, Nov. 20, 1905,

Van Hermann 3373.
Sida spinosa L. {S. angustifolia Lam.), Santiago de las

Vegas, June 25, 1916, Johnston 854.
Wissadula periplocifolia (L.) Presl, Manacas (Prov. Santa
Clara), Nov. 11, 191 5, Johnston 185; Camagiiey (Prov.
Camagiiey), Nov. 28, 1915, Johnston 366.
A short-cycle species without pycnia, very common throughout
the warmer regions of the earth, on many malvaceous genera.

It has been seen on the following collections from Cuba in the
phanerogamic herbarium of the N. Y. Bot. Garden: on Ahutilon
ahutiloides (Jacq.) Garcke {A. lignosum Rich.), Santiago de Cuba,
1899, V. Havard 78, March 10-25, 1912, Britton, Britton &
Cowell 12SQ1; on Ahutilon hirtum (Lam.) Sweet, Cienfuegos, Aug.
I3> 1895, Rohert Comhs 462; Valley of the San Juan (Prov. Pinar
del Rio), March 19, 1903, Britton, Britton & Shafer 277; on Anoda
hastata Cav., Matanzas, March 16, 1903, Britton, Britton & Shafer;
near San Luis (Prov. Oriente), Feb. 15, 17, 18, 1902, Pollard &'
Palmer 2Q3; on Sida procumhens Sw., Rio Almendares to Playa de
Marianao (Prov. Habana), Dec. 22-23, i9io> Percy Wilson
9506.

Other West Indian islands known for the species are Jamaica,
Porto Rico, St. Thomas, St. Croix, and the Bahamas, but it prob-
ably occurs on many others.

148 Semi-centennial of Torrey Botanical Club

82. PucciNiA MALVACEARUM Mont. in Gay, Hist. Chile 8: 43.

1852.
On Malvaceae:

Malvastrum coromandelianum (L.) Garcke, Santiago de las
Vegas, May 10, 1906, Cook.
This widely distributed short-cycle rust, which does not pro-
duce pycnia, is here first reported for the West Indies. It was
collected on the same host in Venezuela, July 15, 1913, F. L.
Stevens 2861, 2q8j. It was found in Jamaica at Mandeville, on
M. corchorifolium (Desv.) Britton, Oct., 1892, T. D. A. Cockerell
44, and again at the same place, Feb. 23, 1915, E. W, D. Holway
228.

83. Puccinia LuDWiGiAE (E. & E.) Holw. N. Am. Ured. i: 72.

1907.

On Onagraceae:

Isnardia repens (Sw.) DC, Jucaro, Isle of Pines, Feb. 20,

1 91 6, I, Britton, Britton & Wilson 14624.

This is the first record for this long-cycle species outside of the
United States. The host is also a new one.

84. Puccinia Psidii Wint. Hedwigia 23: 177. 1884.
On Myrtaceae:

Jamhos Jamhos (L.) Lyons {Eugenia Jamhos L., Jamhosa
vulgaris DC), El Yunque, Baracoa (Prov. Oriente),
March 10, 1903, II, III, Underwood Earle 1381; same,
II, III, Holway; Baracoa (Prov. Oriente), April 14, 1916,
Johnston ^06; Candelaria (Prov. Pinar del Rio), Jan.

1917, II, Home.

The synonymy and various notes are given in the Stevens' list
of Uredinales of Porto Rico (Mycologia 7: 239-240. 1915).
The full life history of the rust is unknown. Uredinia are com-
mon and often accompanied by telia. Quite likely only pycnia
are needed to complete the stages.

85. Puccinia Hydrocotyles (Link) Cooke, Grevillea 9: 14.

1880.

On Ammiaceae (Umbelliferae) :
Hydrocotyle australis Coult. & Rose.
This imperfectly known rust is doubtless autoecious. Aecia

Arthur and Johnston: Uredinales of Cuba 149

have sometimes been referred to it, but they should more likely
be associated with Uromyces Scirpi (Cast.) Burr. There are a
number of reasons for thinking that the present species possesses
only pycnia, uredinia, and telia.

The record for Cuba is based upon a phanerogamic collection
in the herbarium of the N. Y. Bot. Garden from Guanabaco (Prov.
Habana), April 2, 1910, II, Britton, Earle &' Wilson 624Q. No
other West Indian station was known, until the 1916 collections
by Whetzel and Olive in Porto Rico, although the rust is common
along the eastern coast of both North and South America, notably
in Central America.
86. Puccinia Johnstonii Arthur sp. nov.
On Sapotaceae:

Dipholis salicifolia (L.) A. DC, San Diego de los Bafios

(Prov. Pinar del Rio), Feb. 7, 1915, II, III, Johnston 177

(type).

Sideroxylon foetidissimum L., Santiago de las Vegas, Feb.
27, 1916, II, III, Johnston 48Q.

Uredinia hypophyllous, scattered, pustular, small, 0.2-0.3
mm. across, subepidermal, tardily naked, at first opening by a
pore, becoming pulverulent, cinnamon-brown, the overarching
epidermis quite persistent; paraphyses peripheral in a single row,
erect, arising from a short membranous base, 2 or 3 cells deep, terete,
colorless, 10-13 M broad by 37-77 m long, the inner wall thin, about
I jjL, the outer wall thicker, 3-7 )u ; urediniospores ellipsoid or broadly
obovate, 23-29 by 37-48,0 ; wall cinnamon-brown, 1.5-2 thick,
thicker above, 5-9 /jl, with a lighter umbo, sparsely and strongly
echinulate, the pores 4, sometimes 3, equatorial.

Teliospores in uredinial sori ellipsoid or oblong, 23-30 by 33-
40 ju, rounded at both ends, not constricted at septum; wall dark
chestnut-brown, uniformly 3 jjl thick, sparsely and coarsely echinu-
late, the points 1-2 /z long; pedicel colorless, rough, 10 by 16-23 /jl,
the upper partswelHng in water up to 16 jjl, often attached some-
what obliquely.

An interesting species, showing some resemblances to Pro-
spodium in the character of wall and sculpturing of the teliospores,.
Truly echinulate teliospores are a novelty among Uredinales.
The circle of short and erect paraphyses, having a tissue-like base
of somewhat elongated cells, is clearly a transitional structure be-
tween a simple ring of paraphyses and a membranous peridium
having ostiolar cells larger than the others.

150 Semi-centennial of Torrey Botanical Club

In naming this distinctive species the senior author takes it
upon himself to use the name of Mr. J. R. Johnston in recognition
of the important service he has rendered mycology in making
known the rust flora of Cuba. It was at his suggestion that this
summary of present knowledge was undertaken, the first attempt
made to give a full list of Cuban rusts. And it has been through
his untiring efforts in securing material and making observations
that this considerable showing is possible. More than one third
of the number (40 species) is based entirely upon material supplied
by Mr. Johnston, and nearly two thirds of the total number of
collections are by him. In addition to this, and even more im-
portant, he has made many observations which have added to our
knowledge of life histories of different species.

87. PucciNiA CONCRESCENS Ellis & Ev. ; Arthur, Mycologia 7:
240. 1915.

Puccinia compacta Kunze; Bubak, Hedwigia Beibl. 42 : 30. 1905.
Not Berk. 1855, de Bary 1858, or Thiim. 1875.
On Asclepiadaceae :
Asclepias curassavica L.
This rather common rust of tropical America has not yet been
secured in Cuba by a mycological collector, but occurs on a speci-
men in the phanerogamic herbarium of the N. Y. Bot. Garden,
collected at Baracoa, Jan. 24-29, 1902, Pollard, Palmer &' Palmer
II.

The species is short-cycled and a lepto-form. The history of
the name is reviewed in the Stevens list of Uredinales of Porto
Rico (Mycologia 7: 240-242. 191 5), where a description is given.

88. Puccinia Gonolobi Rav. ; Berk. Grevillea 3: 54. 1874.
On Asclepiadaceae:

Philibertella claiisa (Jacq.) Vail, Rio Gavelan (Prov. Santa
Clara), March 26, 1910, Britton, Earle & Wilson 6022;
Baracoa (Prov. Oriente), April 14, 1916, Johnston 55J;
Toa (Prov. Oriente), April 18, 191 6, Johnston S52.
The asclepiadaceous forms of rust intended to be covered by
this name include those with dark sori and ellipsoid teliospores
having the septum generally transverse. The species is not well
defined. It is a short-cycled lepto-form, without pycnia.

Arthur and Johnston: Uredinales of Cuba

151

89. Puccinia obliqua Berk. & Curt. Jour. Linn. Soc. 10: 356.
1858.

Puccinia Cynanchi Lagerh. Bol. Soc. Brot. 7: 129. 1889.
Puccinia sphaerospora Syd. & Henn. Ann. Myc. i: 327. 1903.
On Asclepiadaceae :

Fischeria crispiflora (Sw.) Schl., near Los Indios, Isle of
Pines, May 20, 1910, 0. E. Jennings 43Q (host no. 438)]
Santa Barbara, Isle of Pines, Feb. 23, 1916, Bntton,
Britton & Wilson 14784; Santa Fe, Isle of Pines, Feb.
27, March i, 1916, Britton, Britton & Wilson 15 108;
Saetia (Prov. Oriente), April 9, 1916, Johnston 551.
[Metastelma penicillatum Griseb.?] without locality (Prov.
Oriente), 1857, Charles Wright.
The species is autoecious, and similar in habit and gross appear-
ance to Puccinia Gonolobi Rav., but as here used is intended to
include those forms having mostly globoid teliospores, with moder-
ately thick walls, and oblique septa in most instances. The type
of the species was collected in Cuba by Charles Wright on some
undetermined host, locality not given. Upon request the ma-
terial at the Kew herbarium was examined, and the opinion given
that the host appears to be a species of Metastelma, probably M.
penicillatum Griseb. There are, however, at least two collections
at Kew. The type cited in the Fungi Cubenses (page 356) is no.
281. This may be the one which is thought to be on Metastelma
penicillatum, and is listed aboye. Another collection bears the
number 288, but is without year or locality other than "Cuba."
A leaf from this collection, kindly sent to the senior author from
Kew, well answers to the statement with the original description
as a leaf "of some plant resembling chickweed."

90. Puccinia crassipes Berk. & Curt. Grevillea 3: 54. 1874.
On Convolvulaceae :

Ipomoea cathartica Poir. (/. acuminata R. & S., not Ruiz &
Pav.), Santiago de las Vegas, Aug. 20, 1904, I, Earle &'
Wilson 1 140 (Barth. Fungi Columb. 2456), Sept. 3, 1904,
I, Abarca 1361, Nov. 20, 1904, I, III, Earle 51 go.

Ipomoea triloba L., Santiago de las Vegas, Nov. 25, 1916,
I, Johnston Q23, Nov. 29, 191 7, III, Johnston q^2.
An autoecious species with aecia and telia, but having no

152 Semi-centennial of Torrey Botanical Club

uredinia. It flourishes especially in warm regions, and has been
reported from St. Croix and Porto Rico, and also from the ad-
joining continent of North America.

91. Puccinia megalospora (Orton) comb. nov.

Allodus megalospora Orton, Mem. N. Y. Bot. Card. 6: 198.
1916.

On Convolvulaceae :

Ipomoea Carolina L., Hanabanilla Falls, Trinidad Mts.
(Prov. Santa Clara), March 1-2, 1910, I, Britton, Earle
& Wilson 4827; Santa Barbara, Isle of Pines, Feb. 12-
March 22, 1916, I, Britton, Britton & Wilson 14786.
This autoecious rust resembles Puccinia crassipes in the ab-
sence of uredinia, and also in gross appearance, but possesses much
larger aeciospores, and teliospores more thickened at the apex.
The Cuban stations are the only ones known for it outside of
Mexico.

92. Puccinia Lantanae Farl. Proc. Amer. Acad. Sci. 18: 83.

1883.

On Verbenaceae:

Lantana Camara L., Holguia (Prov. Oriente), March 17,

1903, Holway (Barth. N. Am. Ured. 64^).
Lantana involucrata L. (L. odorata L.), hills near Santiago

de las Vegas, May 14, 1903, Baker 286g; Playa Marianao

(Prov. Habana), Oct. 31, 191 5, Johnston 240; Vanadero

(Prov. Matanzas), Feb. 8, 191 6, Britton & Wilson 14049;

Saetia (Prov. Oriente), April 8, 1916, Johnston 557.
Lantana reticulata Pers., Marianao (Prov. Habana), Aug.

13, 1916, Johnston 870.
Lippia dulcis Trev., Paso Estancia (Prov. Oriente), March

3, 1916, Johnston 680, May 3, 1916, Johnston 703.
Lippia stoechadifolia H.B.K., San Luis (Prov. Oriente),

February, 1902, Pollard & Palmer 308; Santiago de Cuba

(Prov. Oriente), March 6, 1903, Holway (Barth. N. Am.

Ured. 749).

Priva lappulacea (L.) Pers., El Yunque, Baracoa (Prov.
Oriente), March 10, 1903, Holway.
This short-cycle species, common on many hosts in tropical
America, has been detected also on Lantana trifolia L. in the

Arthur and Johnston: Uredinales of Cuba

153

phanerogamic herbarium of the N. Y. Bot. Garden, Santiago de
Cuba (Prov. Oriente), April 26, 1902, 5. H. Hamilton 46, and in
the same herbarium on L. reticulata Pers., El Moro to Cojimar
(Prov. Habana), Dec. 8, 191 o, P. Wilson 9134; San Juan, Isle of
Pines, March 15, 17, 1916, Britton, Britton &' Wilson 14981.

It has also been taken in Jamaica, Porto Rico, St. Thomas and
the Bahamas.

93. PucciNiA Urbaniana P. Henn. Hedwigia 37: 278. 1898.
On Verbenaceae:

Valerianodes jamaicensis (L.) Medic. (Abena jamaicensis
Hitch., Stachytarpheta jamaicensis Vahl), Santiago de
Cuba (Prov. Oriente), March 6, 1903, Holway; Batabano
(Prov. Habana), March 20, 1906, Baker 2y6y; Conso-
lacion del Sur (Prov. Pinar del Rio), Feb. 2, 191 5, John-
ston 152; Santiago de las Vegas, Oct. 2, 191 5, Johnston
1^4; Baracoa (Prov. Oriente), April 14, 191 6, Johnston
606.

A short-cycle species occurring also in Porto Rico and the
Bahamas, as well as in southern Florida.

94. PucciNiA SALViicoLA Diet. & Holw. Bot. Gaz. 24: 33. 1897.
On Lamiaceae (Labiatae) :

Salvia occidentalis Sw., Marianao (Prov. Habana), Feb, 6,
1916, Johnston 442; San Pedro, Isle of Pines, Feb. 12-
March 22, 191 6, II, Britton & Wilson 142070; Ceballos
(Prov. Camagiiey), March 24, 191 6, II, Johnston 524;
Sabanilla (Prov. Oriente), April 22, 1916, II, Johnston 593.
Only uredinia were found, and these not abundant. The
species also occurs in Porto Rico and Jamaica. The full life cycle
is unknown, but probably pycnia and aecia are sometimes formed.

95. PucciNiA medellinensis Mayor, Mem. Soc. Neuch. Sci.

Nat. 5:497. 1913-

On Lamiaceae (Labiatae):

Mesosphaerum pectinatum (Poir.) Kuntze (Hyptis pectinata

Poir.), Cienfuegos (Prov. Santa Clara), Nov. 3, 1915,

Johnston ig6; Matanzas (Prov. Matanzas), Feb. 7, 1916,

II, Britton, Britton & Wilson 13995.
Mesosphaerum suaveolens (L.) Kuntze (Hyptis suavcolens

Poir.), Scintiago de Cuba (Prov. Oriente), March 6, 1903,

154 Semi-centennial of Torre y Botanical Club

II, Holway; Marianao (Prov. Habana), Oct. 31, 191 5,
Johnston 241; Cienfuegos (Prov. Santa Clara), Nov. 3,

1 91 5, II, Johnston 244; Tetas de Camarioca (Prov.
Matanzas), Feb. 7, 9, 191 6, II, Britton, Britton & Wilson
1408Q] San Juan, Isle of Pines, March 15, 17, 1916, II,
Britton, Britton & Wilson 15 461.

Apparently a common autoecious species throughout the West
Indian islands, but on M. suaveolens only uredinia have been found.
In the North American Flora (7: 212-213. 1912) this species is
confused with Eriosporangium tucumanense (Speg.) Arth., a South
American species not yet found in North America. The descrip-
tion following that name applies to P. medellinensis , but should
give the pores of the urediniospores as 2 or sometimes j.

96. PucciNiA Hyptidis (M. a. Curt.) Tracy & Earle, Bull. Miss.
Exp. Sta. 34: 86. 1895.

On Lamiaceae (Labiatae):

Mesosphaerum capitatum (L.) Kuntze {Hyptis capitata
Jacq.), Baracoa (Prov. Oriente), March 9, 1903, II,
Holway; Paso Estancia (Prov. Oriente), May 3, 191 6, II,
Johnston dg^.

This autoecious species has not yet been takea in the West
Indies in any but the uredinial stage, although known from the
southern United States on M. rugosum {Hyptis radiata) bearing
pycnia, aecia, and telia, as well as uredinia. The other West
Indian stations are in Jamaica and Porto Rico. The record of
Hyptis suaveolens, as a host under this species in the North Ameri-
can Flora (7: 212), is an error. It belongs under P. medellinensis,
as given above.

97. PucciNiA insititia Arth. Mycologia 7: 248. 1915.
On Lamiaceae (Labiatae):

Mesosphaerum lantanifolium (Poir.) Kuntze {Hyptis lan-
tanifolia Poir.), Santa Ana, Isle of Pines, March 20,

191 6, II, Britton & Wilson 15668.

This long-cycle species is founded on a collection from Brazil,
and has also been found in Porto Rico. Only uredinia and telia
are known.

Arthur and Johnston: Uredinales of Cuba 155

98. PucciNiA Leonotidis (P. Henn.) Arth. Mycologia 7: 245.

1915.

On Lamiaceae (Labiatae) :

Leonotis nepetaefolia (L.) Br., Botanic Garden, Havana,
March 21, 1903, II, Holway (Barth. N. Am. Ured. 781);
Vedado (Prov. Habana), Nov. 23, 191 6, II, Johnston Q22.
This autoecious species, occurring throughout the warmer
regions of the world, has been found in America only in the uredin-
ial stage. Aecia and telia have been seen from Africa. The
other West Indian islands represented are Jamaica, Porto Rico,
and the Bahamas.

99. PucciNiA GLOBOSIPES Pcck, Bull. Torrey Club 12: 34. 1885.
On Solanaceae:

Lycium carolinianum Walt., Rio Gavelan (Prov. Santa
Clara), March 26, 1910, II, Britton, Earle & Wilson 6028
(host no. 6027).
An autoecious species with pycnia, uredinia, and telia, now
first reported for the West Indies. It is elsewhere known from
Alabama, Kansas, and Utah to the Mexican boundary, and as far
west as southern California.

100. Puccinia Adenocalymnatis (P. Henn.) comb. nov.
Vredo Adenocalymnatis P. Henn. Hedwigia 35: 249. 1896.
Puccinia aequinoctialis Holw. Ann. Myc. 3: 22. 1905.

On Bignoniaceae :

Cydista aequinoctialis (L.) Miers {Bignonia aequinoctialis

L.), Baracoa (Prov. Oriente), March 13, 1903, II, III,

Holway (Barth. N. Am. Ured. 525).
An imperfectly understood species. Only three collections
are known, two being from the West Indies, and on the same host.
The collection here cited shows a few teliospores and a good de-
velopment of uredinia, and a collection from Porto Rico shows
uredinia only. The beginning stage of the life cycle has not yet
been discovered. A collection from Brazil on Adenocalymna,
E. Ule po2, shows uredinia that appear to be the same as those of
the West Indian form. The spores on the fragmentary part of the
specimen examined are a trifle smaller, but the measurements
given by Hennings are the sa'ne as those taken from the West
Indian material. As Adenocalymna is closely related to Cydista,

156 Semi-centennial of Torrey Botanical Club

there appears no good reason for doubting the identity of the two
rusts, and they are therefore here united under one name. These
rusts were inadvertently placed by the senior author (Mycol. 9:
83. 191 7) under Puccinia cuticulosa Arth. {Uredo cuticulosa
E. & E.), a Nicaraguan rust, recently found to be identical with
Prospodium appendiculatum (Wint.) Arth.

101. Puccinia Ruelliae (Berk. & Br.) Lagerh. Tromso Mus.

Aarsh. 17: 71. 1895.

On Acanthaceae:

Blechum Brownei (Sw.) Juss., Santiago de las Vegas, Feb.
10, 1916, II, III, Johnston 44^.
This species possesses all spore forms. It occurs also in Porto
Rico, Martinique, and in Central and South America on the same
host. Only uredinia have been seen on North American collec-
tions on this host until the present collection came to hand, which
gave a few telia. It is now evident that the rust, heretofore listed
as P. Blechi Lagerh,, is identical with that on Ruellia, and other
hosts, often listed under the name P. lateripes Berk. & Rav.

102. Puccinia lateritia Berk. & Curt. Jour. Acad. Sci. Phila.

2:281. 1853.

On Rubiaceae:

Borreria laevis (Lam.) Griseb., Baracoa (Prov. Oriente),

April 15, 1 91 6, Johnston ^03.
Hemidiodia ocimifolia (Willd.) K. Schum., El Yunque,
Baracoa (Prov. Oriente), March 10, 1903, Holway.
The species is a short-cycle form without pycnia. It is a
common rust of warm regions, and is known from other West
Indian islands on the genera Diodia, Ernodia, Mitracarpum, and
Spermacoce. It has been reported from Jamaica, Porto Rico, and
the Bahamas.

103. Puccinia Xanthii Schw. Schr. Nat. Ges. Leipzig i: 73.

1822.

On Ambrosiaceae :

Xanthium longirostre Wallr., Santiago de las Vegas, June,
1905, Home; July 21, 191 5, P. Cardin (Johnston 168),
Oct. 1916, Johnston J, 4, 5, ij, 75, 16, j/; Baracoa (Prov.
Oriente), April 15, 1916, Johnston ^95-

Arthur and Johnston: Uredinales of Cuba

157

Xanthium saccharatum Wallr., Santiago de las Vegas, Oct.

1916, Johnston 2, 6, g, 10, 12, 14.
Xanthium intermediate between X. longirostre and X.

saccharatum^ Santiago de las Vegas, Oct. 1916, Johnston

J, 8, II,

The species has been detected on the first host also on a phan-
erogamic specimen in the herbarium of the N. Y. Bot. Garden,
collected at Paso Estancia (Prov. Oriente), Aug. 30, 1909, /. A.
Shafer 16^2.

The species, which is short-cycled and without pycnia, is a com-
mon American rust, but rather rare in the West Indian islands,
the other stations being in Jamaica, Porto Rico, and Bermuda,
and in each instance on X. longirostre.

104. Puccinia fuscella sp. nov.
On Carduaceae:

Vernonia menthaefolia Less., El Yunque, Baracoa (Prov.
Oriente), March 10, 1903, II, III, Holway (type), April
18, 1916, II, III, Johnston 584.

Uredinia hypophyllous, scattered, roundish, punctiform, mi-
nute, 0.1-0.2 mm. across, early naked, pulverulent, light cinna-
mon-brown, ruptured epidermis noticeable; urediniospores broadly
ellipsoid or obovoid, 23-29 by 26-32 ^; wall light cinnamon-brown,
moderately thick, 2-3 /x, echinulate, the pores indistinct, probably
4-6, and scattered.

Telia hypophyllous, scattered, or crowded in groups of two or
three sori, irregularly roundish, small, 0.2-0.3 mm. across, rather
early naked, dark chestnut-brown, ruptured epidermis incon-
spicuous; paraphyses peripheral, hyphoid, colorless, thin-walled,
short and inconspicuous; teliospores oblong, 21-27 by 40-48 /z,
slightly constricted at septum, rounded or obtuse at both ends;
wall cinnamon-brown, 1.5-2 /i, thicker above, 5-9// including a
semihyaline umbo, minutely verrucose above, appearing smooth;
pedicel colorless, two thirds length of spore or less, fragile.

Closely related to Puccinia Vernoniae Schw., but differs in the
uredinial pore-arrangement and other minute characters. The
type collection was issued as no. 772 in Bartholomew's North
American Uredinales, under the name P. Vernoniae.

105. Puccinia evadens Harkn. Bull. Calif. Acad. Sci. i: 34.
1884.

Eriosporangium evadens Arth. Result. Sci. Congr. Bot. Vienne
343. 1906.

158 Semi-centennial of Torrey Botanical Club

On Carduaceae:

Baccharis sp., Santa Clara (Prov. Santa Clara), March 22,
1903, II, III, Holway.
The species is autoecious, pycnia, aecia, uredinia, and telia all
being known, and is found in southern California and Arizona to
Central Mexico. The collection here listed is the only one known
from the West Indies.

106. PucciNiA ABRUPTA Diet. & Holw. Hedwigia 37: 208. 1898.
On Carduaceae:

Viguiera helianthoides H.B.K., Santiago de las Vegas, Feb.

28, 1906, III, Baker 2682; Marianao (Prov. Habana),

Aug. 13, 1916, II, Johnston 86q.
The species is probably autoecious, although no pycnia or
aecia have yet been seen. The collection by Baker is unique in
producing a fusiform swelling of the stem full two inches long and
twice the normal diameter of the stem. The uredinia have been
rarely seen. It was detected in the phanerogamic herbarium of the
N. Y. Bot. Garden, on the same host from Sierra de Anafe (Prov.
Pinar del Rio), Dec. 1911, II, P. Wilson. The species is now first
reported for the West Indies.

107. PucciNiA Helianthi Schw. Schrift. Nat. Ges. Leipzig i:

68. 1822.
On Carduaceae:

Helianthus annuus L., Santiago de las Vegas, June 15, 1905,
II, Home.

The collection of this autoecious rust, the only one seen from
the West Indies, was obtained in the garden of the Cuban Experi-
ment Station. It shows an abundance of sori, but only uredinia.
The species is cosmopolitan, and produces pycnia, aecia, uredinia,
and telia in the life cycle.

108. PucciNiA CoNOCLiNii Seym. Bot. Gaz. 9: 191. 1884.
On Carduaceae:

Ageratum maritimum H.B.K., Caleta Cocodrilos, Isle of
Pines, March 8, 191 6, II, III, Britton, Wilson & Leon
15311-

Eupatonum villosum Sw., Gibara (Prov. Oriente), March
I5» I903» II> Holway; Santiago de las Vegas, April 5,

Arthur and Johnston: Uredinales of Cuba 159

1906, II, Home, Feb. 28, 191 6, II, S. C. Bruner {Johnston
476).

This autoecious rust occurs on many hosts in the warmer parts
of America. Its aecia are not known, although Aecidium roseum
Diet. & Holw. was for a time supposed to belong with it, and it has
often been listed as P. rosea. A collection from Cuba was found
in the phanerogamic herbarium of the N. Y. Bot. Garden, from
Cabanas Bay (Prov. Oriente), on E. villosum, March 17-20, 191 2,
II, Britton & Cowell 128 16. Porto Rico is the only other West
Indian island at present represented.

109. Puccinia Synedrellae p. Henn. Hedwigia 37: 277. 1898.
Puccinia solida Berk. & Curt. Jour. Linn. Soc. 10: 356. 1869.

Not P. solida Schw. 1832.
Puccinia Tridacis Arth. Bull. Torrey Club 33: 156. 1906.
Puccinia Eleuther anther ae Diet. Ann. Myc. 7: 354. 1909.
On Carduaceae:

Eleuther anther a ruderalis (Sw.) Sch. Bip., without locality
(Prov. Oriente), 1856-7, Charles Wright 276; Vivijagua,
Isle of Pines, Feb. 28, 29, 1916, Britton, Britton &f Wilson
15071, March 18, 20, 1916, Britton &' Wilson 15608;
Maravi, Baracoa (Prov. Oriente), April 18, 1916, Johns-
ton 604.

Emilia sonchifolia (L.) DC, Ceballos (Prov. Camagiiey),
Nov. 24, 1915, Johnston J04; Canet (Prov. Camagiiey),
Dec. I, 1 91 5, Johnston 280; Nuevitas (Prov. Camagiiey),
Dec. 3, 1 91 5, Johnston 383; San Pedro, Isle of Pines,
Feb. I2-March 22, 191 6, Britton &' Wilson 14465; Colum-
bia, Isle of Pines, March 19, 21, 1916, Britton, Britton &
Wilson 15793; Paso Estancia (Prov. Oriente), May 3,
1916, Johnston 704; Baracoa (Prov. Oriente), April 15,
1 91 6, Johnston 585, April 17, 191 6, Johnston 566, 628.

Neurolaena lobata (L.) R. Br., El Yunque, Baracoa (Prov.
Oriente), March 12, 1903, Holway; Jucaro, Isle of Pines,
Feb. 20, 1916, Britton, Britton & Wilson 14613.

Synedrella nodiflora (L.) Gaertn., Cojimar (Prov. Habana),
Aug. 24, 1910, Britton, Earle & Gager 6272 (host no.
6271)', Santiago de las Vegas, Oct. 2, 1915, Johnston 155;
Taco Taco (Prov. Pinar del Rio), Sept. 17, 19 16, Johnston
877-

160 Semi-centennial of Torrey Botanical Club

Tridax procumbens L., Punta Brava (Prov. Pinar del Rio),
Nov. 15, 1904, Baker & 0' Donovan 40JQ; Matanzas
(Prov. Matanzas), Feb. 7, 191 6, Britton, Britton &
Wilson 13997; Columbia, Isle of Pines, Feb. 20, 1916,
Britton y Britton & Wilson 14664.
This species, common in the West Indies, is a short-cycle lepto-
form, without pycnia. It has been detected on phanerogamic
specimens in the herbarium of the N. Y. Bot. Garden, on Eleuthe-
ranthera ruderalis, near Gerona, Isle of Pines, May 8, 1904, A. H.
Curtiss 488; Guantanamo Bay, March 17-30, 1909, N. L. Britton
224s; on Tridax procumbens, Santiago de Cuba, March, 1903,
Underwood & Earle 125; and on Neurolaena Idbata, from "Cuba
Orientale," 1856-7, Charles Wright 772.

The type of Puccinia solida B. & C. was collected by Charles
Wright in ''Cuba Orientale," 1856-7, and the host was first de-
termined in January, 1910, from the specimen in the Curtis her-
barium at Harvard University, by B. L. Robinson of the Gray
herbarium, who found it to be E. ruderalis.

Other West Indian islands now represented are Jamaica,
Porto Rico, St. Domingo, Guadeloupe, Martinique, Grenada,
Antigua, Barbados, and Tortola, being the most extensive West
Indian record for any species of rust up to the present time,
no. PucciNiosiRA PALLIDULA (Speg.) Lagerh. Tromso Mus.
Aarsh. 16: 122. 1894.
On Tiliaceae:

Triumfetta semitriloba L., Itabo (Prov. Matanzas), Nov.
12, 191 5, Johnston 178; Ceballos (Prov. Camagiiey),
Nov. 25, 1915, Johnston 303; Minas (Prov. Camaguey),
Dec. 2, 1915, Johnston 372; Santiago de las Vegas, Dec.
3, 191 6, Johnston 930.
A short-cycle species, not very conspicuous, and probably more
common than the few collections known would indicate. It also
occurs in Porto Rico, Jamaica, Guadeloupe, and in Guatemala and
South America.

III. Endophyllum circumscriptum (Schw.) Whetzel & Olive,
Am. Jour. Bot. 4: 49. 1917.
Aecidium circumscriptum Schw.; Berk. & Curt. Jour. Acad. Sci.
Phila. 2: 283. 1853.

Arthur and Johnston: Uredinales of Cuba

161

Aecidium Cissi Wint. Hedwigia 23: 168. 1884.
On Vitaceae:

Cissus sicyoides L., Baracoa (Prov. Oriente), April 14,
1916, Johnston 502.
This short-cycle rust has been considered until recently a stage
of a heteroecious species. It is known from Jamaica and Porto
Rico, and from Panama and the northern border of South America.

112. BoTRYORHiZA HiPPOCRATEAE Whctzcl & Olivc, Am. Jour.

Bot. 4: 47. 1917.
On Celastraceae :

Hippocratea volubilis L., Alto Cedro (Prov. Oriente),
March, 1903, Underwood &' Earle 1636; Baracoa (Prov.
Oriente), April 15, 191 6, Johnston 633.
A peculiar white-spored rust, formed on hypertrophied areas.
It also occurs abundantly in Porto Rico, but has not been reported
elsewhere.

113. Aecidium Pisoniae sp. nov.

On Nyctaginaceae :
Pisonia aculeata L., Ceballos (Prov. Camagiiey), Nov. 25,
1 91 5, Johnston 2g8.
Pycnia amphigenous, few, on discolored spots, noticeable,
subepidermal, globose, 96-128 /x in diameter; ostiolar filaments
64-80 /I long.

Aecia hypophyllous, crowded in groups 3-5 mm. across, cupu-
late, 0.1-0.2 mm. in diameter; peridium colorless, the margin
somewhat recurved, erose, the peridial cells rhomboidal in radial
section, 12-16 by 16-34 ;u, abutted, the wall evenly thick, 1.5-2 )u,
the outer wall smooth, the inner wall very finely and closely ver-
rucose; aeciospores globoid or short-oblong, 15-16 by 16-23/1;
wall nearly colorless, thin, about I /x, very finely and closely ver-
rucose.

The rust has been collected only once. Early in July, 1 91 6,
the junior author visited the locality where he had first found it the
year before, and discovered that the forest had been cut down and
burned off preparatory to planting cane. The host is abundant
in the vicinity, but no rust could be detected upon other plants of
it. The germination of the spores has not yet been studied, and
the assignment to the genus is based upon superficial characters,
only. If correctly placed, it is probably heteroecious.

162 Semi-centennial of Torrey Botanical Club

114. Aecidium passifloriicola p. Henn. Hedwigia 43: 168.

1904.

On Passifloraceae :

Passiflora rubra L., Ceballos (Prov. Camaguey), July 6,
1916, Johnston 862.
This rust appears to be quite local. It has been collected in
Porto Rico and Jamaica, and also in Peru, S. A. It is undoubtedly
heteroecious. According to observations made by Whetzel and
Olive in Porto Rico during March and April, 1916 (Mycologia 9:
75. 191 7), it probably is the aecial form of Puccinia Scleriae
(Paz.) Arth., a rust which has not so far been found in Cuba.

115. Aecidium Tournefortiae P. Henn. Hedwigia 34: 338.

1895.

On Boraginaceae :

Tournefortia hirsutissima L., Baracoa (Prov. Oriente),
April 14, 1916, Johnston 505; Paso Estancia (Prov.
Oriente), May 3, 191 6, Johnston yo6.
Tournefortia peruviana Poir., Ceballos (Prov. Camaguey),
July 6, 1916, Johnston 863.
The rust is probably the aecial stage of some heteroecious
species. It has been observed on a phanerogamic specimen in the
herbarium of the N. Y. Bot. Garden, collected at Vento (Prov.
Habana), July i, 1904, P. Wilson 5/(5. It occurs also in Porto
Rico and in South America.

116. Aecidium tubulosum Pat. & Gaill. Bull. Soc. Myc. Fr. 4:

97. 1888.
On Solanaceae:

Solanum torvum Sw., Aguacate (Prov. Habana), March 23,
1903, Holway; Jamal (Prov. Oriente), April 21, 191 6,
Johnston 513.

A very abundant rust in some localities in the tropics, and
doubtless heteroecious with the alternate form on a grass or sedge.
It is known also from Jamaica, Porto Rico, as well as from Central
and South America.

117. Aecidium simplicius sp. nov.
On Bignoniaceae :

Tecoma pentaphylla (L.) Juss., Taco (Prov. Oriente), April
18, 1916, Johnston 518.

Arthur and Johnston: Uredinales of Cuba

163

Aecia hypophyllous, scattered or indefinitely grouped, cupu-
late, short, 0.08-0.15 mm. in diameter; peridium white, margin
erect, lacerate, the peridial cells rhomboidal, 2 1-29 /x long, slightly
or not overlapping, the outer wall rather thick, 4-7 ju, the inner wall
somewhat thinner, 2.5-4 ju, rugose; aeciospores globoid, 21-26 by
25-29 ijl; wall colorless, rather thin, 1-1.5 m» very finely and closely
verrucose.

The material on which this species is founded is scanty, being
only a few small leaves from seedlings. The aecia on them are,
however, quite numerous. In reference to the host, and the place
where it was found, the collector has the following to say.

"In regard to the host there seems to be little chance for mis-
take. The seedlings have 3-5 leaflets identical with those of
Tecoma. The leaflets are very narrow but so are they on the
mature flowering plants in this locality, that is to say on the first
branches. The later leaflets become of a normal width. In one
case the same fungus was found on one leaf of a plant 2 feet high,
while the most of them had only the cotyledonary leaves or per-
haps the second and third pair affected.

"The location of this plant was unique — in the basin at the
foot of a waterfall about a hundred feet high, with constantly
moist condition about the basin. Seedlings were growing on the
moss-covered rocks, and mature plants overhanging the boulders
in the river. It is regretable that the locality is so isolated."

Assuming that this form is a genuine aecium, for the spores
have not yet been germinated, the probability of its being autoe-
cious or heteroecious remains an open question.

118. Aecidium Farameae Arth. Bull. Torrey Club 42 : 592. 1915.
On Rubiaceae:

Faramea occidentalis (L.) A. Rich., San Diego de los Banos
(Prov. Pinar del Rio), Aug. 31-Sept. 3, 1910, Britton,
Earle & Gager 6855.
Only the type collection cited above is known. The germina-
tion of the spores has not yet been observed, and it is by no means
certain that it is not a species of Endophyllum. The more suc-
culent part of the host is often greatly distorted by the rust.

Form-genus Uredo, with paraphyses imbricated to form a pseudo-peridium,
or with cells united into a peridial membrane, mostly forms belonging to Uredin-
ACEAE (Melampsoraceae), nos. 119-121.

164 Semi-centennial of Torrey Botanical Club

119. Uredo Artocarpi R. & Br. Jour. Linn. Soc. Bot. 14: 93.
1873.

Physopella (?) Artocarpi Arth. N. Am. Flora 7: 103. 1907.
On Artocarpaceae :

Artocarpus incisa L.f., Baracoa (Prov. Oriente), April

14, 1916, Johnston ^00.
Castilla elastica Cerv., Santiago de las Vegas, Jan. 30,
1 91 6, Johnston 464.
Heretofore the only American station known for this tropical
rust was in Porto Rico on Artocarpus communis. Only uredinio-
spores have yet been detected, and its affinities are very uncertain.

120. Uredo Coccolobae P. Henn. Hedwigia 35: 253. 1896.
On Polygonaceae :

Coccolohis Uvifera (L.) Jacq., Marianao (Prov. Habana),
Feb. 6, 1916, Johnston 440; Santiago de las Vegas, May
13, 1 91 6, Johnston 6gg.
The rust also occurs in Porto Rico, and in South America.

121. Uredo jatrophicola Arth. Mycologia 7: 331. 1915.
On Euphorbiaceae :

Jatropha Curcas L., Santiago de las Vegas, Oct. 24, 1915,
Johnston 12^; Saetia (Prov. Oriente), April 8, 1916,
Johnston 514.

Jatropha gossypifolia L., Soledad, Cienfuegos (Prov. Santa
Clara), Nov. 5, 191 5, Johnston 208; Baracoa (Prov.
Oriente), April 14, 1916, Johnston 587.
This unconnected form has been found in the phanerogamic
herbarium of the N. Y. Bot. Garden on Jatropha gossypifolia
from Havana, April 7, 1903, J. A. Shafer 86, and from Rio Togaba,
Trinidad (Prov. Santa Clara), March 15, 1910, Britton & Wilson
5549-

The rust is also known from Porto Rico and St. Domingo.

Form-genus Uredo, with paraphyses absent, or if present, free and peripheral,
mostly forms belonging to Aecidiaceae (Pucciniaceae), nos. 122-140.

122. Uredo Gymnogrammes P. Henn. Hedwigia 34: 337. 1895.
On Polypodiaceae :

Pityrogramma calomelanos (L.) Link {Gymnogramma calo-
melanos Kaulf.), El Yunque, Baracoa (Prov. Oriente),
March 10, 1903, Holway.

Arthur and Johnston: Uredinales of Cuba

165

This imperfectly known fern rust has been collected in Jamaica
and Porto Rico on numerous hosts.

123. Uredo paspalicola p. Henn. Hedwigia 44: 57. 1905.
Uredo Stevensiana Arth. Mycologia 7: 326. 1915.

On Poaceae:

Bamhos vulgaris Schrad., Santiago de las Vegas, Jan. 29,

1916, Johnston 424.
Leptochloa domingensis (J acq.) Trin., Antilla (Prov. Oriente),

April 8, 1916, Johnston 542.
Paspahim conjugatum Berg., Herradura (Prov. Pinar del
Rio), March, 191 7, Home.
This imperfectly known, pale-spored. South American rust,
recently recognized from Porto Rico and Guatemala, is now first
recorded for Cuba. The second Cuban host is a new one for the
species.

124. Uredo Fuirenae P. Henn. Hedwigia Beibl. 38: 70. 1899.

On Cyperaceae:

Fuirena umbellata Rottb., Batabano (Prov. Habana),
Oct. 3, 1904, Baker & Wilson 2214 (host no. 2213); Her-
radura (Prov. Pinar del Rio), March 30, 1907, Earle 652;
Siguanea, Isle of Pines, March 12, 191 6, Britton, Britton
& Wilson 15387.
This imperfectly known rust has been found in Porto Rico,
Brazil and India, in each instance on Fuirena umbellata, and only
with urediniospores.

125. Uredo superior Arth. Bull. Torrey Club 31:5. 1904.
On Cyperaceae:

Fimhristylis ferruginea (L.) Vahl, Saetia (Prov. Oriente),
April 8, 191 6, Johnston 646.
The species is elsewhere known only from Porto Rico.

126. Uredo Dioscoreae P. Henn. Hedwigia 35: 255. 1896.
On Dioscoreaceae :

Dioscorea sp., El Yunque, Baracoa (Prov. Oriente), March

12, 1903, Holway.
Rajania cordata L., Toa (Prov. Oriente), April 18, 1916,

Johnston 554.

The reexamination of all West Indian collections of Dioscorea

166 Semi-centennial of Torrey Botanical Club

rust seems to indicate that, while there is some variation as to
size of urediniospores and thickness of walls, yet all may be con-
sidered to be one species, and also to be the same as the type ma-
terial of Uredo Dioscoreae P. Henn. from Brazil. The surface of
the spores is conspicuously echinulate (not "verrucose," as erro-
neously stated in the Uredinales of Porto Rico (Mycologia 7 : 320.
1 91 5)), and the pores indistinct but probably two and equatorial,
or somewhat superequatorial.

Puccinia valida Arth., on Dioscorea convolvulacea, from Jalapa,
Mexico, has uredinia that in both gross and microscopic appear-
ance agree quite well with the West Indian material. The chief
differences, aside from being intermixed with the telia, are the
darker and thicker walls of some of the urediniospores. There
are good reasons for thinking that all the West Indian collections
belong to some Uromyces or Puccinia, possibly to the Mexican
species of Puccinia.

127. Uredo gynandrearum Corda, Icones Fung. 3: 3. 1839.
On Orchidaceae:

Habenaria maculosa L.
This imperfectly known orchid rust was observed on a phanero-
gamic specimen in the herbarium of the N. Y. Bot. Garden, col-
lected on the side and top of El Yunque (Prov. Oriente), Dec. 30,
1910, /. A. Shafer 7992. It is known also from Porto Rico and
Trinidad, and from Central and South America.

128. Uredo nigropunctata P. Henn. Hedwigia 35: 254. 1896.
On Orchidaceae:

Bletia patula Hook. •
This imperfectly known orchid rust was observed on a phanero-
gamic collection in the herbarium of the N. Y. Bot. Garden, col-
lected upon El Yunque, Baracoa (Prov. Oriente), March, 1903,
Underwood & Earle 929. It is also known from Porto Rico,
Haiti, the Bahamas, and from Florida and South America.

129. Uredo Cherimoliae Lagerh. Bull. Soc. Myc. Fr. 11: 215.

1895.

On Annonaceae:

Annona reticulata L., Santiago de las Vegas, March 2, 1916,
Johnston 492.

Arthur and Johnston: Uredinales of Cuba

167

Annona squamosa L., Santiago de las Vegas, June 25, 1916,
Johnston 848, Nov. 3, 191 7, Johnston pjj.
The species has, heretofore, been known only from Ecuador,
S. A., and on A. Cherimolia. It is well characterized, and easily
separated from other species of Uredo on similar hosts, by the
thin-walled spores and the greatly thickened extremities of the
paraphyses.

130. Uredo bauhiniicola P. Henn. Hedwigia 34: 98. 1895.
On Caesalpiniaceae (Cassiaceae) :

Bauhinia heterophylla Kunth, Aguacate (Prov. Habana),
March 23, 1903, Holway; Guanajay (Prov. Pinar del
Rio), Sept. 13, 1904, Earle 1499; Candelaria (Prov. Pinar
del Rio), Jan. 191 7, Home.
The type of this imperfectly known rust was obtained by Ule
in Brazil, on Bauhinia ruhiginosa Bong. It is here first reported
for North America. The spores have three equatorial pores,
usually to be seen without difficulty. Paraphyses are absent.
The form is probably a stage of some Uromyces.

131. Uredo Hymenaeae Mayor, Mem. Soc. Neuch. Sci. 5: 585.

1913-

On Caesalpiniaceae (Cassiaceae):

Hymenaea Courbaril L., Ceballos (Prov. Camagiiey), Nov.
26, 1 91 5, Johnston 296.
This imperfectly known rust occurs also in Porto Rico and
South America. It may belong to the genus Ravenelia.
1^2. Uredo Arachidis Lagerh. Tromso Mus. Aarsh. 17: 106.
1894.
On Fabaceae:

Arachis hypogea L., Santiago de las Vegas, Sept. 27, 1915,
Johnston 164.

This slightly known rust is also reported from Porto Rico,
Guadeloupe, and Grenada, as well as from Trinidad and the con-
tinent of South America. It has also been sent to the senior
author by Mr. W. Robson from Montserrat, British West Indies,
where some seasons it has become a menace to the peanut crop.
133. Uredo Cabreriana Kern & Kellerm. Jour. Myc. 13 : 25. 1907.
On Fabaceae:

Erythrina glauca Willd.

168 Semi-centennial of Torrey Botanical Club

This imperfectly known rust was found in the phanerogamic
herbarium of the N. Y. Bot. Garden, collected at Paso Real (Prov.
Pinar del Rio), 1906, Abarca & 0' Donovan 2634. It is known also
from Porto Rico and from Guatemala, on the same host.

134. Uredo Erythroxylonis Graz. Bull. Soc. Myc. Fr. 7: 153.

1891.

On Erythroxylonaceae :

Erythroxylon havanense Jacq., San Antonio de los Bafios
(Prov. Habana), Nov. 21, 1904, Baker 41 2y (Barth.
Fungi Columb. 228/'); San Diego de los Banos (Prov.
Pinar del Rio), Feb. 7, 191 5, Johnston 174; Soledad,
Cienfuegos (Prov. Santa Clara), Nov. 5, 1915, Johnston
ig8; Santiago de las Vegas, Oct. 24, 191 5, Johnston 131;
Cerros de Vivijagua, Isle of Pines, Feb. 28-29, 1916,
Britton, Britton & Wilson 15023; Taco Taco (Prov. Pinar
del Rio), Sept. 17, 191 6, Johnston SyS.
This imperfectly known species was also detected on a phanero-
gamic collection of the same host in the N. Y. Bot. Garden, from
Sierra de Anafe (Prov. Pinar del Rio), Dec. 28, 191 1, Percy Wilson
ii56g. The only other station in North America for this South
American rust is on Mona Island, a small island not far from
Porto Rico.

135. Uredo Saviae sp. nov.

On Euphorbiaceae:

Savia sessiliflora (Sw.) Willd., San Juan, Isle of Pines,
March 15, 17, 1916, Britton, Britton & Wilson 15453.

Uredinia hypophyllous, scattered, oval or oblong, 0.1-0.3 mm.
long, subepidermal, rather tardily naked, cinnamon-brown, rup-
tured epidermis usually overarching and conspicuous; peridium
and paraphyses wanting; urediniospores angularly obovoid, usually
triangular above, 16-20 by 23-29 /i; wall cinnamon-brown, thin,
about I ju, closely echinulate, the pores 3, approximately equatorial,
in the projecting angles.

The spores have an unusual appearance from the position of
the pores in the three corners. Seen from above the spores appear
triangular. They are usually narrowed and sometimes shrunken
below the pores, giving a pyriform appearance from the side.

Arthur and Johnston: Uredinales of Cuba 169

136. Uredo Sapotae sp. nov.

On Sapotaceae:

Achras Sapota L. (Sapota Achras Willd.), Santiago de las
Vegas, March 5, 1916, Johnston 4Q3 (type).

Uredinia hypophyllous, scattered or somewhat gregarious in
close groups of a few each on small discolored areas, 0.5-1 mm.
across, oval, 0.1-0.3 mm. long, subepidermal, rather tardily naked,
cinnamon-brown, pulverulent, opening by a lateral rupture of the
epidermis which remains as an evident flap; peridium and para-
physes none; urediniospores in one view triangular, when revolved
one fourth broadly obovate, 18-22 by 21-26 ju; wall cinnamon-
brown, moderately thick, 1.5-2^1, closely echinulate, the pores 2,
opposite and close to the hilum, in the obovoid view of the spore
seen in the lateral walls.

The same fungus on the same host was collected by E. W. D.
Holway at Nassau, Bahamas, March 2, 1903.

137. Uredo Lucumae sp. nov.

On Sapotaceae:

Lucuma nervosa A. DC, Santiago de las Vegas, June 25,

1915, O, II, 146 (type), Feb. 23, 1916, II, March 5,

1916, O, II, 4Q4, all by Johnston.

Pycnia amphigenous, numerous, on discolored spots 5-15 mm.
across, punctiform, honey-yellow becoming brown, subcuticular,
hemispherical in section, 80-100 broad; ostiolar filaments
wanting.

Uredinia amphigenous, rarely only hypophyllous, surrounding
and among the pycnia on purplish-brown spots, usually crowded,
irregularly roundish, 0.1-0.3 mm. in diameter, subepidermal,
early naked, pulverulent, the hypertrophied tissues forming a
protective structure for the sorus ; urediniospores globoid or broadly
ellipsoid, 24-35 by 35-42 fi, larger when wet by swelling of the
gelatinous layer; wall lamellate, the inner portion firm, golden-
or cinnamon-brown, 2-3 /x thick, the outer portion pale, swelling
to 4-9 M thick, the cuticle bearing coarse, conical tubercles, the
pores indistinct.

The distinctive character of this rust suggests its relation to
Prospodium or Uropyxis. The Uromyces Lucumae Diet., from
Brazil, is a wholly unlike species, judging from the description.

138. Uredo Operculinae Arth. Mycologia 9: 95. 191 7.
On Convolvulaceae :

Operculina dissecta (Jacq.) House (Convolvulus dissectns

170 Semi-centennial of Torre y Botanical Club

Jacq.), vicinity of Santiago de Cuba, Feb. 14, 1892,
Pollard & Palmer 271.
This most unusual form of uredinia constitutes a species here-
tofore known only from the type collection taken in Porto Rico.

139. Uredo Cephalanthi Arth. Bull. Torrey Club 29: 231. 1902.
On Rubiaceae:

Cephalanthus occidentalis L., Vivijagua, Isle of Pines,
March 18-20, 191 6, Britton &' Wilson 1560Q.
Until this collection came to hand the species has only been
known from the type material from southern Florida. It is an
inconspicuous form, with applanate sori and no protecting struc-
tures.

140. Uredo proximella Arth. Mycologia 7: 324. 1915.
On Cichoriaceae :

Lactuca intyhacea Jacq.
This rust has been found on a phanerogamic specimen in the
herbarium of the N. Y. Bot. Garden, collected at Guantanamo
Bay (Prov. Oriente), March 17-30, 1909, N. L. Britton 2 161. It
is also known from Porto Rico and St. Domingo. The rust ap-
pears to belong under the genus Puccinia, but no teliospores have
yet been discovered.

Index to Uredinales
New and newly combined names are in bold face type

Aecidium circumscriptum iii

Cissi III

desmium lo

Farameae 118

punctatum 26

Nymphoidis 69

passifloriicola 114

Pisoniae 113

Rivinae 74

roseum 108

simplicius 117

Tournefortiae 115

tubulosum 116
Allodus megalospora 91
Botryorhiza Hippocrateae 112
Calliospora Farlowii 25
Cerotelium Fici 9

Gossypii 10
Cionothrix Cupaniae 14
Coleosporium Elephantopodis I

Eupatorii 5

Ipomoeae 3

Plumierae 4

Vernoniae 2
Cronartium Byrsonimatis 12

notatum 12

Wilsonianum 13
Dendroecia Lysilomae 20
Endophyllum circumscriptum iii

Rivinae 74

singulare 12
Eriosporangium evadens 105

tucumanense 95
Gymnosporangium guaraniticum 49
Klebahnia Bidentis 54
Kuehneola Fici 9

Gossypii 10

malvicola 11
Nephlyctis transformans 32
Nigredo Columbiana 55
Patouillardiella guaranitica 49
Phakopsora Aeschynomenis 8

Vignae 7

Vitis 6

Arthur and Johnston: Uredinales of Cuba

Physopella Aeschynomenis 8
Artocarpi 119
concors 7
Fici 9
ficina 9
Vitis 6

Prospodium Amphilophii 29

appendiciilatum 27, 100

Lippiae 30

plagiopus 28

tuberculatum 31
Puccinia abrupta 106

Adenocalymnatis 100

aequinoctialis 100

Amphilophii 29

Anthephorae 61

appendiculata 27

Arechavelatae 78

barbatula 77

Blechi 1 01

canaliculata 65

Cannae 71

Cenchri 57

Chaetochloae 60

Chaseana 61

compacta 87

concrescens 87

Conoclinii 108

coronata 62

crassipes 90

cuticulosa 100

Cynanchi 89

Cyperi 65

deformata 58

elegans 32

Eleocharidis 66

Eleutherantherae 109

evadens 105

exitiosa 32

Fuirenae 68

fuscella 104

globosipes 99

Gonolobi 88

Gouaniae 79

graminis 63

Helianthi 107

heterospora 81

Huberi 59

Hydrocotyles 85

Hyptidis 96

inflata 76

insititia 97

invaginata 80

Johnstonii 86

Lantanae 92

lateripes loi

lateritia 102

Leonotidis 98

Lippiae 30

Ludwigiae 83

macropoda 73

malvacearum 82

Puccinia medellinensis 95

megalospora 91

obliqua 89

plagiopus 28

poculiformis 63

Polygoni-amphibii 72

Pruni-spinosae 26

Psidii 84

purpurea 56

Raunkiaerii 74

Rhamni 62

Rivinae 74

Ruelliae loi

rosea 108

salviicola 94

Scirpi 69

Scleriae 114

scleriicola 67

Smilacis 70

solida 109

Sorghi 64

sphaerospora 89

striolata 73

substriata 60

Synedrellae 109

transformans 32

Tridacis 109

tuberculata 31

Urbaniana 93

valida 126

Vernoniae 104

Xanthii 103

Zorniae 75
Pucciniosira pallidula 110
Ravenelia cubensis 22

Humphreyana 21

Indigoferae 15

Lonchocarpi 17

Lysilomae 20

Piscidiae 16

Pithecolobii 19

portoricensis 23

pulcherrima 21

siliquae 18
Sphaerophragmium Dalbergiae 33
Tranzschelia punctata 26
Trichobasis euphorbiaecola 48
Uredo Adenocalymnatis 100

Aeschynomenis 8

Anthephorae 61

Arachidis 132

Artocarpi 119

bauhiniicola 130

Cabreriana 133

Cephalanthi 139

Cherimoliae 129

Coccolobae 120

concors 7

cristata 49

cuticulosa 100

Dalbergiae 33

Desmodii-tortuosi 46

172 Semi-centennial of Torrey Botanical Club

Uredo Dioscoreae 126
Erythroxylonis 134
Fici 9
ficicola 9
ficina 9
Fuirenae 124
Gossypii 10
Gouaniae 80
Gymnogrammes 122
gynandrearum 127
Hibisci 11
Hymenaeae 131
ignobilis 36
jatrophicola 121
Lucumae 137
malvicola 11
moricola 9
nigropunctata 128
notata 12
Operculinae 138
paspalicola 123
proximella 140
Sapotae 136
Saviae 135
Sissoo 33
Stevensiana 123
striolata 73
superior 125
Vignae 7
Zorniae 75

Host

Abena jamaicensis 93
Abutilon abutiloides 81

hirtum 81

indicum 81

lignosum 81

permoUe 81
Acacia Farnesiana 18
Acanthaceae loi
Achras Sapota 136
Adenocalymna sp. 100
Aeschynomene americana 8
Ageratum maritimum 108
Amaranthaceae 40, 73
Ambrosiaceae 103
Ammiaceae 85
Amygdalaceae 26
Amygdalus Persica 26
Andropogon halepensis 56
Anoda hastata 81
Annona Cherimolia 129

reticulata 129

squamosa 129
Annonaceae 129
Anthephora elegans 61

hermaphrodita 61
Apocynaceae 4
Arachis hypogea 132
Artocarpaceae 9, 119
Artocarpus communis 119

Uromyces appendiculatus 44

bidenticola 54

Bidentis 54

Celosiae 40

columbianus 55

Commelinae 39

Cupaniae 49

Dolicholi 45

dolichosporus 52

Eragrostidis 35

Euphorbiae 48

gemmatus 51

Hedysari-paniculati 46

Hellerianus 53

Howei 50

ignobilis 36

insularis 43

jamaicensis 41

Janiphae 47

leptodermus 34

Lucumae 137

major 36

Medicaginis 42

Medicaginis-falcatae 42

Neurocarpi 43

proeminens 48

Rhyncosporae 37

Scleriae 38

solidus 46
Uromycladium (?) cubense 24

INDEX

Artocarpus incisa 119
Asclepiadaceae 50, 87, 88, 89
Asclepias curassavica 50, 87

nivea 50
Avena sativa 62

Baccharis sp. 105
Bam bos vulgaris 123
Banisteria laurifolia 77
Bauhinia divaricata 41

heterophylla 130

rubiginosa 130
Bidens leucantha 54

pilosa 54

Bignoniaceae 27, 28, 29, 32, 100, 117
Bignonia aequinoctialis 100
Blechum Brownei loi
Bletia patula 128
Boraginaceae 52, 115
Borreria laevis 102
Byrsonima coccolobifolia 12
crassifolia 12

Caesalpiniaceae 21, 22, 23, 130, 131
Caesalpinia bahamensis 21

pulcherrima 21

Rugeliana 21
Cajan Cajan 45
Cajanus indicus 45

Arthur and Johnston: Uredinales of Cuba

173

Cannaceae 71
Canna indica 71

Cardiospermum microcarpum 78
Carduaceae i, 2, 5, 54, 55, 104, 105,

107, 108, 109
Castilla elastica 119
Cassia emarginata 23

robinaefolia 22
Cassiaceae 21, 130, 131
Cayaponia racemosa 53
Celastraceae 112
Cenchrus echinatus 57

viridis 57
Cephalanthus occidentalis 139
Chaetochloa geniculata 60

imberbis 60

onurus 60

purpurascens 60

setosa 60

verticillata 60
Chamaesyce hirta 48

hypericifolia 48
Cichoriaceae 140
Cissus rhombifolia 13

sicyoides iii
Clitoria rubiginosa 43
Cocqolobis Uvifera 120
Commelinaceae 39
Commelina longicaulis 39

nudiflora 39
Convolvulaceae 3, 51, 90, 91, 138
Convolvulus dissectus 138

nodiflorus 51
Cucurbitaceae 53
Cupania americana 14

glabra 14

macrophylla 49
Cydista aequinoctialis 100
Cyperaceae 37, 38, 65, 66, 67, 68,

125

Cyperus ferax 65

Dalbergia Amerimnum 33
Dalea domingensis 25
Desmodium Scorpiurus 46

tortuosum 46
Diodia sp. 102
Dioscoreaceae 126
Dioscorea convolvulacea 126
Dipholis salicifolia 86
Dolichos Lablab 44
Dolicholus reticulatus 45

Eleocharis capitata 66

geniculatus 66
Elephantopus mollis i
Eleutheranthera ruderalis 109
Emilia sonchifolia 109
Eragrostis tephrosanthos 35
Ernodia sp, 102
Erythrina glauca 133
Erythroxylonaceae 134

Erythroxylon havanense 134
Eugenia Jambos 84
Eupatorium macrophyllum 5
106, villosum 108

Euphorbiaceae 47, 48, 121, 135
Euphorbia heterophylla 48

hirta 48

hypericifolia 48

pilulifera 48

Fabaceae 7, 8, 15, 16, 17, 25, 33, 41, 42,
43, 44, 45, 46, 75, 132, 133

Faramea occidentalis 118

Ficus Carica 9
Combsii 9

Fimbristylis ferruginea 125

Fischeria crispiflora 89

Frangulaceae 79, 80

Fuirena simplex 68
unibellata 124

Gaya occidentalis 81
Gossypium acuminatum 10
Gouania domingensis 79, 80

lupuloides 79, 80

polygama 79, 80

tomentosa 79
Gymnogramma calomelanos 122

Habenaria maculosa 127
Helianthus annuus 107
Hemidiodia ocimifolia 102
Hibiscus syriacus 11
Hippocratea volubilis 112
Holcus halepensis 56

Sorghum 56
Hydrocotyle australis 85
Hymenaea Courbaril 131
124, Hyptis capitata 96

lantanifolia 97

pectinata 95

radiata 96

suaveolens 95, 97

Ichthyomethia Piscipula 16
Indigofera suffruticosa 15
Ipomoea acuminata 3, 90

Carolina 91

cathartica 3, 90

Learii 3

mutabilis 3

stolonifera 3

triloba 90
Iresine angustifolia 73

Celosia 40, 73

celosioides 73

elatior 73

paniculata 40, 73
Isnardia repens 83

Jacquemontia nodiflora 51
tamniflora 3

174 Semi-centennial of Torrey Botanical Club

Jambos Jambos 84

vulgaris 84
Jatropha Curcas 121

gossypifolia 121

Manihot 47

Labiatae 94, 95, 96, 97, 98
Lablab vulgaris 44
Lachnorhiza piloselloides 2
Lactuca intybacea 140
Lamiaceae 94, 95, 96, 97, 98
Lantana Camara 92

involucrata 31, 92

odorata 92

reticulata 92

trifolia 92
Leonotis nepetaefolia 98
Leptochloa domingensis 123
Limnanthemum Grayanum 69
Lippia dulcis 30, 92

stoechadifolia 92
Lonchocarpus campestris 17

latifolius 17
Lucuma nervosa 137
Lycium carolinianum 99
Lysiloma bahamensis 20

tergemina 20

Malache scabra 11
Malpighiaceae 12, 76, 77
Malvaceae 10, 11, 81, 82
Malvastrum corchorifolium 82

coromandelianum 82
Malvaviscus arboreus sagreanus 1 1
Manihot Manihot 47

utilissima 47
Medicago sativa 42
Meibomia Scorpiurus 46

tortuosum 46
Melanthera brevifolia 55

hastata cubensis 55
Melothria guadalupensis 53
Menyanthaceae 69
Mesosphaerum capitatum 96

lantanifolium 97

pectinatum 95

rugosum 96

suaveolens 95
Metastelma penicillatum 89
Mimosaceae 18, 19, 20, 24
Mimosa asperata 24

pigra 24
Mitracarpum sp. 102
Myrtaceae 84

Neurolaena lobata 109
Nyctaginaceae 113

Olyra latifolia 58
Onagraceae 83
Operculina dissecta 138
Orchidaceae 127, 128

Panicum barbinode 34

fasciculatum 59

sanguinale 60

trichoides 59
Parosela domingensis 25
Paspalum conjugatum 123

virgatum 59
Passifloraceae 114
Passiflora rubra 114
Persicaria punctata 72
Petiveriaceae 74
Pharbitis cathartica 3
Phaseolus vulgaris 44
Philibertella clausa 88
Phytolaccaceae 74
Pisonia aculeata 113
Pithecoctenium echinatum 29
Pithecolobium tortum 19
Pityrogramma calomelanos 122
Plumiera emarginata 4

obtusa 4

rubra 4

Poaceae 34, 35. 36, 56, 57. 58. 59- 60, 61,

62, 63, 64, 123
Poinciana pulcherrima 21
Poinsettia heterophylla 48
Polygonaceae 72, 120
Polygonum acre 72

punctatum 72
Polypodiaceae 122
Priva lappulacea 92
Prunus Persica 26

Rajania cordata 126
Rhamnaceae 79, 80
Rivina humilis 74

octandra 74
Rubiaceae 102, 118, 139
Rynchospora distans 37

Salvia occidentalis 94
Sapindaceae 14, 49, 78
Sapota Achras 136
Sapotaceae 86. 136, 137
Savia sessiliflora 135
Scirpus capitatus 66

lacustris 69
Scleria lithosperma 38

verticillata 67
Setaria setosa 60
Sida angustifolia 81

glutinosa 81

procumbens 81

spinosa 8i
Sideroxylon foetidissimum 86
Smilaceae 70
Smilax havanensis 70
Solanaceae 99, 116
Solanum torvum 116
Sorghum halepensis 56

vulgare 56
Spermacoce sp. 102

Arthur and Johnston: Uredinales of Cuba

Sporobolus indicus 36
Stachytarpheta jamaicensis 93
Stenolobium stans 27, 32
Stigmaphyllon lingulatum 76

periplocifolium 76

reticulatum 76

Sagraeanum 76
Synedrella nodiflora 109
Syntherisma sanguinalis 60

Tecoma lepidota 28

pentaphylla 28, 117
stans 27, 32

Tiliaceae 110

Tournefortia hirsutissima 115

peruviana 115

volubilis 52
Tridax procumbens 109
Triticum sativum 63

vulgare 63

Triumfetta semilobata no

Umbelliferae 85

Vachellia Farnesiana 18
Valerianodes jamaicensis 93
Verbenaceae 30, 31, 92, 93
Vernonia menthaefolia 104
Vigna repens 44
vexillata 44
Viguiera helianthoides 106
Vitaceae 6, 13, in
Vitis vinifera 6

Wissadula periplocifolia 81

Xanthium longirostre 103
saccharatum 103

Zea Mays 64
Zornia diphylla 75

THE PHYSIOLOGICAL PROPERTIES OF TWO
SPECIES OF POISONOUS MUSHROOMS*

By Michael Levine

(with plates I AND 2)

The older literature of mushroom poisoning has been thor-
oughly summarized in the works of Paulet (1793), Gillot (1900),
Ford (1906-7-8), Ferry (191 1), Sartory (1914) and others. With-
in the last decade more careful experimental studies have been
made by mycologists, physiologists, and physicians, which throw
new light on the physiological and toxicological effects both of the
older well-known poisonous forms and certain species hitherto
unsuspected or unknown.

Gillot (1900) studied the physiological effects of the extracts
of eight speciesf of mushrooms on dogs, guinea pigs, and rabbits.
Various quantities (25-100 gm.) of each species were macerated
with ether and filtered. The filtrate was evaporated and dried at
100° for two hours. The residue was either diluted in water and
injected subcutaneously or powdered and then fed to dogs. The
quantities injected varied from 3 to 5 c.c. All species caused
death in guinea pigs except Lactaria rufa, Cantharellus auran-
tiacus, and Hypholoma fasiculare, which were found to be harm-
less.

Ford (191 1), besides making his well-known chemical studies
of the Amantia species, has also investigated haemolysins, agglu-
tinins and toxins found in a number of other species of fungi.
Aqueous extracts of the macerated pilei were studied by injecting
subcutaneously in different animals a given quantity of the ex-
tract before and after boiling. He was thus able to determine the
presence of thermolabile or thermostabile haemolysins, agglutinins,
and toxins. Besides studying eight species of Amanita and

* From the Department of Physiology, Columbia University.

t Clitocyhe nehularis, Cantharellus aurantiacus, Hypholoma sublateritium, H .
fasiculare, Russula sanguinea, R. Queletii, Lactaria vellerea, and L. rufa.

176

Levine: Two species of poisonous mushrooms 177

Amanitopsis volvata he investigated five species from other genera*
in which he found a haemolytic substance. In three other speciesf
of agarics he found heat-resisting haemolysins while in six species
of Entoloma, two species of Hypholoma, and four species of Boleti
he found no haemolysin. In all, thirteen species of fungi showed
the presence of agglutinins which were destroyed by heating to
60-65° C. for one half hour, while eight species contained heat-re-
sisting agglutinins which in this respect resembled the agglutinin
of the fly agaric, Amanita muscaria, as he had earlier described it
(1906, 1909). Ford has made it clear that while extracts of various
poisonous mushrooms when injected subcutaneously into guinea
pigs or rabbits show evidence of toxic properties and also contain
haemolysins and agglutinins, t other poisonous mushroom juices
may lack entirely or be deficient in both haemolysins and agglu-
tinins,! while still other species show haemolysins and agglutinins
and yet are not toxic. || Ford maintains that herbivora are en-
tirely non-susceptible to the action of the most virulent of the
poisonous species of mushrooms when they are introduced into
the stomach. Later Ford in collaboration with Sherrick (1911)
investigated the toxic properties of ten species of Boleti and of
Clitocybe dealhata sudorifica. They found that Boletus felleus,
B. miniato-olivaceus , and B. chromapes have no agglutinin or hae-
molysin but contain some toxic substance which when injected
subcutaneously into the body of the guinea pig caused chronic
intoxication and finally death. This supports in the main the
claim of Collins (1898) who reported a non-fatal case of mushroom
poisoning due to eating Boletus miniato-olivaceus var. sensibilis.
Rabbits were likewise affected by extracts of B. felleus. Four
other species of Boletus {B. affinis, B. ornatipes, B. bicolor and
• B. separans) were found to have haemolysins or agglutinins but

* Clitocybe muUiceps, Hygrophorus pratensis cinereus, H. praiensis albus, H.
marginatus and Lactaria torminosa.

t Inocybe infelix, Galera tenera, and Naucoria firma.

t Amanita phalloides, A. muscaria, Lactaria torminosa, and Inocybe infelix.

^Amanita Morrisii, A. spreta, A. citrina, A. crenulata, Amanitopsis vaginata,
Clitocybe illudens, Lactaria uvida, Russula squalida, Tricholoma ustale, Hygrophorus
pratensis cinereus, H. pratensis albus, H. hypothejus, H. conicus, Entoloma nidorosum.
E. sinuatum, E. salmoneum, E. strictius, E. cuspidatum, E. rhodopolium, Hypholoma
instratum, H. cernuum, Panaeolus retirugis, and Boletus palusler.

II Clitocybe muUiceps, Hygrophorus parvulus, Flamm'ila betulina, Galcra tenera,
and Naucoria firma.

178 Semi-centennial of Torrey Botanical Club

were not toxic to guinea pigs or rabbits, while B. Ravenelii, B.
Roxanae, and Strohilomyces strobilaceus were found to have no
haemolysin or agglutinin and were not toxic to any animal. Ford
and Sherrick substantiated Peck's (191 1) earlier report on the
effect of ingesting Clitocyhe dealbata sudorifica. Extracts of this
fungus injected subcutaneously into the body of a rabbit caused
excessive salivation ; the animal became weak and apparently very
sick, but recovery followed in twenty-four hours. A similar ex-
tract injected into the body of a guinea pig caused death within
fifteen minutes. The authors conclude that the extract of Clito-
cyhe dealbata sudorifica is similar in its effects to that of muscarine
or the pilocarpine series. In a further paper Ford and Sherrick
(191 3) report an attempt to isolate the toxic substances in Clito-
cybe dealbata sudorifica, Inocybe decepiens, and Pholiota autumnalis.
They found that the chemically extracted preparations of all these
mushrooms were toxic to rabbits and guinea pigs.

The extracts of the American forms of Helvetia esculenta gave
negative results when injected into guinea pigs and rabbits, though
the European species shows toxic properties which have been
attributed to helvellic acid, as first described by Boehm and Kiilz
(1885).

Clark and Kantor (191 1) studied the effects of extracts of
Amanita muscaria, Inocybe infida, and Clitocybe multiceps on frogs.
Extracts of these mushrooms were injected into the dorsal lymph-
sac of frogs or were given to them by way of the mouth. The
Amanita extract killed a frog in from foui* to twenty- five minutes
after injection. The Inocybe extract caused lethargy, paralysis,
and finally death. Clitocybe multiceps, unlike the other species of
Clitocybe reported by Ford and Sherrick (1911, 191 3) and Gillot
(1900), proved to be harmless. Clark and Smith (1913) continued
further the investigations of the properties of Amanita muscaria,
Inocybe infida, Clitocybe multiceps, and another species, Clitocybe
illudens, on the hearts of frogs. The heart was exposed and con-
nected with the kymograph so that a record of the normal con-
traction was made. The ext^-act in physiological salt solution
was next applied to the heart. In the case of the application of
the extracts of the Amanita, Inocybe, and Clitocybe illudens, the
heart slowed up and soon ceased to beat. This effect was then

Levine: Two species of poisonous mushrooms 179

overcome by the application of atropine. When Clitocyhe multi-
ceps was applied no effect was noted. These authors hold that
Clitocyhe illudens and Inocybe infida contain a muscarine-like sub-
stance as had already been pointed out by Ford and Sherrick
(1911, 191 3). Radaisand Sartory (i 914), in an endeavor to deter-
mine the part of the fungus that contains the most toxin, investi-
gated all the structures of Volvaria gloiocephala and V. speciosa.
They found that the bulb and pileus contained the most poison
and the stipe the least.

Murrill (1909) reported a case of poisoning due to ingestion of
Panaeolus papilionaceus and Inocybe infida experienced by a phy-
sician and members of his family. The mushrooms caused a queer
feeling, an increased heart action, excessive perspiration, and
diarrhoea. Another member of the household was prostrated.
Murrill (191 6) later reported the appearance of a new toxic species
in this genus which he called Panaeolus venenosus Murrill. The
material came from commercial mushroom beds. The plants had
been mistaken and eaten for Agaricus campestris, the commercial
mushroom. The persons after eating of the plant became ill,
dazed, and experienced a rapidly decreasing heart action and
dilation of the pupils. A physician diagnosed the case as poisoning
due to muscarine.

In studying the development of Agaricus campestris, I found
in 1916-1917 in the mushroom beds of one of the largest mush-
room growers in the region of New York City great numbers of
the new fungus which Murrill had described as Panaeolus venenosus
Murrill, These plants appeared about the same time as Agaricus
campestris and were scattered among it; and since the spawning
of the different beds was so timed as to give a constant supply of the
market mushroom throughout the winter, Panaeolus venenosus
also appeared regularly from October, 19 16, to May, 191 7. Panae-
olus venenosus (pl. i, figs. 1-8; PL. 2, figs. 9-15) is a small
mushroom with a fulvous or isabelline (Ridgway) colored cap 3-5
cm. in diameter. Its stem is in length about twice the diameter
of the cap and is covered with a white tomentum at its base, while
its upper part is striate and covered with fine scales. The stipe
is slightly darker in color than the pileus. The plants appear in
fairy rings about one to two feet in diameter. Very often two or

180 Semi-centennial of Torrey Botanical Club

three plants grow in a cluster with the bases of their stipes united
(PL. I, FIGS. 1,2; PL. 2, FIG. 9). The odor of this plant is like that
of the commercial mushroom but the two plants could never be
mistaken for each other.

I have studied the effect of infusions of this fungus (I) on
guinea pigs and rabbits when injected into or fed to these animals;
(II) on the heart-beat and blood-pressure of a cat; (III) on a
skeletal muscle such as the gastrocnemius of the frog, and finally
(IV) on the sciatic and vagus nerves of the frog.

Material

While Panaeolus venenosus grew rather abundantly in the
mushroom beds where I made my collections, no great bulk of
it could be obtained at any one time. The largest quantity I
collected at any one time weighed from twenty-five to forty grams.
I have as yet made no attempt to isolate the toxic element of the
fungus in a pure state, but its common occurrence in commercial
mushroom beds makes a knowledge of the general physiological
effects highly desirable from a practical point of view. It is also
of much interest to compare its physiological effects with those of
the other well-known types of poisonous fungi. With two excep-
tions (extracts Nos. 5 and 6) I used for my experiments infusions
of this plant and the different infusions listed below give some idea
of the amount of material available and the relative toxicity to be
expected of the plants, so far as their juices are concerned. The
plants collected were brought to the laboratory where they were
washed in tap water to remove the soil particles and rinsed in dis-
tilled water. They were then dried by applying filter paper to
them and after that they were weighed and the juices were ex-
pressed.

Infusion No. i was made by macerating 25 gm. of Panaeolus
venenosus in 25 c.c. of distilled water. The filtrate was diluted
with sufificient water to make up 25 c.c. of fluid. The filtrate was
yellowish orange in color at first but after standing for four hours
its color changed to dark brown.

Infusion No. 2 consisted of the juices of 25 gm. of P. venenosus
ground with sand in a mortar to which 75 c.c. of distilled water
had been added. The filtrate from the entire mass was used.

Levine: Two species of poisonous mushrooms 181

The plants used in making this infusion are shown in plate 6,
figures 1-6.

Infusion No. 3 was made by grinding 13 gm. of Panaeolus
venenosus to which sufficient water was added to make a filtrate
of 50 c.c.

Infusion No. 4 was made from 25 gm. of Agaricus campestris
treated as described for infusion No. 3. This infusion served as
a control.

No. 5 was an extract made by grinding 5 gm. of P. venenosus
which had been dried at room temperature and 45 c.c. of distilled
water. The filtrate was evaporated at 25° C. giving .5 gm. of a
brownish-yellow mass. This was mixed with 10 c.c. of distilled
water.

Extract No. 6 was made by rubbing up the residue on the filter
obtained in making extract No. 5, with absolute alcohol and filter-
ing. The filtrate was evaporated and the mass was shaken up with
25 c.c. of distilled water.

Infusion No. 7 was made like extract No. 5, but with the excep-
tion that the filtrate was not evaporated. There were 30 c.c. of
the filtrate, which was greenish yellow at first and gradually be-
came a darker greenish yellow.

Infusion No. 8 was made by thoroughly crushing 10 gm. of
fresh P. venenosus to which was added 40 c.c. of distilled water.
This was allowed to stand for 24 hours before filtering.

Infusion No. 9 was made by rubbing 25 gm. of mature P.
venenosus (pl. 7, figs. 9-15) with 25 c.c. of distilled water. The
pulpy mass was allowed to stand for 24 hours before filtering.
The filtrate was no further diluted but sufficient sodium chloride
was added to make the solution isotonic with the muscle tissue of
the turtle.

Infusion No. 10 was made of Panaeolus retirugis (iL. 7, figs.
16-19); 12 gm. of plants of this species were macerated with an
equal quantity of distilled water. Sufficient sodium chloride was
added to make the solution isotonic with turtle muscle tissue.
The mass was filtered 24 hours after maceration. The filtrate
was black in color and had the odor of crushed lawn grass {Poa
pratensis) .

Infusion No. 11 was made of a species of mushroom which re-

182 Semi-centennial of Torrey Botanical Club

sembled in all respects P. venenosus with the exception that it had
a comparatively short stipe and no marked tomentum at the base
of the stipe. This plant I shall describe later as a form of P.
venenosus. To 15 gm. of these plants was added 15 c.c. of dis-
tilled water. The filtrate was diluted to make 20 c.c. of fluid and
sufficient sodium chloride was added to make it isotonic with the
muscle tissue of the turtle.

Infusion No. 12 was made of mature specimens of P. venenosus.
20 gm. of fresh mushrooms were macerated with 30 c.c. of .7 per
cent saline solution. The liquor was permitted to stand for
twenty-four hours before filtering.

Infusion No. 13 was made of 10 gm. of P. venenosus macerated
with 10 c.c. of .8 per cent saline solution three days after the plants
were collected. The infusion was filtered twenty-four hours later.

The effects of such aqueous infusions of Panaeolus venenosus
on guinea pigs and rabbits were tested, first, by intraperitoneal
injections; second, by subcutaneous injections; and third, by
feeding the infusions or the fresh mushroom mixed with small
pieces of lettuce to the animals. In making the injections aseptic
methods were followed.

The effect of intraperitoneal injections on guinea pigs

AND rabbits

The results of the intraperitoneal injections are given in
table I. The quantity of infusion injected varied from 3^ to
2 per cent of the body weight of the animal. In some experiments
(see table) the infusion was used as made. In other cases i c.c.
of the infusion was mixed with 4 c.c. to 9 c.c. of physiological salt
solution, distilled water or tap water. The results obtained by
injecting the given quantities of Panaeolus venenosus infusion
invariably caused the death of the guinea pig in about twenty-four
hours. After the injection was made the animal usually was
restless and moved about in an apparently dazed condition. This
was followed by a period of lethargy, from which he never
recovered. During this period he generally assumed a char-
acteristic position. The head was retracted between the shoulders
and the eyes were closed. When the eyes were open the
animal usually appeared stupid and dazed. Noises made by

Remarks

7:05 P.M.,
gave prema-
ture birth to
3 young

Urine
discharged
before death
was appar-
ently normal

Autopsy

sSut

Hemorrhag-
ic area in ab-
dominal wall
in region of
injection; in-
testine dis-
tended, not
injured

Hemorrhag-
ic areas near
pyloric end
of stomach

Hemorrhag-
ic areas near
pyloric end
of stomach
and along en-
tire length of
small intes-
tine

No hemor-
rhages; vital
organs nor-
mal

Time of
death

ID " t-^ 0 d 00

Sense responses

qonox

Responds
to touch ;
startled

Responds
to touch;
startled

Sui

Not dis-
turbed by
noise

Not dis-
turbed by
noise

No re-
sponse to
noises

Eyes
closed

Respiration

Increased

Increased

Markedly
increased

Markedly
increased

General
behavior

Immediate
crouching;
head retract-
ed; lethar-
gic

Animal in
stupor,
moved rest-
lessly

Retracted
head;t stu-
pefied; spas-
modic mo-
tion of head.

3:50 P.M.,
posterior por-
tion of body
paralyzed

Same as no.
3; more
marked; pa-
ralysis ap-
peared early

Injection

UOIS

-njtii

M MM N

"D'O

UI i^JIJ

-uBn^

0 in rj. dv

0 M M N

■ fO <N

Animal *

0 00 0

•ui3 UI

d do d

Date

MM 1>
^ 0\ 0\ M
M *1 1

sn 0 dv

M ^ *^

M MM <S

o

M N CO

sSui

c3 Pi-:^

^ _ — ' w

i-, o

73 ^

> -1-1 X<

^ - ^ .2 Si . - ^,

^2"^ ^ > ^ §

1- (U

,9 G

Ph 2"

9.

Sui

§2

tc CI

a 2

W o

•y^ ^ a o

q >

03 O

03 w g3

S a E a
.2 2

a; ^ CO ^ 03 «
o3 03 cj (D .n ■-
n b :3 T3 s u

S 2 I § § 2

a o Si S Tj t!

03

a; -o

5 ^

03

O 03

N T3
03 S

UOIS

-njui

•O'D

ui Aiii
-UBn5

X3S

•luS UI

"2 3

§2

sSui

Si «

0 O a; ^ .S

1 S « § -S 'S ^ -

^ b ^

O -M

*J o

o .5

C 1)

03

CX CO

" S
o 2

^a

CO

S rO ° w u
w 9^ N C 3J ^

<D O

=3 o

w 7,

UOIS

-njui

Injection

•D'O UI

4 tap
water

9 water

o

r?

ro

7t

*
p

Young

1 Young

Old

Anil

•uiS ui

o

a

00

00 1

G.P.

G.P.

G.P.

Date

0\

,1.1917

i>

0\

cs

ro

No.

00

0\

0

w 5

Remarks

1 Autopsy

1 ■

sSui

Stomach
and intestines
inflated; whit-
ish-yellow
patches on
liver

No local
injury; in-
testine dis-
tended with
gas

M l-l

OS 0\

M M

<N dv

Time of
death

OS o\

M M

(N* 0\

IN

ro to

Sense responses

qonoj,

Respiration

Respiration
increased

Respiration
markedly
increased

General
behavior

Stupor;
head retract-
ed and in-
active

Effect im-
mediate ;
stupid; dazed.

6 P.M., un-
able to stand;
refuses food

Injection

UOIS

-njui

M

•3-D

ut Aju
-UBn^

fO IN 15

w

IM •<£

O O
ro O

N ro

Animal*

xag

Old
Young

•ui3 UI

lO o
Tt- Os

OS

O

Dale

ci «5

o
"A

M '-'

1

> -5

03 03
6 S

g o ^ S lo ^ .-J

.2 OS d

S w 03 S

0\ o

2^ C H fl M Cl

^' Si 2^1 2^1

.2 5s S ^ -S ^ I
-u M c; rt o3

o ^

A ...

^' . 1 ^ . ^3:2

5< 53 a 8 1^
,s o

OJ +->

6 .8%
saline

12 tap
water

SO

10
to

4 tap
water *

0

M

to

10

Ti

ro

10
ro

CO

0

Ti-
ro

Id
0

Young

Young

1

0
to

Os
ro
•0

so

0
o\
p<

Os
00
ro

Oh'

6

6

Ph'

d

d

G.P.t

OS
to

(N

Os

to

OS

pi
pj

2.12.1917

M

o\

pi

ci

M

N

ro

10

ft 00

a -

d ^

Levine: Two species of poisonous mushrooms 187

tapping the cage failed to stimulate. It was only when the animal
was caught that it showed signs of life and its efforts to resist were
very futile. In a number of cases, Nos. 3 and 4, table i, and
others, it appeared that soon after the injection the hind portions
of the body responded feebly when touched and there were spas-
modic contractions of the head and fore legs. In all cases after
an intraperitoneal injection was made the rate of respiration in-
creased considerably. On approaching death there was generally
a period of feeble respiratory movements, followed by rapid breath-
ing, which often appeared labored. In all cases where death oc-
curred an autopsy was performed, primarily to determine whether
or not the injection had caused mechanical injury to the intestines
or other abdominal organs. In no cases was there any trace of
such injury of the intestine or stomach or any other abdominal
organ. There was, however, a distension of the intestine due to
gases and the pyloric end of the stomach and intestine showed
hemorrhagic areas.

Injection of extract No. 6 proved to be harmless.

A number of control experiments (see table i) were performed
simultaneously with the Panaeolus tests. Tap water, .8 per cent
physiological salt solution, and infusions of Agaricus campestris
were used. As in the Panaeolus tests a quantity of liquid equal
to I to 2 per cent of the body weight of the animal was
injected. The animal shortly after the injection fed and appeared
normal in every respect and remained so. The animal in control
experiments No. 5, table i,was later injected (March i, 1917) with
4 c.c. of infusion No. 3. Immediately afterward he developed
all the symptoms oi Panaeolus intoxication and died within twenty-
four hours. An autopsy failed to show any abnormality other
than those described above.

The above experiments show clearly that intraperitoneal in-
jections of Panaeolus venenosus prove fatal to guinea pigs and
rabbits and that relatively small quantities bring about this result.
The symptoms in many respects are not unlike those caused by
other mushroom poisons as reported by Ford and others. The
extracts of dry material give the same results as the infusions made
by fresh plants; both are equally toxic.

188 Semi-centennial of Torrey Botanical Club

The effect of a subcutaneous injection of Panaeolus

venenosus

Different quantities of Panaeolus venenosus infusion were in-
jected subcutaneously into guinea pigs and rabbits. In all twelve
tests were made and table 2 gives the results of some of the more
interesting ones. The early symptoms noted for intraperitoneal
injections appeared in these experiments but the effects were not
so pronounced. There was the lethargic state, during which the
animal did not feed and responded very poorly to external stimuli.
These effects, however, wore off more or less quickly; the time of
recovery varying with the individual animals. The sensory re-
sponses became normal within 48 hours in the most poisoned
animals. In all injected animals there was evidence of continued
local irritation lasting a number of days after the injection. The
animal persistently licked the skin where the injection was given.
About a week after the injection, the skin in the region of the
wound broke down and a lesion was formed covering an area of
3 to 4 cm. This sloughing off of the skin invariably occurred when
a subcutaneous injection was given. The skin healed completely
three to four weeks later. No death occurred as a result of a
subcutaneous injection.

It appears from the experiments that, unlike the intraperitoneal
injections, the subcutaneous injection of Panaeolus venenosus in
the quantities employed does not produce fatal results. However,
temporary intoxication results and a sloughing off of the skin in
the region of injection is sure to follow in both guinea pigs and
rabbits. In this respect Panaeolus venenosus infusion behaves
like the extract of Clitocybe multiceps reported by Ford (1911).
Small quantities of the Panaeolus venenosus infusion are sufficient
so to affect the skin (see experiment No. 3 — table 2).

The effects of feeding Panaeolus venenosus to guinea

PIGS

Guinea pigs refuse the fresh pilei, but when these are crushed
and mixed with lettuce leaves they will eat small quantities.
When fed lightly with their customary food for twenty-four hours
previously they take the Panaeolus more readily. In this way it
was possible to feed them 5 gm. of the plants at one time. In-

Levine: Two species of poisonous mushrooms 189

0\ 73

3 - o

§ M 75 O

■si

^ On

Si ,

o

O -O lO 03 g .« .S 0\
S ^ o; ^ ^ g I

03 o

i> o o

Ss g s

IT) 6* O

o ^

>> CO

§1^

<D O

r dj w ^

o3 CO ^ r; c

(u o

■3 2 " » ^ 5

-^^ U U. CO

a S .2 J-

03 +j 4J O

^ CO o ^-H aj

>^ 03

.§ -2 • -

o ^
o

rGoj ir^dj rsi-i

- O CO ^

M 2; g M

^ 2 ^

a;

^ X! ^

D. 0) O

+ ^ -2

+ 'I

^ c

^ I"

190 Semi-centennial of Torrey Botanical Club

fusions of Panaeolus venenosus were also administered by way of
mouth by means of a medicine dropper. In all, four experiments
are recorded in table 3. The results show that these fungi fed to
guinea pigs will make the animal stupid and appear dazed but
recovery follows in an hour or two. Feeding of the infusion to
guinea pigs gave uniform results except in one case, No. 4, re-
corded in TABLE 3. In all, eight such experiments were made with
infusions Nos. 2 and 3 and extract No. 5. None, however,
caused death. The guinea pig in experiment No. 4 (table 3)
behaved after the feeding as if it had been injected intraperito-
neally and died five hours after feeding. The autopsy showed
hemorrhagic areas along the intestinal wall and at the pyloric
end of the stomach. It is possible that this animal died of pneu-
monia. It is quite likely that some of the fluid may have gone
into the lungs.

It appears from these experiments that feeding of Panaeolus
venenosus at least causes temporary intoxication with increased
respiration, but recovery follows within several hours. It has
been claimed by Ford (191 1) that herbivora are not susceptible to
mushroom toxins when given to them by way of the mouth.

The effect of an intravenous injection of an infusion of Panaeo-
lus venenosus was tried on rabbits. Small quantities, 2 c.c. of a
mixture of i c.c. of infusion No. 12 and 20 c.c. of .8 per cent phys-
iological salt solution, were injected into the marginal vein of the
right ear. The rate of respiration began to increase shortly after
the injection until the breathing became labored. The animal
became deeply intoxicated and its senses dulled; all food offered
after the injection was refused. Although he began to feed on
the following day, this intoxicated condition lasted for three
days, when the effect of the injection began to disappear. Nine
days later the animal appeared normal. Two further experiments
were made in which 2 c.c. of solution No. 13 were injected intra-
venously. Neither proved fatal, although severe intoxication
was induced, which lasted four to five days, during which time the
symptoms described above were observed. It is quite likely that
with larger doses of the infusion No. 13 the results would be
fatal.

Levine: Two species of poisonous mushrooms

191

Remarks

Animal
recovered
after i
hour

Recov-
ery fol-
lowed
after 2 H
hours

Recov-
ery fol-
lowed
after i
hour

Rate of
respira-
tion above
normal,
then ir-
regular

Autopsy

Finding

Hem-
orrhagic
areas
along in-
testinal
wall

Date

On

d

CI

Time of
death

5 hours

after

feeding

Sense responses

Touch

Must be
touched
to be
aroused

Must be
touched
to be
aroused

Must be
touched
to be
aroused

Hearing

Dull

Dull

Dull

Unaf-
fected by
noises

Sight

Eye§
closed

General
behavior

Stupefied

Stupefied

Stupefied

Animal
appeared
sick after
feeding;
inactive ;
paralysis
of hind
legs ,

Food

Quantity

Amount
of P. ven-
enosus
eaten, i
gm.

5gm.

4C.C.
4C.C.

Kind

Young
P. veneno-
sus and
lettuce
leaves
chopped

Old and
young P.
venenosus
and let-
tuce
leaves

Old and
young P.
venenosus
and let-
tuce
leaves

*Infu-
sion no.
2 + tap
water

Animal

bex

^

Age

Young
Young
Young
Old

^ be

0 0 t~- 10
m ^ 0

Kind

PU Pii Ah

d d d • d

Date

On OS 0\ 0\

M M H M

On M w On

CN M M M

O 1 H

192 Semi-centennial of Torrey Botanical Club

The effect of Panaeolus infusion on muscle
The effect of an infusion of Panaeolus venenosus and of P.
retirugis was tested on the gastrocnemius muscles of the frog.
Animals weighing 30-45 gm. were used in these experiments.
The muscles were excised so that the skin covering was left
around the muscle to prevent drying out and at the same time
to form a sac into which the fluids, the effects of which were to
be studied, could be placed. The method is as follows. The skin
around the tendon of Achilles was cut. It was then rolled back
over the gastrocnemius muscle to the knee, the tendon was severed,
the muscle was carefully lifted away from the tibia and the latter
was cut off near the knee. The skin was then redrawn over the
muscle and the cut end of the skin was tied to the tendon so that a
considerable part of it protruded below the skin. The femur was
cut off near the hip, leaving the gastrocnemius attached to it.
The protruding tendon was used as a point of attachment for
one electrode and the femur for the other. Both gastroc-
nemius muscles were similarly excised. One skin sac was filled
with .6 per cent physiological saline which served as a con-
trol, the other was filled with the mushroom infusion. The
muscles were then suspended in moist-air chambers as an extra
precaution and were attached to writing levers. The muscles
were simultaneously stimulated electrically at intervals of two
seconds by single induction shocks and the contractions were
recorded on a slowly moving drum. The experiments were con-
tinued until the muscles were completely exhausted. Calculations
were then made of the duration of the power to contract, the total
number of contractions, and the total amount of work that was
performed. The right and left muscles were used alternately for
the Panaeolus infusion.

Ten experiments were performed, from which it appears clearly
that Panaeolus infusion hastens fatigue and reduces the working
capacity of the muscle. The total number of contractions of
which the muscle was capable was reduced by 51 per cent
and the total amount of work by 55 per cent. The length of
time the treated muscle was capable of working was reduced by
51 percent. The length of time, the total number of contractions,
and the amount of work done by the normal muscles were each rep-

Levine: Two species of poisonous mushrooms 193

resented by lOO per cent. A case of extreme difference between
the normal muscle and the one injected with Panaeolus infusion
when compared with respect to the length of time the muscles
were able to work before fatigue set in is shown in experiment 2
(table 4). Here the normal muscle worked for a period of 28.9

TABLE 4

Action of Panaeolus infusions on muscle
Normal Panaeolus extract '

1-2 c.c. of .6 per cent NaCl injected into 1-2 c.c. of various Panaeolus infusions
skin sac around gastrocnemius muscle. injected into skin sac around gastro-
cnemius muvscle.

Dur. of

Total no.

Amount of

Dur. of

Total no.

Amount of

Expt.

Muscle

work,

of con-

work done.

Infus.

work.

of con-

work done,

min.

tractions

in gm.m.

min.

tractions

in gm.m.

I

R.

30.15

689.83

15-67

9

9.21

210.72

2.09

2

L.

28.9

610.08

14.22

9

5-31

112.09

.605

3

R.

29.68

695-59

17-78

9

12.81

300.26

5-28

4

L.

47-5

II34-3

25.41

10

28.9

690.13

18.91

5

R.

35-01

828.3

17-52

II

18.75

443-6

9-58

6

L.

47-18

mo. 71

18.70

II

17.96

419.0

8.078

7

R.

24-53

569-59

14.878

12

11.09

257-5

3-17

8

L.

27.9

681.59

15.81

12

22.18

541-85

11-85

9

R.

32.9

745-5

18.73

12

15-15

343-29

7-63

10

L.

30.7

719-3

10.645

12

22.6

528.8

10.44

Average

32.445

778.479

16.9363

16.396

384-724

7-7633

Per cent

100%

100%

100%

49%

49%

45%

minutes, while the injected gastrocnemius worked only 5.31
minutes. The total number of contractions was greater for the
normal muscles than the injected ones and the greatest difference
between them was observed in experiment No. 6 (table 4). In
all cases, without an exception, the amount of work done by the
normal muscle was greater. These experiments show conclusively
that the infusion of Panaeolus venenosus has a harmful effect on
the working power of a muscle.

The effects of Panaeolus infusion on blood pressure and

PULSE

Infusions of Panaeolus venenosus and P. retirugis were injected
into the venous circulation of a cat to determine the effect of the
infusion of these fungi on the blood pressure and pulse. A large
female cat (weight 1,550 gm.) was used. She was anaesthetized

194 Semi-centennial of Torrey Botanical Club

with ether and after tracheotomy had been performed a mercury
manometer was connected with the carotid artery. After the
blood pressure was recorded on a kymograph the right femoral vein
was exposed and i c.c. of infusion No. 9 (P. venenosus) diluted with
4 c.c. of .8 per cent saline solution was injected with a hypodermic
needle. The fluid was allowed to flow into the vein by force of
gravity. Shortly after the injection there was a slight increase
of the blood pressure approximating 2 mm. which lasted three
minutes and then became normal. The number of heart-beats
prior to the injection was 131 per minute, and in less than a minute
after the injection the heart-beat increased to 165. Fifteen minutes
later 5 c.c. of infusion No. 11 (P. venenosus) was injected into the
same vein in the same way as described above. The injection
was followed by a gradual increase until a pressure of 2 mm. above
normal was recorded. The pulse rate remained unchanged.
Twenty-two minutes later (3:00 P.M.) 5 c.c. of infusion No. 10
(P. retirugis) was injected into the left femoral vein and in three
seconds the blood pressure dropped gradually 14.5 mm. below the
normal and then slowly increased to 11 mm. above the normal,
from which point it slowly diminished until a point below the
normal was attained from which there were but slight deviations.
The pulse rate immediately before the injection was 144 per minute
and after the injection increased to 165-170. At 3:09 P.M. an-
other injection of 5 c.c. of infusion No. 10 (P. retirugis) was given
through the same vein. This time there was only a decrease in
the blood pressure and as in the preceding cases this was followed
by an increase which was very slight in proportion to the decrease.
The return to normal required a little over 10 minutes. This
experiment was followed at 3 :17 P.M. by an injection of 3 c.c. of
infusion No. 11 {P. venenosus), and a record was obtained similar
to that of the preceding case. The pulse rate was 169 per minute.
Six minutes later an injection of 3 c.c. of .8 per cent physiological
salt solution was introduced through the same vein. No change
in the blood pressure or in the pulse rate occurred. After the first
injection it was noted that there was a gradual decline in the
blood pressure. The rate of respiration was not increased and a
uniform count of 19 per minute was made. These results do not
agree with those described above for rabbit and guinea pigs.

Levine: Two species of poisonous mushrooms 195

It is quite evident that infusions of P. venenosus and P. retirugis
have a distinct effect on the blood pressure of a cat. Whether
the decrease and increase is due to locaHzed or general dilatation
or constriction of the blood vessels, I did not attempt to deter-
mine. It is evident, however, that the changes in the blood pres-
sure are not due to the heart-beat. As far as my experiments go,
they tend to show that the heart-beat is accelerated by Panaeolus
infusion in small quantities and this effect is quite lasting. The
acceleration may be due to the paralysis of the vagus termina-
tions, as later experiments tend to show. As noted below, applica-
tions of Panaeolus infusions to exposed hearts of frogs and turtles
show no effect. In this respect the infusions I have used are like
cocaine (Mosso, 1887, 1890) and nicotine, coniine, and piperidine
(Moore and Row, 1898).

The effect of Panaeolus infusion on motor nerves
In these experiments the sciatic nerve and the gastrocnemius
muscle of frogs weighing 45-50 gm. were used. The muscle was
excised in the manner described above so that the skin was left as
a covering about it. The muscle was then attached to a writing
lever and the sciatic nerve was stretched over two pairs of elec-
trodes thoroughly protected from currents of air. Single induc-
tion shocks were used to stimulate the nerve. After a number of
normal contractions of the gastrocnemius muscle were recorded by
stimulating the nerve, infusion No. 13 (P. venenosus) was applied
to the central end of the nerve by means of a camel's-hair brush.
Immediately after this the nerve was stimulated and normal con-
tractions of the gastrocnemius muscle were obtained. After 5 to
10 applications of the extract to the nerve in a period of 3 to 5
minutes, stimulating by the make shock failed to produce a
contraction of the muscle and the effects of the break shock
became less and finally disappeared. On stimulating the nerve
through the peripheral electrodes normal contractions were ob-
tained by both make and break shocks. Moving the central
electrode successively from its original place to 3^, and ^ of
the distance between it and the peripheral electrode and stimu-
lating the curve at each point gave in each case normal contrac-
tions of the muscle, while stimulating the nerve centrally to the
first central point gave no reaction.

196 Semi-centennial of Torrey Botanical Club

Applying Panaeolus infusion successively at 4 or 5 points
between the original central and peripheral electrodes showed that
the infusion had to be applied more often and for a longer period
before electrical stimulation of the nerve failed to produce a
muscular response and the infusion had very little effect when
applied to the nerve 5 to 8 mm. from the muscle. These observa-
tions were repeated many times and in all cases application of P.
venenosus infusion to the cut end was more effective than on the
muscle end of the nerve. A .3 per cent isotonic solution of
cocaine applied to the sciatic nerve and gastrocnemius muscle prep-
aration gave me in the main similar results to that obtained with
Panaeolus infusion. The effects of cocaine observed here are like
the results obtained by Mosso (1890) in his work on the effect of
cocaine on motor nerves, although Mosso's method of experimenta-
tion was different from mine. The interesting fact is that Panaeolus
infusions showed the local paralyzing effects so well known for
cocaine.

The effect of Panaeolus infusions on the heart and
vagus nerves

The increased rate of heart-beat noted in injection of Panaeolus
infusions into the circulation of the cat led to the investigation of
the effect of this infusion on the heart and vagus nerves of the frog
and the turtle. Preliminary experiments were made on frogs.
After pithing the animals the two vagus nerves were exposed, and
electrodes were attached. Each nerve was tested for its power
to inhibit the heart-action. The apex of the heart was attached
to a writing lever and the contractions were recorded on a slowly
revolving drum. After normal and complete inhibition on stimu-
lating each nerve had been recorded, a Panaeolus infusion was
painted on the right vagus with a camel's-hair brush at a point
between the electrodes and the heart. No effects on the heart
were noticed and after 30 seconds the vagus was stimulated with
the electric current. The heart-beat instead of stopping became
slower than usual and the contraction became greater. After
three to four minutes electrical stimulation produced no effect
even on the rate of the heart-beat (see graphs Nos. i and 2).
Later the effects of the Panaeolus infusion slowly disappeared and

198 Semi-centennial of Torrey Botanical Club

the inhibiting power of the nerve was regained. When the
Panaeolus infusion was painted directly on the heart the contrac-
tions were in no way affected but the inhibiting function of both
vagi was abolished. If the electrodes were then applied to the
sinus there were the usual accelerated contractions followed by
inhibition. Repeated application of the Panaeolus infusion to the
heart of the turtle followed by stimulation of the sinus gave
stronger contractions than the normal followed by inhibition.
The effects of infusions Nos. 9, 10, and 12 were studied on turtles.
With each infusion similar experiments were performed. It was
found, however, that infusion No. 9 abolished the inhibiting action
of the vagus more quickly than No. 12, and that No. 10 (P. reti-
rugis) was more effective than Nos. 9 or 12. It was further ob-
served that the complete failure of the vagus nerve to cause inhi-
bition differed with the different turtles used, although the in-
fusions were applied in the same manner and with equal frequency.
The rapidity of recovery of the vagus nerves from the Panaeolus
infusions to a great extent depended upon the animal.

Tests were made to determine the extent of the influence the
infusion had on the nerve. The normal heart inhibitions were
recorded on stimulating the vagi; then, as in the case of the frog,
one of the nerves was painted with Panaeolus infusion at a point
between the electrodes and the heart. The nerve was stimulated
electrically at intervals of one minute. When stimulation failed
to produce inhibition the electrodes were moved to a point between
the heart and the point of application of the infusion. In all cases
stimulating at the new point caused inhibition. From such ex-
periments, I am of the opinion that the Panaeolus infusion acts
only locally. Painting the heart of the turtle with infusion abol-
ished the vagus action.

The similarity in the behavior of Panaeolus infusion to cocaine
made it desirable to compare the two substances on the vagus of
the same animal. After recording heart inhibitions from the
right and the left vagus the former were painted with infusion
No. 9, and the latter with .3 per cent cocaine. The results ob-
tained were identical. With cocaine, however, the effect was more
rapid and the recovery more slow than for Panaeolus infusion,
although it must be borne in mind that the amount of the active

Levine: Two species of poisonous mushrooms 199

element in my infusion is entirely unknown. While no chemical
study of the active principle of Panaeolus venenosus or P. retirugis
has been made, it appears from these experiments that they con-
tain some substance which has a temporary and local anaesthetic
effect on nerve tissue, especially the vagus nerves, and in this
respect is like cocaine. Furthermore, these experiments show that
P. venenosus and P. retirugis contain a substance or substances
with similar physiological and toxicological properties. The most
interesting fact about the infusions from these species is that they
differ entirely from the extracts made of other fungi so far investi-
gated. According to Ford and Clark and their collaborators,
Amanita muscaria, the poisonous Russulas, Clitocybes, Inocybes,
Boleti and other species of poisonous mushrooms contain a toxic
substance which physiologically acts like muscarine, i. e., when the
extracts of these mushrooms are applied to the vagus nerve, the
nerve-endings are stimulated and cause in turn the inhibition of
heart-action. The infusions of Panaeolus species studied here
show no effect on the hearts of the frogs or turtles except to anaes-
thetize the part of the nerve to which they are applied.

Summary

1. No attempt was made to isolate the toxic substances in
Panaeolus venenosus and P. retirugis. Infusions of the mushrooms
were used in testing their toxic properties.

2. Infusions made by macerating P. venenosus with an equal
weight of water produced death of guinea pigs and rabbits when
3^ to 2 per cent of their body weight of the infusion was injected
intraperitoneally.

3. Subcutaneous injections of these infusions into guinea pigs
and rabbits caused milder symptoms than those observed in
animals injected intraperitoneally; these injections did not cause
death, but without exception caused the sloughing off of the skin
in the region of the injection.

4. Intravenous injections of relatively small quantities of in-
fusion into the circulatory system of the rabbit caused symptoms
similar to those observed when the animal was injected intra-
peritoneally, but no death was recorded and the animals recovered
completely within a few hours.

200 Semi-centennial of Torrey Botanical Club

5. Infusions injected into the circulatory system of the anaes-
thetized cat produced a decrease in blood pressure followed by an
increase in blood pressure, from which condition recovery followed.
The rate of the heart-beat was increased.

6. Immersion of the gastrocnemius muscle of the frog in infu-
sions caused the muscle to fatigue sooner than the normal muscle,
and its working capacity was markedly reduced.

7. Application of the infusions to the cut end of the sciatic nerve
of the frog affected the nerve so that its electrical stimulation
failed to produce a contraction of the gastrocnemius muscle. In
this respect the infusion acts like cocaine.

8. Application of the infusions to the vagus nerves of the frog
and turtle affected the nerves at the point of application so that
electrical stimulation at that point failed to cause heart inhibition.
Painting the heart with infusion appeared to have no effect on the
heart-beat but paralyzed the vagus-nerve terminations. The
action of the infusion in this differs from the extracts of all other
fungi so far investigated.

To Professor Frederic S. Lee and Professor R. A. Harper I
wish to extend my thanks for many helpful suggestions and criti-
cisms.

Literature cited

Boehm, R., & Kiilz, E. 1885. Ueber den giftigen Bestandtheil der
essenbaren Morchel {Helvetia esculenta). Arch. Exp. Path. u.
Pharm. 19: 403-414.

Clark, E. D., & Kantor, I. L. 1911. Toxicological experiments with
some of the higher fungi. Mycologia 3: 175-188./. i. pi. 52.

Clark, E. D., & Smith, C. S. 1913. Toxicological studies on mush-
rooms, Clitocybe illudens and Inocybe infida. Mycologia 5 : 224-
232. pi. 91.

Collins, F. S. 1899. A case of ^c»/e/2i5 poisoning. Rhodora i : 21-23.
Ferry, R. 191 1. Etude sur les amanites — Les amanites mortelles.

Suppl.Rev. Mycol. 1911.
Ford, W. W. 1906. The toxicological constitution of Amanita phal-

loides. Jour. Exp. Med. 8: 437-450.
Ford, W. W. 1907. A clinical study of mushroom intoxication.

Johns Hopkins Hosp. Bull. 18: 123-130.
Ford, W. W. 1908. The pathology oi Amanita phatloides intoxicsition.

Jour. Infect. Dis. 5: 116-132.

Mem. Torrey Club

Volume 17, plate i

1-8. PANAEOLUS \'ENENOSUS IMurrill
16-19. PANAEOLUS RETIRUGIS Fr.

Mem. Torrey Club

9-15. PANAEOLUS XENHNOSUS Murrill

Levine: Two species of poisonous mushrooms 201

Ford, W. W. 1909. The distribution of poisons in the Amanitas.

Jour. Pharm. and Exp. Ther. i: 275-287.
Ford, W. W. 1911. The distribution of haemolysins, agglutinins and

poisons in fungi, especially the Amanitas, the Entolomas, the

Lactarius, and the Inocybes. Jour. Pharm. and Exp. Ther. 2:

285-318.

Ford, W. W., & Clark, E. D. 1914. A consideration of the properties

of poisonous fungi. Mycologia 6: 167-191.
Ford, W. W., & Sherrick, J. L. 1911. On the properties of several

species of the Polyporaceae and of a new variety of Clitocybe,

Clitocyhe dealbata sudorifica Peck. Jour. Pharm. and Exp.

Ther. 2: 549-558-

Ford, W. W., & Sherrick, J. L. 19 13. Further observations on fungi,
particularly Clitocybe sudorifica Peck, Pholiota autumnalis
Peck and Inocyhe decipiens Bresad. Jour. Pharm. and Exp.
Ther. 4: 321-332.

Gillot, V. 1910. Etude medicale sur I'empoisonnement par les

champignons. Lyon.
Moore, B., & Row, R. 1898. A comparison of the physiological

actions and chemical constitution of piperidine, coniine and

nicotine. Jour. Physiol. 22: 273-295.
Murrill, W. A. 1909. A new poisonous mushroom. Mycologia i:

211-214.

Murrill, W. A. 1916. A very dangerous mushroom. Mycologia 8:
186-187.

Mosso, U. 1887. Ueber die physiologische Wirkung des Cocains.

Arch. Exp. Path. u. Phar. 23: 153-208.
Mosso, U. 1890. Ueber die physiologische Wirkung des Cocains.

Arch. Physiol. 47: 553-601.
Paulet, J. J. 1793. Traite des champignons. Paris.
Peck, C. H. 1911. Report of state botanist, 1910. Bull. N. Y.

State Mus. 150: 43.
Sartory, A. 1914. Les champignons veneneux. Nancy.

explanation of plates I AND 2
Plate i

Figs. 1-8. Panaeolus venenosus Murrill.
Figs. 16-19. Panaeolus relirugis Fr,

Plate 2

Figs. 9-15. Panaeolus venenosus Murrill.

LIFE-HISTORY STUDIES IN SCLEROTINIA

By F. J. Seaver and W. T. Horne

New York Botanical Garden University of California

(with plate 3)

For several years past a species of Sclerotinia has been observed
by the writer in a certain stretch of woods in the upper end of Van
Cortlandt Park, New York City, on the rootstocks of wild gera-
nium. Although this has been seen abundantly in this particular
region, it has not been detected by us in other localities where the
wild geranium grows. The apothecia usually appear early in the
spring about the latter part of April or early in May and disappear
early in June. A search of the records showed no species of Sclero-
tinia listed for this host, so that the writer was uncertain whether
this represented an undescribed species or some old species on a
new host. It did not appear, however, to agree well with any
described species, and it was finally decided to publish it in order
to bring it to the attention of mycologists. Before doing this it
was thought advisable to locate the conidial stage, if possible, in.
order to make the description more complete, and during the spring
of 191 7 this work was undertaken.

From our knowledge of other species of Sclerotinia, it was*
thought that the conidial stage might be located as a parasite on
the leaves or other living tissue of the host. A careful search of
the region in which the fungus occurred on various occasions
showed nothing on the living plants which could be suspected of
being the conidial stage of this fungus. A collection of infected
rootstocks, however, which had been brought into the laboratory
and placed in a moist chamber after a few days showed a most
luxuriant growth of a species of Botrytis. This appeared in dense
tufts not only on the rootstocks and rootlets, but even covered
ever the outside of the apothecia like a coat of fur. While none of
the Botrytis was apparent on the plants when brought into the
laboratory, it invariably appeared within a few days on the roots

202

Seaver and Horne: Studies in Sclerotinia 203

and rootstocks of those plants infected with Sclerotinia^ while
similar rootstocks from regions where the Sclerotinia was absent
failed in every case to develop this type of Botrytis, The conidio-
phores often appeared in dense tufts, these often springing from
minute black sclerotium-like bodies, although the latter were not
always evident. Thinking that this fungus might be an omniv-
orous saprophyte, the rootstocks of other kinds of plants from
the same region were placed in moist chambers, but failed to pro-
duce this fungus. From these rough observations it was suspected
that the Botrytis might have some connection with the Sclerotinia.

It was noted that the ascospores were always in excellent ger-
minating condition when brought into the laboratory, and it was
decided to attempt to culture out the fungus. One of us (Horne),
who happened to be working at the New York Botanical Garden
at this time, kindly offered to culture the fungus and the following
experiments were conducted by him.

Culture experiments

Crude cultures were made by touching the tufts of Botrytis
spores, as shown in the accompanying plate, on the small roots of
wild geranium with a sterile needle and then bringing the needle
into contact with an autoclaved potato plug placed in a test tube.
A vigorous fungus grew promptly, developing in somewhat the
same way as Botrytis vulgaris, but readily distinguished from that
species on detailed examination. A few days later a crude culture
was made by touching the top of an apothecium of the Sclerotinia
of wild geranium with a sterile needle and with this inoculating a
drop of sterile water on a sterilized slide. The Sclerotinia spores
were abundant and no Botrytis spores were observed in the drop
on examination with the i6 mm. objective. A transfer was made
from this drop to a potato plug, as with the Botrytis inoculation.
After somewhat more than one week, these cultures were examined
and both were producing the characteristic Botrytis spores and
microconidia.

On May 20, pure cultures of the Sclerotinia of wild geranium
were made as follows: a dried herbarium specimen of the Sclerotinia
collected in Van Cortlandt Park during the present season was
moistened by placing a small drop of sterilized water on the upper

204 Semi-centennial of Torrey Botanical Club

surface, using a loop needle and sometimes slightly rubbing the
surface of the hymenium with the loop. Some of the material
was then transferred to a drop of sterile water on a sterilized slide.
Poured plate cultures were made in the usual way, being inocu-
lated directly from the drop prepared on the slide, using prune-
juice agar and Shear's cornmeal agar. Before making the poured
plates, the drop used for inoculation was carefully inspected with
the 1 6 mm. objective and no Botrytis spores were seen in it, in fact
no spores were observed except those of the Sclerotinia.

After about twenty hours the agar plates were inverted and a
number of germinating spores were marked. Only the colonies
of mycelium clearly arising from one spore and well separated
were marked. While the original spore had become considerably
swollen and not recognizable with absolute certainty in some cases,
it appeared from the figure of the mycelium that the Sclerotinia
spore had given rise to the growth in every case marked. Later
in the same day five of these colonies were transferred to slant
tubes of prune agar. The following day six more plantings were
made from separate colonies growing from marked single asco-
pores. By this time the colonies had become very complex and
were plainly visible. Three days from the making of the poured
plates the colonies were confluent and vigorous and the charac-
teristic Botrytis spores had commenced to be formed. Apparently
all of the colonies originating from the ascospores gave rise to the
Botrytis spores and there were no contaminations, all of the colonies
being of the same sort.

Of the single-spore transfers to prune-juice agar, three of
those made on the first day failed to grow, presumably the young
mycelium had been caught on the needle used in the transfer,
since some had not been found on the slant after making the trans-
fer. The remaining eight single-spore cultures developed very
uniformly and all produced abundant Botrytis spores.

On June 24, plantings were made from each of the eight pure
cultures on prune-agar slants to sterilized geranium rootstocks and
to sterilized potato plugs. On the potato plugs the growth was
identical with that originally secured from crude plantings of the
spores of Botrytis and Sclerotinia. After four days, Botrytis
spores could be seen with a lens in nearly all of the cultures and

Seaver and Horne: Studies in Sclerotinia 205

yellowish sclerotia were beginning to form on some of the geranium
rootstocks, but no distinct sclerotia were observed on any of these
or older potato cultures. Although Botrytis spores appeared in all
of these, they were much more abundant on the rootstocks than
on the potato plugs. On the rootstocks the spores were so abun-
dant as to be evident to the naked eye, and the masses were very
similar to those obtained on infected rootstocks brought from the
field and placed in moist chambers, as shown in the accompanying
plate. Check experiments were kept and these failed to show any
evidence of the Botrytis.

The infected rootstocks will be kept with the hope of securing
mature apothecia, but these will probably not appear until spring,
as is the case in nature, and it is too soon to predict what the result
of this study will be. However, the production of Botrytis
directly from the ascospores of the Sclerotinia confirms field obser-
vations on the connection of the two fungi. The production of
apothecia would add still more interest to the investigation.

Sclerotinia (Stromatinia) Geranii sp. nov.

Conidial stage (Botrytis) occurring on the roots and rootlets of
the host, being especially abundant when left in moist chamber
for a few days and even developing on the outside of the apothecia,
usually appearing in tufts and often springing from minute sclero-
tium-like bodies, although the latter are not always present, dark
brown in mass at maturity; conidiophores reaching a length of i
mm. or more and a diameter of 10-15 yu, pale brown, sparingly
septate and branched, the conidia borne in rather large masses
like bunches of grapes; conidia subglobose or pyriform, the small
end representing the point of attachment, reaching a diameter of
10 M or rarely as large as 12 /z, slightly longer than broad, at first
smooth, becoming quite strongly roughened, pale brown with
transmitted light.

Apothecia springing from the partially decayed rootstocks in
clusters of variable numbers, stipitate, shallow-cupshaped or
subdiscoid, reaching a diameter of i cm. or rarely larger, pale-
brown externally; hymenium concave or nearly plane, a little
darker than the outside of the apothecium; stem reaching a diame-
ter of 2 mm. and often reaching a length of several cm., though
often short and occasionally almost wanting, the length varying
with the depth to which the rootstocks of the host are buried; asci
cylindric or subcylindric, 8-spored, reaching a length of 120-140^

206

Semi-centennial of Torrey Botanical Club

and a diameter of 8-10 ^t; ascospores hyaline, ellipsoid or almond-
shaped, 4-5 )u X 12 ju, usually containing two very small oil-drops.
[Plate 3.]

On the rootstocks of wild geranium {Geranium maculatum).
The type collected in woods in the upper end of Van Cortlandt
Park, New York City, May, 191 7.

The subgenus Sttomatinia has been raised to the rank of a
genus by Boudier, although it is not commonly regarded as such.
If the genus Sttomatinia is considered distinct from Sclerotinia
our plant would be designated as Stromatinia Geranii.

description of plate 3

Fig. I. Photographs of the Botrytis stage on the underground parts of wild
geranium, natural size, with drawings of sporophores and conidia enlarged.

Fig. 2. Photograph of apothecia, natural size, with drawing of ascus with spores
and germinating ascospores, enlarged.

All drawings made with the aid of the camera lucida to a common scale using a
one-inch eye-piece and a one-sixth objective.

Mem. Torrey Club

VOLUI\-E 17, PLATE 3

SEAVER AND HORNE: STUDIES IN SCLEROTINIA

THE INDIVIDUALITY OF THE BEAN POD
AS COMPARED WITH THAT OF
THE BEAN PLANT

By Helene M. Boas

The New York Botanical Garden

This paper is a preliminary report on some experiments bearing
on the question to what extent parts of a plant, apparently alike,
may possess an individuality similar to that of individual plants.

It is a well-known fact that organs that are homologous may
be differentiated from one another. For example, the first pair of
leaves of the bean are, as a rule, simple, those following are com-
pound. The foliage leaves of reproductive parts are commonly
different in shape, size, and other characters from those of vege-
tative parts. In the composites the outer flowers of a head may be
of quite a different type from the inner and often the fruits pro-
duced by them are morphologically distinct.

Many instances could be cited in which lateral organs have been
found to differ from terminal organs in the number or shape of
their parts. It will be sufficient here to mention the fact that the
number of flowers in the inflorescences of Compositae and Umbel-
liferae, and the number of stamens and carpels and other floral
parts in other forms, have been found to exhibit differences ac-
cording to their position on the plant. Flowers have been found
to differ in size, whether produced early or late in the season; as,
for example, in tobacco, and in the composites it is a common
observation that flower number of the heads decreases with the
advance of the season.

Often, however, the differentiations are not of sufficient mag-
nitude to be readily detected and we are then inclined to speak of
variations. In such instances a detailed study frequently shows
organs apparently alike to be different. The differences may be
slight, but regular and significant. In other words, homologous
parts may show individuality, comparable to the individuality
shown by entire plants.

207

208 Semi-centennial of Torrey Botanical Club

In the winter of 191 5 an experiment bearing on this point was
planned and the data recorded are from thirty-eight plants of the
Longfellow variety of string-bean grown in the summer of 1916.
The plants had 736 pods and 3,462 beans. The individuality of
the bean pod and of the bean plant chosen for expression is the
shape of the beans. An index, the thickness of the bean divided
by the width measured from the hilum to the opposite side, has
been used not because any particular importance is to be attached
to this measure, but because it was convenient. An index of this
sort is preferable to weight or to a single measurement as thickness
or length because purely physiological differences due to nutrition
are more probably eliminated. It was observed that the beans
varied considerably in form, some being flat and others much
thicker. This is expressed in the index, which varies from about
70 to about no.

If all the pods on a plant were alike and the variations among
the individual beans of a plant purely chance variations, the
average index of each pod, assuming each pod to contain a large
number of beans, would be the same as the average index for the
plant as a whole. There would then be no variation among the
pods or, in other words, they would all be alike. If, however,
there are differences among the averages for the pods, the pods
show individuality.

We have first to consider in how far the plants are alike or dif-
ferent. As we might expect, the average index is not the same for
all plants. It ranges from 86 to 98, the average being 94 with a
standard deviation of zt 3.4. In other words, the plants are not all
alike, but have individuality, some with thicker and others with
flatter beans.

We may treat the separate pods on a plant in the same way.
The fact that the pods contain a small number of beans and not all
pods the same number must be taken into account in the calcula-
tion, in order that the results obtained may be comparable. If
the variability of the individual beans on a plant is expressed by
(7, the correlation between the beans in a pod by and the number
of beans in a pod by n; then s, the observed variability of the beans
in a pod, is expressed by the following formula:

Boas: Individuality of the bean pod

209

n — I

= (i — r)

n

From this calculation the amount of correlation of the beans within
a pod (r) has been determined to be + 0.29,

It appears from this that the pods exhibit individuality just as
the plants do. The individuality of the pods is of almost the same
order as that of the plants, the average standard deviation for the
beans within a pod being =t 4.3 and for the individual plants ± 34.

These studies do not show what the individuality of the pods
is due to, whether to purely chance physiological differences as
differences in nutrition or to definite morphological differentiation
according to position of the pod on the plant. As has been men-
tioned before, there are many instances in which differences be-
tween homologous organs have been correlated with morphological
position or time of development. I believe that the morphological
position of the pod bears some relation to its individuality. I
cannot now present quantitative results, but the material studied
so far shows that there are slight differences in the form of the
beans of the lowest and uppermost branches.

A further question to be investigated is whether such mor-
phological differences as discussed in this paper may be inherited.
As far as I know, there has been no work that would either prove
or disprove the inheritance of such differences. Correns made
some attempt to grow offspring from ray- and disk-flowers of
Dimorphotheca pluvialis, but his material was meager and the
results inconclusive. In pure-line breeding small individual dif-
ferences between plants have been selected out and have main-
tained themselves for successive generations, and it is at least to
be tested whether similar pure lines may not be obtained from a
single individual.

I have material at hand, and I hope at some later time to be
able to report on these questions.

THE EVOLUTION OF CELL TYPES AND CON-
TACT AND PRESSURE RESPONSES
IN PEDIASTRUM

By R. a. Harper

Columbia University

I desire to present at this time an outline of the evolutionary
development in the genus Pediastrum, noting especially also the
relation of the forms of the various cell types which characterize
the species to the intercellular biogenetic reactions through which
the colonies get their characteristic configurations. We shall find
in the subgenera a series of well-marked groups, in each of which
the particular initial primitive cell form foreshadows all the types
which have appeared in the evolution of this particular series.
The differences which characterize the species in many cases
pass over into each other by very finely graded variations, so
that there has been the greatest possible confusion and uncer-
tainty among systematists as to whether certain types should be
considered species, varieties, or mere form races. An orthogenetic
trend of development can be recognized in the Diactinia and
Tetractinia especially and the forms illustrate such series of con-
tinuous variants as Jennings ('i6) has produced by selection in
Difflugia corona.

The genus, as a whole, on the other hand, presents a series of
groups, the subgenera, which just as plainly differ by discontinuous
characters. In most cases it is necessary to assume a return to
the primitive undifferentiated cell type of the simplest species
in order to conceive the method of origin of the subgenera.

An orthogenetic trend for the whole genus can perhaps be
recognized in the simple tendency to develop spinous projections
on the body of the cell, but we need to know more of the relation
of such changes of form to the ultimate constitution of the cell
before we can be sure that there is any common background for
the tendency which has led to the formation of the one-spined,

210

Harper: Cell types and responses in Pediastrum 211

two-spined and more or less deeply lobed, three-spined, and four-
spined forms which are recognized as Monactinia, Diactinia,
Triactinia, and Tetractinia respectively. It is not obvious on the
basis of our present knowledge why an ancestral type which had
developed a unispinous form is more likely to have descendants
with two- or three- and four-spined cells than with long cylindrical
cells like Hydrodictyon or spindle-shaped cells like Scenedesmtis,

In most cases again in the subgenera we are confronted with a
series of continuous variations in the configuration of the colonies,
which, however, is broken at what appear to be critical points at
which a further modification of the form of the cell leads to a
quite characteristic change in the symmetry of the whole colony.
In some cases, cell forms which have apparently tended to a large
degree of asymmetry in the colony, when modified to a certain de-
gree, make the achievement of equal contact and pressure relations
and a higher degree of symmetry possible, as in the transition from
Pediastrum simplex to P. triangulum and from P. Ehrenhergii to P.
Rotula. In other cases a more extreme development of a particular
cell form may make a new configuration of the colony necessary
with a symmetry much more difficult to achieve, as in the transi-
tion from P. asperum to P. clathratum. Throughout, as I have
already pointed out ('15), we have the conflict of these ortho-
genetic tendencies in the evolution of the cell form and the law of
cell reproduction by bipartition, giving the geometrically pro-
gressing series of cell numbers, 2, 4, 8, 16, 32, etc., with the prin-
ciple of least surfaces requiring an entirely different series of
numbers, i, 7, 19, 37, 61, etc., for its full expression.

I am presenting elsewhere ('18) the results of a study of the
organization, reproduction, and heredity of Pediastrum asperum,
together with further observations on the variations in a series of
seven colonies of P. Boryanum. The results there described are
assumed in this paper. These two species represent respectively
the forms with spines well developed orf both peripheral and in-
terior cells of the colony and rather large intercellular spaces, and
those with small or no intercellular spaces and the spines little
developed on the interior cells. In both species, I have pointed
out, the cells seem in some degree directly adapted to the forma-
tion of bilaterally symmetrical plate-shaped colonies of sixteen

212 Semi-centennial of Torrey Botanical Club

cells. There is a considerable number of species of Pediastrum
in which the cell form is by no means so obviously adapted to a
symmetrical configuration of the colony as a whole, species in
which in some cases the lobing of the cells, as noted, is carried to
extremes or is of a kind calculated to result in more unstable
conditions of equihbrium and a greater tendency to asymmetry
in the colonies. This increase in the length of the spines and
the correlated large size of the intercellular spaces by favoring
floating may be adaptive for species tending to assume the plank-
ton habit of life. These species in their relations to P. asperum
may be considered as extreme or aberrant, though representing
natural and easily conceived modifications of the cell form shown
in it and in P. Boryanum. Such types illustrate the operation of
orthogenetic tendencies in the production of results which are
non-adaptive from the standpoint of the organism as it was
situated when the tendency first appeared, but may become adap-
tive under new environmental conditions.

P. asperum apparently represents a climax type viewed from
the standpoint of the possibilities of developing a least surface
configuration with unit 'cells derived by bipartition. The deeply
four-lobed form of the cells permits the best possible approxima-
tion to conformity with the circular outline and the intersection
of all boundaries at 120" as found in the corresponding nine teen-
unit least surface configuration.

The two-spined series has by far the largest number of species
and is unquestionably the most common type, though the single-
spined species, P. simplex, is at times found in great abundance
and in almost pure growths. In general, and in all its variations,
the two-spined form of cell seems to be better adapted to the
formation of the anomogenous cell group with the bipartition
series of cell numbers as contrasted with normal least surface
groups. There is general agreement that the delimitation of
species must be based on cell form and we have quite a series of
types differing primarily in the degree of development and the
form of the two spines.

We shall get a clearer understanding of the significance of
cell form in such groups if we compare the conditions found in
other species of the Diactinia with those in P. Boryanum and P.

Harper: Cell types and responses in Pediastrum 213

asperum and further with those in reprCwSentatives of the Anomo-
pedium group, P. integrum, the Monactinia, the Triactinia, the
Tetractinia, etc. I have not been able to study the reproduction
and colony formation in these species and can only compare
their adult forms with those I have more fully studied.

A . Anomopedium. — Pediastrum integrum Nag.

I have seen only a few specimens of P. integrum. The indi-
vidual shown in figure i (*i8) has but seven cells, the missing
cell perhaps lies at some point above or below the other cells of
the colony though it can not be made out in the photograph nor
was it discovered in the specimen before photographing. The
species is figured by Nageli ('49) as occurring in 4-, 8-, 16-, 32-, etc.,
celled colonies, all of whose marginal cells have two very rudi-
mentary spines. All the colonies figured by Nageli, except three
of those with four or eight cells, are very irregular and he states
that the regularly concentric arrangement of the cells which is
common in other species of Pediastrum is exceptional in P.
integrum. He also states that it is common to find the cells in
two layers. Braun ('55) does not figure the species.

Nageli gives the habitat of the species as wet cliffs. Nitardy
('14) reports that he has seen but one specimen. His figure shows

Fig. I. Pediastrum integrum Nag. Perhaps the same as P. muticum Borge.
Spines more prominent than in the eight-celled colony figured in 'i8, fig. i. X
about 300.

that it was quite irregular and he notes that it was plainly two-
layered. He explains his failure to find more specimens as due to
his having been concerned especially with material from lakes, pools,
etc., and notes the agreement of most authors that P. integrum is
found on rocks over which water trickles rather than in the deeper
waters of pools, etc. The question at once arises whether the ir-
regular form of the colonies in two layers is not due to environmen-

214 Semi-centennial of Torrey Botanical Club

tal conditions which check the vigor of the swarm-spores just as it
is checked in agar cultures of P. Boryanum with resulting irregu-
larity of the colonies and reduction of the spinous projections. It
seems hardly possible that P. integrum is only a habitat form of P.
Boryanum but it would be interesting to grow it in water cultures
along with P. Boryanum and observe its behavior as to cell shape
and the symmetry of the colonies. I have not had the species in
numbers sufficient for such experiments. The specimen shown in
FIGURE I shows the thirty-two cells in one plane but the arrange-
ment is quite irregular and asymmetrical, though there is a fair ap-
proach to the concentric circles. The short papillae seem to be quite
reularly directed radially outward in the colonies so far figured, but
the tendency to have certain cells out of the plane of the colony is
marked and either indicates that the poles of the transverse axes
of the cells and the affinities which they represent are relatively
less strongly developed or that the cells are unable for environ-
mental reasons to achieve their normal orientation and inter-
relations in the colony. P. integrum may be a species whose
colonies regularly fail to achieve their typical development and
yet are able to maintain themselves. I am more inclined to
believe, however, that it is a primitive type in which the polarities
and cell differentiations characteristic of the Monactinia and
Diactinia are not fully achieved. Whether or not P. integrum
is a good species, it may certainly be regarded as representing in
the form of its cells a primitive type out of which the better
adapted cell forms of the Diactinia have been developed.

B. Monactinium. — Pediastrum simplex Meyen.

The species of Pediastrum whose cells show a single spine are
common and widely distributed. The evidence of their varia-
bility is well shown in the fact that De-Toni ('89) believes they
can all be included in one polymorphic species, P. simplex, while
various other authors have recognized P. duodenarium (Bailey)
Rabenh., P. clathratum Lemm., P. triangulum (Ehrenb.) A. Br.,
P. Sturmii Reinsch, etc. P. simplex has been described as in-
cluding forms both with and without intercellular spaces.

It is impossible to determine from the evidence in the litera-
ture as to the independence of these various forms and yet there

Harper: Cell types and responses in Pediastrum 215

can be no doubt that in different localities and in different seasons
one or another of them may be found in almost pure growths and
to the exclusion of the others. It seems clear that a particular
type of cell form tends to perpetuate itself, but with much fluctua-
tion. Nitardy (*I4) includes all the forms under the name P.
triangulum (Ehrenb.) A. Br., dropping the name P. simplex
because it has been used so variously by different authors.

I have photographs of both eight- and sixteen-celled colonies.
A comparison of the sixteen-celled colonies of P. simplex, as I
find it, with those of any species with well-developed two-spined
cells suggests at once the greater capacity of the latter to form
symmetrical colonies with the bipartition series of cells 4, 8, 16,
32, 64, etc. The two common arrangements in my observation
seem to be, first, an irregular group of five in the center and eleven
around them (fig. 2), and, second, an irregular group of four in
the center with twelve around them (fig. 3).

The peripheral cells show their polarity by the regularity
with which the spine is turned outward. Whether they are
flattened enough to indicate the existence of a second differentia-
tion in a transverse axis, I have not been able to determine.
The arrangement of the inner group of cells is, as noted, quite
irregular in all the colonies I have seen. Several intercellular
spaces are commonly present, but show no constancy as to size,
shape, or position. They may be from three- to six-sided. It
is generally quite impossible to determine which side of any one
of these interior cells tends to be produced into the spine. In two
cases, however, I have observed an interior cell with a well-
developed spine projecting into an intercellular space. One of
these is shown in figure 4. Whether the tendency to form the
one-spined form is as fully fixed in P. simplex as is the correspond-
ing tendency to form two spines in the Diactinia is not clear.
It is possible that the lack of adaptation in the one-spined cell
form to the production of symmetrical colonies is correlated with
a failure to fix this cell form so firmly in heredity and that with
an increased fixity of cell type such symmetrical forms as Nitardy's
P. triangulum (Ehrenb.) A. Br. become possible. De Wildeman
C93)> however, has observed both irregular and symmetrical
colonies and includes them all in P. simplex. De Wildeman

216 Semi-centennial of Torrey Botanical Club

figures likewise only one symmetrical sixteen-celled colony ('93,
pi. IQ. f. g) against five which are irregular, but the form of the
cells in all of them is essentially the same.

Nitardy (^14) has had an abundance of the more symmetrical
forms and their variants which have been variously named simplex,
duodenarium, etc., and, as noted, puts them under the name P.
triangulum (Ehrenb.) A. Br. with two varieties, angustum and
latum. The latter in its cell form (Nitardy, pL 8. f. 5) agrees
with the forms I have found and the two varieties seem fairly
well marked in the case of his figures 3 and 5, plate 8. In some
of his other figures the distinction is not so clear, but the drawing
is rather crude and it is hard to judge. In his variety angustum

Figs. 2, 3, and 4. Pediastrum simplex Meyen, sixteen-celled colonies showing
the cell arrangement 5 + 11 and 4 + 12, X about 175. Figure 4 shows one of the
central cells with a well-developed spine, pointing downward, but not very clearly
down in the reproduction.

the lobes of the cell are more slender and form large intercellular
spaces. The sixteen-celled colonies may have a very definite bi-
laterally and radially symmetrical arrangement of their cells with
four in the center in a square, surrounded by four groups of three
(Nitardy, '14, pi. y. f. 5), or two groups of four and two groups of
two (Nitardy, '14, pi. g. f. 20), or five in the center, a pentagon
surrounded by eleven — four pairs and three (Nitardy, ^14. pi. 5.
/. i). These symmetrical figures are from his own observations
on material from Griinewaldsee near Berlin and are very fine
illustrations of the capacity of the swarm-spores of this species
to achieve delicately balanced equilibrium relations such as are
necessitated by the one-spined cell form. I have not seen these
symmetrical forms and have taken from Nitardy ('14) figures
27, a, hy etc., to illustrate the very interesting bilateral and radial
symmetry which these sixteen-celled colonies may show.

\

Harper: Cell types and responses in Pediastrum 217

My material corresponds more nearly with Nitardy's var.
latum and Nitardy gives no evidence that his var. angnstum
ever produces dense colonies with small irregular intercellular
spaces like the forms of P. simplex I have figured. I shall refer
to these bilaterally symmetrical colonies and their variants as P.
triangulum (Ehrenb.) A. Br. (Nitardy, '14, pL 4. f. 4, 7, 8; pi. 5.
f. I, 2; pi. 6, etc.) and to the irregular forms I have figured as P.
simplex (De Wildeman, '93, pi. ig.f. 14, and perhaps Nitardy, '14,
pi. 6. /. j), leaving unsettled the question whether the one type
can arise from the other directly as De Wildeman and others have
supposed. Nitardy says he has not seen the forms without inter-
cellular spaces such as De Wildeman has figured and evidently
about Berlin the symmetrical form and its variants occur pre-
dominantly. It is not impossible, of course, that the irregular
forms are merely the expression of a lack of vigor at the swarming
period.

The regularity of the peripheral series in P. simplex even when
the central cells are asymmetrically placed is doubtless due to
the fact that the outer series of cells seems to come to rest sooner
than those in the interior of the swarming group, as I have noted
elsewhere ('18). The colonies of P. simplex with the irregular
central groups (figs. 2, 3, 4) are particularly interesting as illus-
trating a case in which while symmetrical relations of contact
and pressure are apparently impossible for all the members of the
cell colony they are none the less quite perfectly achieved for a
portion of the colony — the peripheral series — so far as their inter-
relations are concerned.

The eight-celled colonies of P. simplex offer fewer possibilities
in the complexity of their intercellular relations and there is also
much greater uniformity of type. By far the commonest arrange-
ment is one cell in the center surrounded by seven cells (figs. 5
and 6). It is evident here that the single cell can hardly fill a
circle made of seven instead of six cells (s€e figs. 5 and 9) and
intercellular spaces tend to appear and may sometimes be quite
large. In figure 6, however, the central cell seems quite to fill
the center of the group of eight. I have never seen a colony of
this species with two cells symmetrically placed in the center
and surrounded by six cells, as is so commonly the case in the

218 Semi-centennial of Torre y Botanical Club

eight-celled colonies of P. Boryanum. The markedly oblong form
of the body of the central cells in P. Boryanum is plainly quite
impossible for P. simplex.

I have seen a number of colonies in which (fig. 7) all eight
cells were arranged in a very perfect circle about a central rounded
space. Meyen ('29) and other authors since have figured such
forms. Nitardy has found it in his P. triangulum (fig. 27c).
There evidently is the tendency here to achieve a symmetrical
arrangement of one sort or another — a tendency which is quite
independent of the presence or absence of any adaptation in the
form of the cells to the production of such symmetrical inter-
relations. We may assume, as in the other species, that this
tendency is based on the effort of the cells in the swarming period
to achieve a position in which their contact and pressure relations
will be equal and balanced in as many directions as possible or
that at least such pressure relations as are achieved shall be as
nearly as possible mutually compensatory, as in the ring-shaped
colony. The significance of occasionally achieved chance con-
figurations is well illustrated in these cases. The symmetry of the
circle is here very perfectly illustrated in figure 7, but the chance

Figs. 5, 6, and 7. Eight-celled colonies of Pediastrum simplex Meyen. 5,
central cell does not fill space enclosed by peripheral cells. 6, no intercellular space.
7, ring-shaped colony, X about 150.

that out of a swarm of eight free-swimming cells attempting to
achieve interrelations of equal or balanced contact and pressure
such a circle will be achieved would seem very remote. The
chance for seven about one in a free-swimming group is much
greater and affords a sufficiently close approximation to symmetry
to make unlikely any very radically different configuration when
once it is achieved. A statistical study of the relative abundance
of such colonies as are shown in figures 5, 6, and 7 might throw

Harper: Cell types and responses in Pediastrum 219

light on the question as to the relative abundance of individuals
of the highest vigor as compared with those of high vigor but not
the maximum.

It is evident that when in swarming the ring form is by accident
once achieved it tends to persist, since it gives, as shown by its
contours, the most perfectly symmetrical interrelations possible
for the eight simplex cells. Just why the cells should tend to
find their final resting position in a situation of equal or balanced
pressures and contacts instead of unequal or unbalanced pressures
and contacts is the same question here as in the case of other
coenobes. There are obviously two factors or sets of factors
involved in all these adjustments- First, it is plain that during
the slow and protracted writhings of the swarm-spores of vigorous
colonies the direct physical tendency of such viscid, semi-fluid
droplets to adhere and yet as far as possible round up and assume
a least surface configuration will have the fullest possible oppor-
tunity to come to expression. In the random movements of the
swarm-spores these constantly acting physical relations will tend
to maintain any accidentally achieved position which is comform-
able with them and to act as a check on any movement unconform-
able with them. Results of this sort, however, will be chiefly
in evidence in the later stages of colony formation. The general
arrangement of the swarm-spores in a plate or ring must be regarded
as the result of the interrelations of the cells as free motile organ-
isms involving polarities, tropisms, etc., such as are observed in
other morphogenetic processes.

The grouping of eight cells in a ring rather than in a plate of
seven about one in the case of P. simplex, which we are consider-
ing, brings out most clearly the relations of two divergent types
of activity. The unconformability of a group of seven units
about one — instead of six about one — with the principle of least
surfaces as applied to the whole group is what prevents any
swarm-spore that accidentally comes into the center of such a
group (fig. 6) from achieving equal contact and pressure relations
with all the cells about it and thus leads to its changing its posi-
tion until in the present case it makes one of a ring of eight (fig. 7) .
This is a matter of intercellular reactions involving contact and
pressure, polarities, tropisms, etc., the physical unconformability

220 Semi-centennial of Torrey Botanical Club

providing for the stimuli. It is obvious that the mere molecular
pulls involved in least surface phenomena could never lead directly
to such violent changes of group relations, though they may be
of final importance in determining the exact contours of the ring
when once it is blocked out, as it were, as a result of the inter-
cellular reactions which control the earlier swarming movements.

It is then by no means enough, even when, as here, the in-
herited cell form is not especially involved, to identify offhand
the results of these complex physiological reactions as simply
the expression of the physical principle of least surfaces which,
at least as at present stated, is based on intermolecular relations.
If the facts are as indicated we perhaps have here in the relations
of the complex reactions of these simple organisms to the wide-

FiGS. 8, 9, and lo. Pediastrum simplex Meyen, of the form known as P. Sturmii
Reinsch. 8, with small intercellular spaces, X about 200. 9, with a large inter-
cellular space, X about 250. 10, sixteen-celled colony, eleven cells empty, simplex
type, and five cells about ready to form swarmspores, Sturmii type, X about 200.

spread symmetries of form, rounded contour, etc., which are based
on the physical principle of least surfaces, a suggestion at least as
to the origin of the so-called aesthetic satisfaction of higher
organisms in physical symmetry and balance of configuration or
artistic composition. This capacity to react to inequalities in
contacts and pressures or unbalanced pressures from the environ-
ment might be ascribed to a sense of symmetry and classed as one
of the fundamental properties of living cells. These ring-shaped
colonies of P. simplex certainly suggest a notable delicacy of
response to pressure and contact stimuli.

Pediastrum Sturmii Reinsch, characterized by plumper, more
rounded cells with spines supposed to be solid instead of hollow, may

Harper: Cell types and responses in Pediastrum 221

be represented by the form I have shown in figures 8, 9, and 10.
In my opinion, these are merely colonies of P. simplex approaching
the reproductive stage, but I have not so far observed swarm-spore
formation in the species. Nitardy ('14, p. 178) regards a warty
surface, pentagonal peripheral cells, and the non-tapering form
of the spine as important characters of P. Sturmii.

Figure 10 gives quite convincing evidence on this point.
In this colony five of the cells are well advanced toward reproduc-
tion while the remaining eleven have remained immature. The
mature cells have the form characteristic of P. Sturmii; the
immature cells are like those of the ordinary colonies of P. simplex
which have not yet reached the reproductive stage, but it is to
be noted that the cells in this colony are much smaller than in
the other two. The change in form of the cells as they approach
the period of reproduction is very marked. Their rounding up
leads to the at least partial withdrawal of material from the
lobes and a narrowing of their bases. Failure to recognize these
changes may be responsible for confusion as to the real character
of P. Sturmii. Professor B. M. Davis has kindly shown me, and
permits me to refer to, as yet unpublished figures showing the
reproduction of P. simplex which seem to me not inconsistent with
this view.

In figure III have been able to bring out faintly, in my
original prints, the curious bristle-like appendages at the ends of
the spines, which like the similar structures in plankton diatoms
tend to keep the organism afloat (see Petersen, '11, and Zacharias,
'03). As shown here, it is clear that the individual setae may be
widely divergent or almost parallel. The suggestion that they are
movable on their bases or points of insertion in the spines is very
obvious. The nodular or pear-shaped swellings at their bases are
also very conspicuous in the case of the widely spread group from
the spine of cell a and may very well be contractile droplets of
cytoplasm functioning as motile organs.

This colony is further interesting from the fact that the cell a
seems relatively large as compared with the cells next to it. The
whole colony also shows only fifteen instead of sixteen cells.
As Nitardy and others have emphasized, the law of bipartition
and the resultant 4-, 8-, 16-, 32-, etc., cell numbers are very firmly

222 Semi-centennial of Torrey Botanical Club

fixed in the colonies of Pediastrum. It is possible that the six-
teenth cell lies above or below the plane of the colony in this case
and hence does not appear in the photograph. On the other
hand, it may be that cell a did not undergo the third division and
has remained larger than its fellows.

Fig. II. Pediastrum simplex Meyen. Fifteen-celled colony, one cell, a, larger
than the others; bristles and pear-shaped basal bodies show on some of the spines
in the original, but are practically lost in reproduction. X about 400.

Nitardy ('14, p. 183) figures a specimen of P. Boryanum with
fifteen cells, one of wl^ich is much larger than the others and
contains a double pyrenoid. There is also a faint line running
across the middle of the cell and Nitardy is convinced that the
large cell has arisen by fusion of two swarm-spores. He notes
that this is the only case of anomaly in cell number which he has
observed in twenty years of study of the group. It would seem
that the large cell might equally well be due to a failure of one of
the cells to complete the third division. The larger size of the
cells in the sixteen-celled daughter of a thirty-two-celled colony
(*i8, FIG. 21) as compared with the thirty-two-celled daughter of
the same mother is conspicuous.

Harper: Cell types and responses in Pediastrum 223
C. Diactinium.

In the Diactinia we have the largest and most common of the
subgenera of Pediastrum. The deHmitation of species, however,
seems in high degree difficult and uncertain. De-Toni ('Sg)
recognizes eight species in the group with eight varieties under
P. Boryanum (Turp.) Menegh. and seven varieties under P.
duplex Meyen, many of which are regarded by other authors as
good species. Nitardy (^14) believes the whole series can be best
regarded as three species with three varieties under P. Boryanum
(Turp.) Menegh. and three varieties under P. pertusum Kiitz.

Braun C55) gives no figure of his P.- pertusum Kiitz. var.
dathratum, but refers to figures of Meyen ('29), Hassall ('45)
and Corda ('39, pL 3, f. 18). Corda's figure of P. diodon is cer-
tainly unreliable. Meyen's figures, though poor, seem to agree
rather better with Lagerheim's figures of his var. reticulatum

Fig. 12. Pediastrum Boryanum (Turp.) Menegh. Sixteen-celled colony typical
arrangement, form with slender equal spines, X about 300.

Fig. 13, P. asperum, sixteen-celled colony, typical form, X about 425.

Fig. 14. P. duplex Meyen, var. reticulatum Lagerh. Intermediate between
P. asperum and P. clathratum (figs. 15-21), X about 300.

(^82) than with those of De Wildeman (*oo, p. 104, /. ly and 18) j
which are labelled P, duplex var. reticulatum Lagerh., and those of
Chodat ('01, p. 227 and 228) labelled P. duplex Meyen and P.
duplex f. genuinum (A. Br.). Hassall 's figure (^45, pi. g2. f. 4) is
certainly widely different from those of De Wildeman and Chodat.
It seems doubtful whether Nageli ('49), Braun, or Lagerheim had
these 5 + II forms figured by De Wildeman, Chodat, and
Nitardy ('14, pi. 8. f. ii). Both types are found in this country

14

I

224 Semi-centennial of Torrey Botanical Club

and cultures will have to show whether they both can be produced
from the same mother colony. Until the question is settled it is
certainly more convenient to keep them under the old names of
Meyen, Kiitzing, and Braun. I shall call the type shown in
figure 14 p. duplex Meyen var. reticulatum Lagerh. (Meyen,
'29, pi. 4j. f. 16 and 17 \ Lagerheim, '82, pi. 2. f. i), and the
forms shown in my figures 15-21, P. clathratum A. Br., P.
pertusum Kiitzing, may very well be Braun 's var. asperum^
though there may be a form with smooth spines connecting P.
asperum with P. Boryanum. This form of mine (fig. 14) is
plainly Lagerheim's P. duplex var. reticulatum ('82, pi. 2. f. i).

P. clathratum A. Br.

In Pediastrum clathratum A. Br. we have a species of the Diac-
tinia in which the four-lobed cell type has been carried to its extreme
development. It is a fairly common and abundant species, ap-
parently vigorous and well adapted to the conditions it finds. In
the extreme length of its cell lobes as compared with other species
of the Diactinia, P. clathratum is obviously a climax type. What
we may call the body of the cell in P. Boryanum (fig. 12) has gone
over almost completely into the four spinous lobes. The cell is
quite H-shaped, with cross-bars little or no thicker than the arms
(figs. 15-18). As a result, the adaptation of the lobed form of
the cells to the exigencies of colony formation, with cell numbers

im

Figs. 15 and 16. Pediastrum clathratum A. Br. Sixteen-celled colony and type
diagram. Fig. 15 X about 300.

produced by bipartition, works out in quite a different way from
that in P. Boryanum and P. asperum (fig. 13). It is a type in
which the four-lobed form in its extreme development has resulted
in a reduction of stability and compactness in the organization of

Harper: Cell types and responses in Pediastrum 225

the colony as a whole. The result is a light and open structure
which may be better suited to conditions of life in the plankton.

The intercellular angles are hard to measure because of the
very limited areas of contact between the cells but there can be
no question that the variations from 120° are so slight as not to be
accurately determinable by the means I have used. The slender-
ness of the lobes makes possible in the highest degree compensa-
tory curvings and bendings so as to give quite equal pressure and
tension relations between the surfaces of contact of the cells.
The extreme length of the lobes has brought with it a new type of
cell grouping in the colony. I have had an abundance of material
of this form and have not seen a single sixteen-celled individual
with the common cell arrangement i +5 + 10, found in P. Boryanum
and P. asperum.

Nitardy (pL 8. f. ij) refers to P. clathratum, a sixteen-celled
colony with the ordinary arrangement found in P. asperum
I + 5 + 10. The cell form also is plainly that of P. asperum
rather than that of P. clathratum. Chodat's ('01) figures of what
he identifies as P. duplex are both of the clathratum type. His
figure 151 is of an irregular older colony nearer the stage of repro-
duction, but figure 152& shows the type configuration of the cells.

The type arrangement seems to be that shown in the diagram
(fig. 16), five cells surrounded by eleven cells and the center of
the colony an open pentagonal or oblong area. Such a colony is
bilaterally symmetrical about the axis mn, as shown in the
diagram. The outer series of cell contacts is in threes except at
the pole m, where there is a contact between two. The inner
series of contacts is all in twos. The central intercellular space
is, as noted, pentagonal and more or less elongated in the axis of
the colony. With the variation in the shape of the central inter-
cellular space the whole colony becomes either rounder or more
oblong. Compare figures 15 and 17. In correlation with the
length and slenderness of its cell lobes P. clathratum is a climax form
in its development of intercellular spaces. The outer series of
intercellular spaces consists of five inequilateral lens-shaped and
five shield-shaped openings bounded by two and three cells
respectively, with the large oval and four-cornered intercellular
space near the pole w, bounded by four cells and bisected by the

226 Semi-centennial of Torrey Botanical Club

axis of symmetry of the colony. No such configuration is found,
so far as I have observed, in any other species of the Diactinia
so far described, and yet the cell form shown in figure 14 connects
P. clathratum with P. asperum very closely. This figure is from
material collected in Wisconsin and I have quite a series of photo-
graphs showing cells of this form in colonies with 16, 32 and 64
cells, but I have never seen one of these colonies with the 5 + 11
cell configuration of P. clathratum. My figures of P. clathratum
are from material collected at Woods Hole, where the typical
form is common as well as the less developed types of P. asperum,
but I have not found with these forms colonies exactly like those
from Wisconsin.

We have here two types, which, as the confusion in the litera-
ture shows, can be connected very closely by all possible inter-
gradations in cell form and yet it seems clear that either when
the modification of the cell form passes a certain point or as a
result of modifications of the cell polarities a change in the type
configuration of the colony results. There is no good evidence
in the literature that colonies with the 5 + 11 configuration of
P. clathratum and those with the i + 5 + 10 configuration of
P. asperum can arise from the same parent colony and, as noted
above, it must be left to further culture work to show whether
this is possible.

Figs. 17, 18, and 19. Fediaslrum clathratum A. Br., irregular colonies. 17,
circular colony, X about 175; figure 18 showing quite clearly the pear-shaped
bodies at ends of spines, X about 200. The colony shown in figure 19 is about
ready for reproduction, X 175.

The configuration of such colonies as those shown in figures
15-17, involving the absence of a central cell, the increased variety
of form in the intercellular spaces, and the increased number of
paired cell contacts is certainly more complex and further removed
from that of a simple i + 6 + 12 least surface group than is the
type of sixteen-celled colony of P. asperum (fig. 13). Greater

Harper: Cell types and responses in Pediastrum 227

delicacy in the contact and pressure responses of the swarming
zoospores is certainly necessary to achieve it. It is the most
highly specialized configuration I have yet observed in any of the
species of the genus, though whether greater delicacy of response
is necessary for its production than for that of the ring-shaped
eight-celled colony of P. simplex is not easy to say.

The colonies are very commonly irregular and indicate very
clearly that the normal contact relations are not necessary for
the development of the typical cell form, as is illustrated by the
interior cell with one free spine shown in figure i8, and by De
Wildeman's (*oo,/. i8, p. 104). I have been able in several cases
to observe the reproduction of the species. The cells become
much swollen but still show a very deeply lobed form as compared
with P. asperum at the corresponding stage. Figure 19 shows a
thirty-two-celled colony which is about ready for reproduction
and figures 20 and 21 show two very irregular young colonies,

Figs. 20 and 21. Young, irregular colonies of Pediastrum clathratum A. Br., X
about 400.

whose birth I observed in a sealed preparation. They are only a
few hours old but the cells show the contours characteristic of the
species. It is of interest to compare these figures with figure 14
as to the forms of the cells.

The whole colony is relatively fragile and is ordinarily bent
and curved so as to make a good photograph impossible. The
bristle-like projections from the ends of its cell lobes are extremely
well developed and are brought out faintly in some of my photo-
graphs. The colonies are very sensitive to currents in the water
and seem almost self-motile at times. It is very difficult to find
one quiet enough for photographing and the varying position
of the apical bristles, now all close together in a parallel pencil

228 Semi-centennial of Torrey Botanical Club

and now widely diverging as shown in my figure of P. simplex
(fig. ii), would seem perhaps to be a factor in the wabbling,
tipping, and trembling movements of the colonies. I have,
however, seen no movement of the bristles. The species illus-
trates the possibility that an orthogenetic tendency which is adap-
tive in a specific particular when moderately developed may in its
extreme development become adaptive in quite a different connec-
tion.

P. angulosum (Ehrenb.) Menegh.

This form (fig. 22) represents a type of the diactinial cell
which seems quite remarkable for its constancy and the name
and species have been less juggled with by descriptive writers
than many others. The characteristically short oblique spines
with the wide sinus between them show very clearly that the

Fig. 22. Pediastrum angulosum (Ehrenb.) Menegh. Irregular sixteen-celled
colony, X about 325.

morphogenetic tendency to the production of such projections
involves other factors than merely those of length. It is a widely
distributed and fairly common form and yet apparently has
developed no such series of fluctuating variants as have P. Bory-
anum and P. pertusum. De-Toni ('89) recognizes no varieties of
it. Nitardy (^14) has apparently never seen it and refers it with-
out adequate evidence to P. Boryanum. The colonies tend to
high cell numbers and in the 32- and 64-celled types have quite
regularly a somewhat reniform outline which is suggested also
in the sixteen-celled colony. That the cell form is in any way
adapted to or determines this configuration of the colony is not
obvious, and the form-determining factors are not so readily
recognizable as in the other Diactinia.

I shall discuss the species further in considering the general

Harper: Cell types and responses in Pediastrum 229

question of the relation of the larger cell numbers to the con-
figuration of the colonies in another paper. The shortness and
form of the spines suggest in some degree those of P. Ehrenbergii,
but the species cannot be regarded as in any proper sense a transi-
tion form between the Diactinia and Tetractinia.

D. Pediastrum tricornutum Borge.

This species, representing perhaps a series of Triactinium
(Nitardy makes it Diactiniopsis) , I have not seen. It differs
characteristically from other types in that the three spines do not
lie in the same plane and hence have no part in determining the
intercellular contacts in the colony. Only eight-celled colonies
have been figured so far — one cell in the center surrounded by
seven. Such a form could hardly be conceived as developing
from the Diactinia by progressive variation. It may have
originated from a form like P. integrum as a representative of
quite a different line of development or it may have connections
with Coelastrum in quite a different series.

E. Tetr actinium. — Pediastrum Ehrenbergii A. Br.

The Tetractinia illustrated by the common P. Ehrenbergii
(figs. 23, 24, 25, 26) are those types in which the two spines or

23

■1 1

" ■]

Figs 23, 24, 25, and 26. Pediastrum Ehrenbergii A. Br., with var^ung degrees
of lobing of the cells. Figure 26 shows a four-celled colony with one cell almost
free but showing none the less the characteristic wedge-shaped form, X about 350.
Fig. 23 X about 1000, Fig. 24 X about 700, Fig. 25 X about 550.

lobes of the Diactinia tend to become more or less deeply bifid.
The species commonly occur in four-, eight-, and sixteen-celled

230

Semi-centennial of Torrey Botanical Club

colonies and consist in the latter of a group of four or five central
cells surrounded by, respectively, twelve or eleven peripheral cells.

In this group again the splitting or doubling of the spinous
projection of the cell in its incipient stages foreshadows the further
development of this character through the whole series. The
incised or bifid tips and the doubling of the spines appear in
graded stages of development which suggest very strongly that
the species have been produced as end members in series of con-
tinuous variants. Under P. Ehrenhergii and its synonyms we
find included by most authors forms in which the degree of lobing
varies widely. In some forms the cells are only bluntly bifid
(fig. 26). In others there is every degree of inequality between
the two points of the bifid spine, suggesting that the forms may
have arisen from the Diactinia by the budding off of an accessory
tooth on the main spine, or at another point on the body of the cell
rather than by splitting the tip of the spine itself (fig. 25).

Nitardy's treatment of the group recognizes the depth of
lobing and the degree of separation of the points as the principal
basis for distinguishing the types and his two species, with a
variety under the first, form what it seems to me is in part at least
a natural series.

In his first species, P. incisum Hassall, however, Nitardy
includes forms with the spines very unequally cleft (fig. 24), one
half frequently much more strongly developed ('14, pi. 5. /. 7
and pi. 7. /. 8) along with others in which the spines are very short
and even blunt ('14, pi. 7. /. 6, 7, and 11). There is no adequate
evidence that all these forms could come from one mother colony.
In the variety P. incisum var. Rota Nit. he includes a natural
group with the spines much more strongly developed and as a
rule quite equally bifid.

In the second species, P. lohatum Nit., he includes what he
regards as the handsomest forms in the whole genus, with strongly
developed lobes deeply bifid {pi. 5.f. 4). This is plainly Braun's
and Cooke's P. Rotula Ehrenb. The five species and four varieties
recognized by De-Toni also show characteristic differences in the
cell form and lobing but reliable figures are not available for
grouping them in an evolutionary series.

I have never seen in P. Ehrenhergii the i + 5 + 10 configura-

Harper; Cell types and responses in Pediastrum 231

tion which is so common in the Diactinia. In the eight-celled
colonies the common arrangement is 1+7? with the central
cell having the appearance of being rather crowded and suppressed
in its development though quite regularly showing the narrow
notch characteristic of the cell form of the species (fig. 24)^.
This apparent crowding of the central cell in a group of i + 7
is quite contrary to what one observes in P. simplex and other
species. It is due to the pronouncedly wedge-shaped form of
the cells. That the cell form is, as in other species, hereditary
and not dependent on the pressure of adjacent cells for its develop-
ment is shown in figure 26, which represents an irregular four-
celled colony, one of whose cells is almost free and has none the
less developed the wedge-shaped form.

The quadrifid cell form apparently does not lend itself to the
formation of symmetrical least surface configurations with regular
intercellular spaces as does the duplex form. I have seen no
colonies in which there was any indication of the utilization of the
quadrifid character in the interior cells of a colony in developing
symmetrical intercellular relations. Braun's figures of P. Rotula
('55j P^' 6' /• 5-^2) suggest that such cases may exist. Braun's

i % '

\ .e=:nl--

L , a ,

Fig. 27, a, b, c, d. Reproduced from Nitardy ('14). a, b, c, P. Iriangulum
(Ehrenb.) A. Br.; d, P. Rotula Ehrenb,, sixteen-celled colony showing bilateral
symmetry.

figure ('55, pi. 6. f. 5) and Cooke's figure (^84, pL 18. f. 2d) are
fine examples of bilateral symmetry in eight-celled colonies of
P. Rotula Ehrenb. In the sixteen-celled colonies the central
group of four or five may make a ring with a four- or five-sided
intercellular space in the center and the peripheral cells may
also make a very perfect ring-formed series of eleven or twelve.
There are, however, no relations of symmetry further than this
general concentricity between the cells of the two series shown

232 Semi-centennial of Torrey Botanical Club

in any figures except one by Nitardy ('14, pi. 5. /. 4), which
I have reproduced (fig. 2'jd). It would seem that the tendency
to lobing which fits so perfectly with the principles of least sur-
faces and binary fission in the simpler forms of the diactinial
type has here gone too far and become distinctly non-adaptive
as far as the symmetrical grouping of the cells is concerned.
The tendency to the lower numbers of cells in the colonies of
these species is, if real, a curious correlation, since the quadrifid
form of the cells would not in any case seem capable of limiting
the number of times they should divide in reproduction.

Discussion

The general relations of the integrum, Monactinium, Diac-
tinium, Triactinium and Tetractinium types suggest at once
certain evolutionary possibilities and limitations in very clear
form owing to the extreme simplicity of the characters involved.
Evolution in the whole group has proceeded by modification of
the cell form. It is quite obvious, as noted, that a species with
one spine could not become gradually modified into a species
with two spines in any other way than by returning to the spine-
less type and then advancing on quite a different line of develop-
ment resulting in typically different intercellular relations in the
colony. We have no evidence of the possibility of transforming
a simplex type into a two-spined type by the gradual development
of a second spine in addition to the one already present or by
splitting the single spine. A form with one long, well-developed
spine and one short, rudimentary spine is not only unknown in
nature but is quite inconsistent with the colonial organization
of the cells in the plate-shaped groups which are characteristic
of the whole genus. The only obvious evolutionary routes from
a one-spined to a two-spined type are either as noted by a change
back to the primitive integrum type and a new start in a character-
istically different direction or by sudden mutational transforma-
tion, perhaps to be considered a reduplication, by which a form
with one spine becomes at once a form with two equally well-
developed spines. There is an analogy here with the reduplications
in the lobing of fern fronds and pinnae by which the common
sports of the Boston fern have been produced (see Benedict, '16).

Harper: Cell types and responses in Pediastrum 233

That it is mere analogy is, of course, obvious from the fact that
the phenomenon is intracellular in the one case while in the other
it involves the morphogenetic behavior of many-celled tissues and
organs. Given this change of cell form and the diactinial type
of colony would result directly, the same polarity and cell inter-
relations being involved in both cases.

We have, further, manifestly orthogenetic groups in most of
the subgenera. Given the tendency to the development of two
spines and the species of the two-spined group are at once fore-
shadowed as are the species of the Monactinia, Triactinia, and
Tetractinia by the presence of the possibility of developing one-
spined, three-spined and four-spined or bifid-spined cells, respec-
tively. Given cells which adhere in groups, at the same time
having a tendency to develop thick spinous projections with
catenoidal deformation of the entire cell body, and the whole
genus is foreshadowed. Such series certainly illustrate one form
at least of the many types of change which have been characterized
as orthogenetic, though the use of such a term is not specially
illuminating in the absence of evidence as to the structural
features of the cells which have determined their characteristic
forms. A fuller cytological study of the cells of Pediastrum may
serve to throw light both on the nature of cell polarities and the
means by which such orthogenetic transformations are brought
about.

The transition from the simplex to the two-spined type, as
noted, could only come about either by a change giving the new
character in functional development at once or by a return to the
primitive integrum type and a new start. The same is true as to the
possibility of change from the two-spined to the three-spined type.
On the other hand, the change from the simplex to the trispinous
form might quite well come about by the gradual development of
two additional spines with or without the degradation of the
single existing spine. To be sure, the three spines do not lie in
the plane of the colony as does the single spine, but the readjust-
ments which this difference between the two types involves are
by no means inconceivable.

It is notable that in Pediastrum clathratum the interior cells show
almost as fully developed lobes as those on the periphery and in

234 Semi-centennial of Torrey Botanical Club

this respect again I am inclined to regard it as a more specialized
type even than P. asperum, though it is obvious that there is
less differentiation between its cells than is found in P. Boryanum,
in which the interior and the peripheral cells differ notably in
their form. In P. clathratum and P. asperum, however, the
hereditary cell form has become apparently so fixed that it comes
to expression even under the difficult conditions of the interior
cells. If, as I have suggested ('i8), the adaptive oblong four-
lobed cell form originated and developed in direct response to the
environmental limitations and stimuli imposed on sixteen cells
adhering in a plate-formed colony and with a tendency owing to
their partially fluid consistency to assume a surface tension form,
P. clathratum certainly represents the most extreme expression of
this evolutionary trend. The advance has been from such un-
specialized and uniform cells as those of P. integrum through P.
Boryanum with its cell differentiations to P. asperum and P.
clathratum where all the cells are much alike again but vastly
more specialized in form.

The relations of P. simplex and P. triangulum, as I am recog-
nizing them, illustrate the same point. In the sixteen-celled
colonies of P. simplex the interior cells differ regularly from the
peripheral cells by the absence of the spine, though as shown in
figure 4 an interior cell will develop a spine whenever it is so
placed with reference to an intercellular space that this is possible.
In P. triangulum both interior and peripheral cells develop spines
and the configuration of the colony is altered accordingly by the
achievement of symmetry relations which permit each cell to
express much more fully its inherited form tendencies. That these
form tendencies are really present equally in all the cells of P.
simplex also is shown in the eight-celled ring-shaped colonies where
every cell has an equal chance to achieve its full morphogenetic
possibilities and the result is a remarkable uniformity in the size
and shape of all the cells.

The development of spines and the four-lobed cell form in
P. asperum seems to have to do with the compactness and surface
tension relations of the cells in the colony as a group, while in
P. clathratum the length of the spines results in a light, open struc-
ture of the colony perhaps adapted to life in the plankton.

Harper: Cell types and responses in Pediastrum 235

Schroeter ('97) and many others have noted that various species
of Pediastrum may be plankton organisms.

The gradual appearance of the bifid spine in the Tetractinia
is certainly a further development of the tendency to lobing
of the cells and the group forms an obviously orthogenetic series,
but here the bifid spine 'is quite unadapted to the development of
cell groups with the bipartition cell numbers. In cases where
symmetry, either bilateral or merely concentric, in the arrange-
ment of the interior cells of the eight- or sixteen-celled groups is
achieved it is at the expense of the bifid tips which appear, if at
all, only as a broadening of the ends of the spine which hinders
rather than helps the achievement of equal contact and pressure
relations among the cells.

The whole Pediastrum group seems well calculated to show that
fixed trends in development do not necessarily imply adaptation,
though frequently resulting in highly specialized structural differ-
entiations which are plainly adaptive from the standpoint of the
life habits of the organism. Further, openness and lightness
with increased surface in the colony as a whole is the same thing
as deep lobing of its body for the single cell. But the develop-
ment of a rounded least surface contour for a group of cells made
up of the bipartition numbers 4, 8, 16, 32, etc., requiring an oblong
form and perhaps favoring the lobing of the cells, is thus quite a
different thing for them from the same tendency to round up
expressed in their individual masses. In order to make a surface
tension group under the given conditions the cells must lose in
some degree their own tendency to assume the surface tension
form and yet, as I have pointed out elsewhere ('18), this anomoge-
nous condition imposed upon the cells in achieving their inter-
relations in the colony becomes then fixed in heredity so that the
cell develops the characteristically lobed form even when as a
result of accident it develops in almost complete freedom from
contact and pressure relations with its sister cells.

I have referred to the interactions by which the type pattern
of the colony is achieved as based on the polarities of the swarm-
spores and their sensitiveness to contact and pressure stimuli.
That there can be no mosaic inheritance in the case of these
colonies formed by groups of free-swimming zoospores is, as I

236 Semi-centennial of Torrey Botanical Club

have pointed out before, sufficiently obvious. It is also clear
that no spatially differentiated representation of the organization
of the colony in the organization of the mother cell could have
any bearing on the method of transmission of the type configura-
tion of the colony. In P. clathratum both the colonies and the
cells are bilaterally symmetrical, both show polar differentiation,
the colony in one axis and the cell in at least two axes, and yet the
polarity and bilateral symmetry of the cell are in no sense repre-
sentative of the polarity and bilateral symmetry of the colony.
Neither predetermines the other directly though there can be no
question here that the cells as independent units build the colony
and their properties determine its properties. Surface tension is
a common factor in determining the form of the cells and through
the adhesion of the cells to each other in determining the rounded
outline of the colony as a whole, but as I have already pointed out
it is the inherited anomogenous consistency of the cells which is
of most significance in determining their form and it is their
motility, polar differentiations, and sensitiveness to pressure and
contact stimuli which make it possible for them to achieve the
highly symmetrical and characteristic interrelations shown in
the pattern of the adult colony. I am discussing elsewhere ('i8)
the possible relation of these contact and pressure interactions in
the primitive ancestral cell group to the development of the form
of the cells on the principle of functional hypertrophy. However
it may be with this question, which involves the difficult problem
of the inheritance of acquired characters, there can be no doubt
that, as noted, in the species as one finds them the cell form
in its major outlines is fixed by heredity and can be achieved by
the cell when free and quite independent of pressure relations
with other cells in the colony. That the typical cell form is
developed to the extent that opportunity offers regardless of how
the cell is placed in the colony is indicated by the perfection of the
free lobe in one of the interior cells of the colony shown in figure
1 8 and by the development of a fairly good spine on one of the
interior cells of the colony shown in figure 4. More extended
evidence on this point is presented in connection with my study
of P. asperum (*i8).

It seems to me, further, clear that the functional polarity

Harper: Cell types and responses in Pediastrum 237

and the capacity of the cells to respond to pressure and contact
stimuli are not fundamental properties present in full degree in
the ancestral types of the group but that these characters have
increased and become specialized with the gradual development
of the highly modified and lobed form of the cells. Simple ad-
hesion of the cells in a palmelloid mass may have been the initial
stage in colony formation. Light reactions may have favored the
development of the plate-like expanded form though this is
achieved now by the polar differentiation and reactions of the
swarm-spores quite independently of the direction of the light.

These reactions to pressure and contact and the resulting form
determinations are typical examples of biogenetic processes in
Hertwig's sense. It is quite possible that such reactions may be
the determining factors in the root behavior which led Noll (*oo) to
assume morphaesthesia as a fundamental phase of morphogenetic
behavior. Morphaesthesia is for Noll the expression of the
capacity of lateral roots to regain a radial direction of growth
after they have been forced out of it by an interposed obstacle —
radial not to the point of origin of the root from the main axis
but radial to the axis from the point at which the root becomes
free from the obstacle. The capacity to regain such a generalized
relation as that of the radius from any point of the axis opposite
to which the root happens to be certainly implies a response to
form-determining stimuli of the most delicate sort. Noll was
inclined to regard it as a sort of direct reaction to the form of the
whole organism by each of its parts. The only physical stimuli
which seem to be involved are the pressure and contact interrela-
tions involving weight relations, tensions due to bending, etc.,
between the cells themselves.

Whether or not such reactions are adequate to account for the
radial growth of lateral roots with their much greater complexity
of structure, there can be no question, it seems to me, that the
assumption of a fundamental capacity to achieve symmetry is
the natural suggestion from a study of the delicately balanced
interrelations of the cells in such types as the sixteen-celled colonies
of P. clathratum and the eight-celled ring-shaped colonies of P.
simplex. Direct action of surface tension on the plastic though
anomogenous cell bodies may account, as noted, for the final

238 Semi-centennial of Torrey Botanical Club

niceties of adjustment, but the grouping as first achieved by the
free-swimming swarm-spores must be admitted to be a matter of
cellular interactions and the major stimuli in such a series of
adjustments must be contact and pressure. That chemotropism
could play a role is hardly conceivable in view of the violent
movements of the swarm-spores in the narrow confines of the
mother vesicle. That in the last stages of swarming equilibrium
should be reached in a situation of as nearly equal pressure and
contact from all sides as is possible may seem to some to be
merely a matter of physical necessity operating on what, from
the conditions in the adult colony, might seem to be inert
gelatinous four-lobed or one-spined masses, but in the fact that
this equilibrium position is achieved by a group of freely swimming
organisms, each with inherited cell-form tendencies which are
certain to come to expression in greater or less degree, no matter
how the cell is finally placed in the group, we find the proof that
nothing less than the assumption of a capacity to respond to and
maintain conditions of equilibrium when once achieved can ade-
quately account for the symmetry of the typical colonies as we
find them.

That the symmetrical spatial interrelations of the cells is no
mere expression of the direct action of the physical principles of
surface tension, adhesion, mutual pressure, etc., is further shown
by the endless number of variations from the type configuration.
There is good evidence here of trial and error by complex organisms
with every type of error as well as degree of approximation to the
typical fixed in the endless variations in detail which can be found
in the configuration of the adult colonies. It is difficult to give
an adequate picture of what one sees in watching the free-swim-
ming swarm-spores darting here and there around and through
the mass and the gradual appearance of order out of confusion
with the coming to rest first of a peripheral series and then of the
interior cells, but that cellular interactions of sensitive tropic
units determine the symmetrical final configuration rather than
the direct operations of surface tension, adhesion, etc., on the one
hand or any mysterious controlling and adaptive principle of
behavior seems to me the obvious suggestion from the facts. The
evidence seems to me adequate for assuming a high degree of

Harper: Cell types and responses in Pediastrum 239

potency for such simple stimuli as contact and pressure between
polarized cell units like those of Pediastrum in initiating and con-
trolling morphogenetic processes.

BIBLIOGRAPHY

'29. Meyen, F. J. Beobachtungen ueber einige niedere Algen-

formen. Nov. Act. Acad. Caes. Leop. -Carol. 14: 771. 1829.
'39. Corda, A. C. J. Almanach de Carlsbad 9: 213. 1839.
'45. Hassall, A. H. A history of the British freshwater algae. 1845.
'49. Nageli, C. Gattungen einzelliger Algen. Zurich. 1849.
'55. Braun, A. Algarum unicellularium genera nova et minus

cognita. Lipsiae. 1855.
*67. Reinsch, P. Die Algenflora des mittleren Theiles von Franken.

13 Taf. Nurnberg. 1867.
*82. Lagerheim, G. v. Bidrag till kannedomen om Stockholmstraktens

Pediastreer, Protococcaceer och Palmellaceer. Ofvers. Kgl. Vet.

Akad. Forhandl. 39^: 47. 1882.
'89. De-Toni, J. B. Sylloge algarum. Vol. I. Patavii. 1889.
'93. De Wildeman, E. Quelques mots sur le Pediastrum simplex

Meyen. Bull. Herb. Boiss. i: 412. 1893.
'95. Chodat, R., & Huber, J. Recherches experimentales sur le

Pediastrum Boryanum. Bull. Soc. Bot. Suisse 5: i. 1895.
'97. Schroeter, C. Die Schwebeflora -unserer Seeen (das Phyto-

plankton). Neujahrsblatt Naturforsch. Ges. Zurich. 1897.
'00. Noll, F. Ueber die Korperforrn als Ursache von formativen und

Orientirungsreizen. Sitzungs. Niederrhein. Gesells. Nat. u.

Heilk. Bonn. 1900.
'00. De Wildeman, E. Les algues de la flore de Buitenzorg. 1900.
'01. Chodat, R. Les algues vertes de la Suisse. Beitrage zur

Kryptogamenflora der Schweiz i^. Bern. 1901.
'03. Zacharias, O. Ueber das Vorkommen von Borstenbiischeln

an den Randzellen bei Pediastren. Biol. Centralbl. 23: 593. 1903.
'11. Petersen, J. B. On tufts of bristles in Pediastrum and Scene-

desmus. Bot. Tidssk. 31 : 161. 1911.
'14. Nitardy, E. Zur Synonomie von Pediastrum. Beih. Bot.

Centralbl. 322 : iii. 1914.
*i6. Harper, R. A. On the nature of types in Pediastrum. Mem.

N. Y. Bot. Card. 6: 91. 1916.
*i6. Benedict, R. C. The origin of new varieties of Nephrolepis

by orthogenetic saltation. Bull. Torrey Club 43: 207. 1916.

240 Semi-centennial of Torrey Botanical Club

'i6. Jennings, H. S. Heredity, variation and the results of selec-
tion in the uniparental reproduction of Difflugia corona. Genetics
i: 407. 1916.

'18. Harper, R. A. Organization, reproduction, and heredity in
Pediastrum. Proc. Am. Phil. Soc. 1918.

DETERMINATION OF ACIDITY IN PLANT
TISSUES

By Herbert M. Richards

Barnard College, Columbia University

For many physiological purposes the determination of the
concentration and amount of acids in plant juices is a matter of
importance. Acid formation or acid splitting may affect the
output of carbon dioxide and thereby alter the gas interchange
relations. It is necessary to be informed as to the amount of
its rise and decline to evaluate properly the energy-releasing
processes connected with respiration. It is known also that the
degree of acidity in an imbibed fluid has an important influence
on the hydratative capacity of colloidal systems and therefore
must affect the colloidal melange of protoplasm. Besides this the
concentration of acids in the vacuole has an important bearing
upon the osmotic coefficient of its contents. Without attempting
then to enumerate all the ways in which a knowledge of the acid
content of plant juices may be of physiological importance, it is
evident that the determination of this factor is necessary.

In this very brief communication it is not intended to give even
a partial survey of the problem, or to touch upon the variety of
methods which have been developed for the isolation of definite
acids. For technical purposes many ways have been devised by
which the acids in various fruits and other plant parts can be
determined and extracted in a manner that is satisfactory for the
results required. That these methods may not be applicable to
some physiological problems is not a criticism of their technical
value; but, at the same time, it is questionable in some instances
whether the acids extracted really represent either in quantity
or condition the acids originally present in the living cell.

In the first place, the manner of obtaining the plant juices
may be considered. The common method is by pressure. As
far as concentration is concerned, the juice squeezed out may

241

242 Semi-centennial of Torrey Botanical Club

represent with some accuracy the maximum concentration of the
soluble substances present. Even here, however, there may be
room for error. With the ordinary means at hand, pressure is
not the easiest thing in the world to standardize and it is possible
that identical samples might yield juices of slightly different con-
centration if subjected to different pressures. The speaker is
quite aware that such errors may have crept into his own work, but
feels rather confident from somewhat empirical tests that have been
made that the error cannot be an important one.

Where, however, what may be termed total acidity is desired,
a single pressure no matter how powerful opens the road to serious
mistakes. By total acidity I mean the total acid content of a
given weight of fresh or air-dry tissue. It is here that it is very
difficult to tell in many published accounts how great have been
the precautions not to leave a considerable percentage of acid in
the rejected pulp. It is to be presumed that in most instances
investigators were fully aware of this danger, but rarely is there
any mention made of the procedure used to obviate the diffi-
culty.

In his own work the writer has found that what appears to be
a very close approximation of the actual total acid content may
be obtained by repeated pressure. After the first juice had been
expressed and the press released the remaining pulp is collected
and copiously moistened with water, which it greedily absorbs.
This is then pressed once more and the process repeated until the
final expressed water shows by titration a practically negligible
amount of acid.

All of the samples so obtained are mixed and made up to a
definite volume, an aliquot portion of which is then titrated. In
regard to the number of times this process is repeated each tissue
will no doubt show its own peculiarities. In some very refractory
tissues it might be an exceedingly difficult matter to satisfy oneself
that the last trace of acid was extracted. In the tissues with
which I have worked, notably the succulents, it has been found
that the water which comes from the fourth pressing is nearly
acid-free. Here again it is impossible to lay down a rule, for the
type of press used might influence the result. Only by actual
testing can we be sure that the acid is extracted.

Richards: Determination of acidity

243

By the method outHned we may obtain first a sample of pure
juice to determine concentration and second the total amount of
acid contained in a given weight of tissue. Where total acidity
alone is required the speaker has used a simple method which is
not in any way original but which by repeated test has been shown
to be satisfactory and to yield very consistent results. Briefly
the process is this. A small sample of the tissue is ground in a
mortar with a little water and carefully washed silica sand. This
is then strained through glass wool : the pulp and wool are again
ground and strained and if there is evidence that the tissue is not
finely enough comminuted by this time the process is repeated.
The various samples are then mixed together and filtered into a
graduated flask, care being taken to wash the filter thoroughly.
An aliquot portion of the known volume is titrated as usual.
Before finally rejecting the pulp it should be tested to determine
if it is acid-free. This process may be carried on with considerable
speed and the whole determination may be completed within
fifteen minutes of the time of taking the sample. Time may also
be gained by the use of a centrifuge in place of the filtering.

One thing, perhaps, is evident in the methods outlined, that
is, the quickness with which the processes may be carried on, and
it is on this point I wish to lay especial stress. We know that the
organic acids with which we are dealing in plants are in many cases
highly unstable and that if considerable time elapses between
their extraction and estimation changes may occur that will
influence the final result. Also the acids are easily affected by
any substances that may be added to the solutions. Conse-
quently, it is requisite to titrate the juices as soon as possible
and in as nearly their original condition as possible in order to
obtain results that are significant from a physiological standpoint.
It is in these regards that many of the methods commonly em-
ployed, no matter how useful they may be for some purposes,
are not always available for the study of the activities of the living
organism. For instance, the addition of alcohol, while it may
serve to clear the juice for the purposes of titration, must un-
doubtedly produce the esterification of some part of the acids.
Similarly other chemical substances will not be without their
effect in altering the original acidity. Besides all this, the time

244 Semi-centennial of Torrey Botanical Club

which must elapse in the many filterings and extractions may
allow a chance for partial disorganization of the unstabler acids.

These considerations have influenced the writer in his own
work to sacrifice the clearness of the solution to be titrated
for rapidity of estimation. It is true that from the standpoint of
the chemist the extracts procured are often cloudy and colored
so that the end point is not so sharp as it would be in a clearer
solution but by the use of rather greater quantities of the indicator
than usual and by accustoming the eye to the behavior of the
specific plant juice it is probable that the results obtained are
more nearly accurate than by a method which in the chemical
sense may be more perfect. I am fully aware of the various
objections which the chemist may bring to the procedures as
outlined, and I admit their inadequacies. For the purposes de-
sired, however, they are more suitable than more elaborate ones.

The two greatest difficulties are, first in the color of the
solutions and second in the precipitation of protein substances
when the neutrality point is approached. For the first there is
at present no very good remedy to be suggested. By the selection
of an indicator the color change of which is compatible with
observation in an already colored solution, something may be
done. The color change of the juice often suggests itself as an
indicator and if one were certain of the neutraHty of its end point
it could be used instead of an indicator. As to the flocculation
of colloidal substances on the approach of the neutral point it
may be said that the precipitate is usually white and does not
interfere as much with the color reaction as might be supposed.
Of course in separating out the protein may adsorb some acid,
but since the precipitate does not appear until the solution is
nearly neutral the amount so occluded cannot be large. The addi-
tion of substances like bone-black for clearing and decolorization
is tempting but open to various objections, the most important
of which is that the bone-black may itself adsorb acids.

The writer would welcome the suggestion of improvements in
the procedures outlined, particularly in the matter of a satis-
factory method of obtaining a perfectly clear and colorless extract
for titration purposes. It so happens that the plants which I have
especially been investigating yield juices which are usually fairly

Richards: Determination of acidity 245

colorless and which contain only a small amount of colloidal
substances which flocculate out in a neutral solution. The last
method outlined was used, with considerable success, during the
summer of 191 7 at Carmel, California, when the acidity of a
number of types of the local plants was determined in addition
to that of the succulent forms which were being investigated in
detail. The results were interesting, but too few in number to
warrant publication at this time.

SIX MISUNDERSTOOD SPECIES OF AMANITA*

By George F. Atkinson

Cornell University

Recent monographs of the Amanitas have not lessened the
obstacles in the way of recognizing the species of this difficult
genus. On the contrary, they have introduced certain elements
of confusion. This is evidenced by a compression of the number
of species in regions where there has not been an opportunity of
studying living plants but only dried material has been examined.
In quite restricted localities where intensive studies have been
made there has occurred a lively splitting process resulting in
the multiplication of species based to some extent on trivial char-
acters, the result of environmental and growth influence. I, wish
here to call attention to several of the species which have been
misunderstood.

Among the pure white Amanitas in the eastern United States
there is one species which is easily recognized from all the others
usually on sight, but with certainty after a microscopic examina-
tion. This is Amanita hisporigera.^ It is a species with a white
volva with apical dehiscence and a prominent limb. The pileus
is smooth, viscid, and glistening white. The stem is pure white
and slender. The basidia are constantly two-spored. The spores
are globose or subglobose.

Its nearest ally is a similar white species with four globose
spores to a basidium. This four-spored species is a robust plant,
interpreted in this country by some as Amanita verna, by others as
a white form of A. phalloides.% Amanita hisporigera is dis-

* Illustrated by lantern slides.

•\ Amanita hisporigera Aikxnson. Bot. Gaz. 41: 348. 1906.

% Amanita hisporigera is placed in Amanita phalloides by Murrill (Mycologia 4:
240. 1912; and N. Am. Fl. 10: 70. 1914), although very different from the
typical Amanita phalloides of Europe. Coker (Coker, W. C., The Amanitas of the
Eastern United States. Jour. Elisha Mitchell Sci. Soc. 33: 1-88. pi. i-6q. 1917)
places Amanita hisporigera as a synonym of Amanita "verna.'' This four-spored
white Amanita of North America was also interpreted by me as Amanita verna (see

246

Atkinson: Six species of Amanita

247

tinguished from the robust, white, 4-spored species, not only
by its 2-spored basidia, but by its more slender form. In nearly
all cases one can distinguish it by size alone from small forms of the
robust, 4-spored species, without a microscopic examination.
However, in several hundred specimens I have examined during
the last ten or more years I have not found a single case of a
variation in the 2-spored character of the basidia. It is a very
distinct genetic type and represents a good species. This inter-
pretation is reinforced by the fact that, in several of the genera
of the agarics, there are a number of constantly 2-spored species.

Another species, Amanita cothurnata,^ is interpreted by some
as specifically identical with Amanita pantherina.\ Amanita
cothurnata is entirely white, or, rarely, in some individuals there is
a tinge of umber over the center of the pileus, or now and then
individuals are found with a slight tinge of yellow at the center.
The volva is circumscissile in both species. The white calyptra
is torn into small floccose patches which are distributed quite
regularly over the surface of the pileus. In Amanita pantherina
the pileus is a dark smoky brown, and these white patches on the
dark background are suggestive of the spotted appearance of the
panther, whence the name pantherina. The lower part of the
volva in both species is "cothurnate" or "booted" concrete with
the stem, the "limb" in Amanita cothurnata terminating in a
thick, regular shoulder or roll, like, the top of a closely fitting
buskin. This species differs from Amanita pantherina chiefly
in its color, and in its more slender habit, as can be seen from
these lantern-slide reproductions of photographs.

At maturity the granular content of the oval, or short-ellipsoid
spores usually disappears and is replaced by a large globose oil drop
of about the same dimensions as the transverse diameter of the
spore. This large glistening oil drop is very distinct in contrast
with the nearly transparent, thin, spore wall, which is rather

Studies of Am. Fungi, Mushrooms, Edible, etc., ist edition, p. 60. /. 59. 1900;
2d ed. 1901). It is, however, quite different from the Amanita verna of France
(Agaricus vernus Fr. ex Bull. Champ. Fr. pi. 108) as I have found from specimens
since collected in France.

* Amanita cothurnata Atkinson, Stud. Am. Fungi, Mushrooms, etc., ist edition,
p. 66./. 68, 69. 1900; 2d ed. 1901.

t Coker, W. C. Jour. Elisha Mitchell Sci. Soc. 33: 46. 191 7.

248 Semi-centennial of Torrey Botanical Club

difficult to see. In specimens of Amanita pantherina, which I
have collected in the Jura mountains in France,* the spores at
maturity, or in dried plants, still retain the granular content.
This condition, however, may possibly vary in some specimens,
and the presence or absence of an oil drop in the spores, may not
be so important a specific character as is usually assumed. But
it is worthy of note that in all specimens of Amanita cothurnata
which I have examined, this change in the spore has taken place.
Another witness of the specific distinction of Amanita cothurnata
is its wide distribution, comparative abundance, and constancy
in character, while typical Amanita pantherina is quite rare in this
country according to my observations. Some of the few indi-
viduals which I have found of this species in the United States
raise some doubt as to their specific identity with Amanita pan-
therina of Europe, and resemble strongly in some respects another
American species, Amanita velatipes. Since the latter species is
interpreted by one student as synonymous with Amanita haccata
of Europe, it is readily seen to what an end such indiscriminate
"lumping" would lead.

Several other American species have recentlyf been added as
synonyms to quite a long array of names of European forms,
cumulatively assembled under Amanita jonquillea Quelet and
Amanita haccata Fr. I will discuss here only two of these, which
are not only specifically distinct from each other, but also from
Amanita jonquillea. These are Amanitopsis albocreata,X and
Amanita velatipes.^

Amanitopsis alhocreata is a white species with now and then
individuals showing a yellowish tint over the center of the pileus,
which is striate on the margin. The annulus is absent. The volva
is circumscissile, the calyptra forming white floccose patches on the
pileus, much as in Amanita cothurnata and Amanita pantherina.
The lower part of the volva is ocreate, concrete with the base of
the stem and fits it like the legging of a boot, but it is not so
prominent, nor does it extend so high up on the stem as is usual

* In 1905. The determination was confirmed by E. Boudier.
t See Coker, W. C. Jour. Elisha Mitchell Sci. Soc. 33: 1-88. 191 7.
X Amanitopsis albocreata Atkinson, Jour. Mycol. 8: 11 1. 1902.
§ Amanita velatipes Atkinson, Stud. Am. Fungi, Mushrooms, etc. ist edition,
p. 63./. 64-67. 1900; 2d ed. 1901.

Atkinson: Six species of Amanita

249

in Amanita cothurnata. Sometimes a portion of the thin marginal
area of the calyptra may adhere to the rim of the ocrea, and
thus resemble the volva limb of Amanita jonquillea, which is not
ocreate but sheathing. In the latter species the volva is partly
circumscissile and partly apical in its dehiscence. The volva is
thin and weak. A portion of the calyptra margin remains at the
base as a thin, free, sheathing limb, while the remaining portion
rests on the pileus in the form of floccose patches. But the essen-
tially differential feature in respect to the volva here is that
the lower portion is not ocreate as it is in Amanitopsis albocreata.
This feature can be seen in the lantern views presented here,
from photographs of Amanita jonquillea which I made from speci-
mens collected by me in the Maritime Alps, at Berre-des-Alpes,
near Nice, in 1905 and also in 19 10. The pileus of Amanita
jonquillea is pale yellow, about the color of jonquils, and the
margin is striate. The spores of Amanitopsis albocreata are
globose to subglobose, while those of Amanita jonquillea are
ellipsoid, as shown in these reproductions from photomicrographs
of the spores.

A thin partial veil and an annulus are present in Amanita
jonquillea. The veil is quite thin and sometimes it is so torn
during the expansion of the plant that a distinct annulus is not
present. In rare cases a delicate annulus may be present in
Amanitopsis albocreata. But in my numerous collections of this
species I have not observed one. However, in all species of
Amanitopsis which I have studied in the fresh state, the ground
•tissue is present, which, in many species of Amanita at least,
forms the partial veil and annulus. This has been demonstrated
in Amanitopsis vaginata through a study of the development of
this species.* As I pointed out at that time, the distinction
between the genera Amanita and Amanitopsis is probably not a
natural one. There are several species of Amanita (and Amani-
topsis also) in which the presence of an annulus is variable. The
ground tissue forming the partial veil, though sometimes abundant,
possesses a very low degree of coherence, with the result that some-
times an ephemeral annulus is present and at other times it is
absent. The facts that a distinct annulus is sometimes wanting

* Atkinson, G. F. The development of Amanitopsis vaginata. Ann. Myc. 12:
369-392. pi. 17-19. 1914-

250 Semi-centennial of Torrey Botanical Club

in Amanita jonquillea, that there is a tinge of yellow in the pileus
of some individuals of Amanitopsis albocreata, with variability in
some of the other characters, are responsible for the opinion
expressed as to the specific identity of these two species. Each
species has its own range of fluctuating variation. The fact that
at the extremes of the range of fluctuating variation one or more
of the characters in different species overlap is not evidence of
their specific identity.

Now we come to Amanita velatipes, another species which is
confused with Amanita jonquillea et al. This is a large and robust
species, about equal in size to Amanita muscaria. The pileus is
usually hair-brown, or umber-brown, sometimes with a tinge of
lemon-yellow, or rarely entirely maize-yellow. The remaining
parts are white. The volva is thick and distinctly circumscissile.
The calyptra breaks into concentric rings, especially near the
margin, and these transversely into irregular areolate patches,
which are usually firm and compact. They are easily freed from
the viscid pileus and commonly warp up around the edge and may
thus soon fall away. The lower portion of the volva remains
concrete with the base of the stipe and is often ocreate. But more
commonly the continued elongation of the stem severs it once or
twice more in a circumscissile manner, thus leaving one or two
stout rings above the bulb. These rings are more rarely checked
transversely into coarse warts, in robust specimens, approaching
then the usual condition in Amanita muscaria. The partial veil
is ample and adheres very firmly to the stipe. It is easily freed
from the gills but clings firmly to the margin of the pileus, for a-
time, and as the plant expands the veil is ripped off the surface
of the stipe and forms an inferior or median annulus. The spores
are oboval and inequilateral in profile, and when mature contain
a large oil drop.

It is difficult to understand a concept of species which would
unite Amanita velatipes with Amanita jonquillea. Its relationships
are far closer to Amanita pantherina and Amanita muscaria. The
smaller forms are exceedingly difficult to distinguish from Amanita
pantherina. Or, shall we say that large forms of Amanita pan-
therina are difficult to distinguish from Amanita velatipes? I
have several times collected, in the vicinity of Ithaca, a large
white Amanita which might be taken for a white form of either

Atkinson: Six species of Amanita

251

Amanita velatipes or Amanita muscaria, If Amanita velatipes is
synonymous with Amanita jonquillea, then Amanita muscaria
and Amanita pantherina must be thrown into the same "melting
pot"!

Two more species remain to be discussed at this time, which
have been misunderstood, and consequently united, in the two
recent treatments of the Amanitas in this country. These are
Amanita Frostiana and Amanita flavoconia."^

Amanita Frostiana] is a beautiful species described by Peck in
1883. This description is brief but very accurate and illustrates
in a striking manner not only Dr. Peck's powers of observation,
but his critical and analytical mind. It is a rather unique pro-
cedure, in a work of a monographic nature, in writing a diagnosis
of Amanita Frostiana, to ignore this original description by Peck,
arid as it appears, leave out of consideration altogether any indi-
viduals which represent this species, and deliberately draw, up a
description of Amanita Frostiana from individuals recognized as
Amanita flavoconia, a very different species. J Amanita Frostiana
is a small to medium-sized plant. The pileus is orange or yellow,
and distinctly striate on the margin. The volva is circumscissile
in dehiscence. The calyptra is separated into numerous floccose
patches on the surface of the pileus. The lower part of the volva
is ocreate. The stem and partial veil are yellow. The spores are
distinctly globose.

Amanita flavoconia^ is also a small to medium-sized species.
The colors are much as in Amanita Frostiana. But it differs from
Frostiana in its smooth, not striate pileus, in its smaller, oboval to
subellipsoid spores, and in its volva all friable. The lower portion
of the volva, that which remains over the base of the stem, is
in the form of a fine yellow powder, with not the slightest sugges-
tion of an ocrea. The lower part of the stem broadens gradually
to the bulb, except in rare cases when, under certain unfavorable
environmental conditions, somewhat depauperate forms may
result, in which the transition from the stem to the bulb is abrupt.

* See Murrill, W. A. Mycologia 5: 76. 1913. and N. Am. Fl, 10: 74. 1914.
Coker, W. C. Jour. Elisha Mitchell Sci. Soc. 33: 65. 191 7.

■f Agaricus Frostianus Peck, N. Y. State Cab. Nat. Hist. 23: 69. 1872.

J See Coker, W. C. Jour. Elisha Mitchell Sci. Soc. 33: 65. 1917- But it is
quite possible his specimens were not typical flavoconia, but belong to a different
species.

^Amanita flavoconia Atkinson, Jour. Myc. 8: no. 1902.

252 Semi-centennial of Torrey Botanical Club

But even under these conditions there is no evidence of an ocreate
volva. Moisture frequently holds portions of the calyptra in
definite patches, but its texture is different from that of the volva
in A. Frostiana. In very rare cases, when the weather conditions
are somewhat drying, the margin of the pileus may be slightly
striate, as sometimes occurs with thin pilei, which normally are
not striate. But this rarely appearing striate margin in Amanita
flavoconia is one of the extreme limits of its range of fluctuating
variation, not at all an indication of its specific identity with a
normally and regularly striate species. But aside from this
feature, the very different spores, and volva, separate flavoconia
very clearly from Frostiana. It is more closely related to Amanita
muscaria, or the form sometimes called formosa.

For a number of years, before I made a critical study of
Amanita flavoconia, I regarded the specimens of this species which
I encountered as belonging to Amanita Frostiana. I remember
that in 1902, when collecting and studying fungi for a week, in
company with Dr. Peck in the vicinity of Lake Piseco, in the
Adirondack mountains, I showed him some specimens which I
had collected during the morning, and said: "Here is an unde-
scribed species of Amanita.'' He examined the specimens crit-
ically for a few minutes and then said: "Yes, it is. Heretofore
I have taken it for Amanita Frostiana."" Amanita flavoconia
appears to have a much wider distribution than Amanita Frostiana
has. It is very common in the Adirondacks; in fact, it appears
to be the most common species of Amanita in that region, while
I have never found Amanita Frostiana there, though it is not
uncommon in the Cayuga region, and probably in all of central
and western New York.

The range of fluctuating variation presented by some of the
characters of these, and many other species of Amanita, is such
that one extreme of the range in a species may now and then
show a tendency toward the constant character of the corre-
sponding structure in a related species. In this way the ranges
of fluctuating variations are linked by this touch, or slight over-
lapping, of the extremes of all the species. If this relation of the
ranges of fluctuating variation, between the different species of
Amanita, is interpreted as indicating specific identity, it would
result in reducing all the Amanitas to a single species.

SOME OBSERVATIONS ON THE DEVELOPMENT
OF PERIDERMIUM CEREBRUM

By B. O. Dodge and J. F. Adams

Columbia University
(with plates 4-6)

The form of Peridermium Cerebrum Peck on Pinus rigida in
the vicinity of Lakehurst, and Toms River, New Jersey, has
presented several points of interest. Specimens of this fungus
on P. virginiana from Bedford, Virginia, have been available for
comparison through the kindness of Professor R. A. Harper.

The prevailing type of infection observed in New Jersey
appears as circular or elongated canker-like swellings on trunks
ranging up to eighteen inches in diameter. Where suckers de-
veloped after trees had been cut down, the recent infections appear
as globular or fusiform swellings. The canker-like swellings on
the trunk are the common form of the fungus. Infections are
frequently found at the base of the tree as well as at varying
heights on the trunk, usually below the region bearing branches.
The trunk infections often consist of 'a number of closely associated
swellings, the outermost being smaller and younger developments.
The galls vary in size and are circular or elliptical in outline.
This difference in the shape of the swellings suggests that the
infection may sometimes progress more rapidly in one direction
than in the other. Stewart,* studying the globoid galls of P.
Cerebrum on Pinus Banksiana, is of the opinion that the fungus
spreads quite as slowly vertically as it does horizontally. In
several cases at Lakehurst protuberant elongated swellings were
found developing parallel with the trunk, and were at least six
times as long as they were wide. On the other hand, the invasion
of the host by the fungus is sometimes more rapid peripherally
than it is vertically. The disease in this material obviously

* Stewart, A, Notes on the anatomj' of Peridermium galls. Am. Jour. Bot. 3:
12-22. 1916.

253

254 Semi-centennial of Torrey Botanical Club

progresses by the development of secondary ovoid slightly pro-
tuberant galls, so that very often at least half of the trunk is
girdled. The more central swellings, when they are closely asso-
ciated, are usually dead. The whole trunk is slightly bent at
this point. This bending is due to the inhibition of growth and
the death and decay of the region of original infection. The
typical condition is that of death at the center and proliferation
at the margin. In many cases the entire affected area is dead,
owing to the action of such agencies as fire, insects, and birds.
Borer and woodpecker injuries in several instances were found
to be the cause of the death of the swellings. Weir* has reported
that borers and wood-rotting fungi, entering the burls on Pinus
divaricata, often hasten the decline of the tree.

While the individual swellings or galls may be circular or
elliptical in outline, the total or final effect of the parasite on the
host sometimes is such as to bring about a fusiform enlargement
of the trunk. An example of this type of infection is shown
in PLATE 4, fig. 2, which is from a photograph of a tree at Lake-
hurst. The infected area is about five feet from the base of the
tree. At the widest part of the swelling the trunk is about
eighteen inches in diameter and tapers from this region so that
the diameter is about four inches less at the limits of the swollen
region. Viewed from the side this tree is seen to be bent or
"kneed" in the manner shown in plate 5, fig. i. This figure is
from a photograph of another infected tree from the same region.
There are at least ten separate swellings on the canker shown in
plate 4, FIG. 2. These were outlined with ink on the photograph
so that their limits may be made out more distinctly in the repro-
duction. If we assume that the whole canker is the result of
one primary infection and that the point of infection is now shown
by the presence of the oldest dead gall shown at the center of the
picture, we see how the fungus has spread in all directions. There
is no bark on this central gall and the exposed wood is dried and
cracked. The second gall, just at the right, is somewhat smaller
and circular in outline. This gall is also dead. Above the central
gall is a large ovoid one that is dead, but not in the advanced stages
of decay. At the right and left in this top row of swellings are

* Weir, J. R. Observations on the pathology of the jack pine. U. S. Dept.
Agr. Bull. 212: i-io. 1915.

Dodge and Adams: Peridermium Cerebrum 255

two living galls, the smaller of which is producing aecidiospores.
The larger one at the left was covered with a thick mass of pitch
at this time. On the lower parts of the infected region are five
or six other galls, the larger one being dead while the others are
alive.

The question as to the manner in which the fungus comes to
attack new regions is an interesting one in view of the fact that
we have these separate galls, all evidently resulting from one
primary infection. The spread of the mycelium appears not
to be by gradual encroachment but rather by sudden migration
induced by the conditions that are to bring about or have already
brought on the death of the tissues of the gall. Some light may
be thrown on this question by a study of the specimen figured in
PLATE 5, FIG. I, which is a side view of a portion of a tree ten inches
in diameter at the cankered region. The marginal gall (at the
right in the picture) is alive, but the other two, both furrowed and
denuded, are in advanced stages of decay. A cross section of
this same specimen is shown in plate 6. The tree was plainly
infected when it was very young, evidently in the growing region
of the stem at a point about two feet from the ground. The
wedge-shaped abnormal discolored gall-tissue can be traced to
within three or four rings of the center. By splitting the central
wedge we find that, further down, the infected area approaches
the very center of the tree. The first gall growth has entirely dis-
appeared, owing to decay. The fungus has spread peripherally by
a series of sudden localized migrations. At the right (above) the
first migration occurred about the tenth year, and about the four-
teenth year at the right-center (below). Both migrations resulted
in the formation of large, lobed or furrowed galls, the wood of
which is now discolored and decayed. Just when the other
migrations took place is difficult to determine. At the upper left
corner a distortion of the annual ring is evident at about the
twenty-sixth year. At the lower left, the wedge-shaped band
of soHd, dark-colored wood begins with the eighteenth year and
spreads out gradually for eight years more before this section
shows the beginning of the globoid gall. It has evidently taken
nine years for the upper swelling at the left to develop, although
the larger amount of the characteristic gall tissue has been formed
during the last three years.

256 Semi-centennial of Torrey Botanical Club

Infections sometimes occur at the base of a tree, as shown in
PLATE 5, FIG. 2. This infection has spread peripherally very
rapidly. The dark area, at the right of the center, is dead, but
the bark still adheres. The other lobes of the gall are producing
aecidiospores.

The prevailing type of trunk infection on P. rigida in the pine
barrens of New Jersey is interesting when compared with those
observed by Weir {I. c.) on P. divaricata. He points out that
Peridermium Cerebrum in dry sandy areas confines itself more
generally to the branches, occurring more rarely on the trunk. In
the pine barrens the older, swellings are very rough in appearance.
Several layers of loose scaly bark are usually found adhering.
This is the tissue that is sloughed off after the development of
spermogonia or aecidia. It adheres most strongly at the margin
of the infected areas. The outer younger swellings possess a
smooth tan-colored layer of cork tissue.

In cross sections of the trunk the dark wood of infected areas
contrasts so markedly with the healthy wood that the time at
which infection took place can be determined fairly accurately.
In eight trees that were cut where the disease was restricted to
the trunk, it was found that infection had occurred when the
trees were from one to four years of age. Cross sections of
typical globoid galls on P. virginiana from Bedford, Virginia,
show that infection usually takes place during the first year's
growth. Stewart (/. c.) has stated that infection on P. Banksiana
usually, if not always, occurs during the first year's growth of the
branch. Where it is possible to trace the annual rings of growth
in P. rigida through the infected and uninfected regions, it is
found that about twice as much wood is formed in the diseased
region as in the healthy.

The mycelium is uninucleated and its intercellular develop-
ment is abundant in the cortex. The hyphae appear to follow
the medullary rays in the cortex as well as in the wood where the
mycelium is more sparingly developed.

Haustoria are commonly developed in the cells of the medullary
rays. They are not exceptionally large and have the usual con-
striction where the cell wall is penetrated. They are of about the
same diameter as the hyphae from which they originate and are

Dodge and Adams: Peridermium Cerebrum

257

sometimes found to be adjacent to the nucleus. Occasionally two
or three haustoria are found in the same cell. The cells of the
cortex are not attacked by haustoria as frequently as are those
of the phloem and medullary rays. Living hyphae with haustoria
are found in wood tissue several years old.

We were not fortunate enough to observe the exudation of
spermatia in the New Jersey material. Sections of material
showed the presence of spermatiophores bearing spermatia as
early as March 25. They form a palisade layer that appears to
be spread over indefinite areas of considerable extent like a caeoma

type of fructification. This
palisade is formed beneath four
or more layers of newly devel-
oped cork cells. The spermati-
ophore primordium consists of
a compact mass of uninucleated
cells of mycelium situated just
above the outer row of cortical
parenchyma cells. Below this

the mycelium is sparingly de- Fig. i. Section through spermogorial
veloped as compared with what primordium. c. cork; plectenchyma;
^ , , , , . . cp, cortical parenchyma.

we find below the aecidmm pri-
mordium. The Virginia material was more favorable for the study
of the spermatial layer. This material consists of the typical glo-
boid swellings as described by most investigators of P. Cerebrum.
The galls are fairly smooth in appearance compared with the New
Jersey material. The development of the palisade layer of sper-
matiophores bearing spermatia was first observed in sections of
material collected February 9. The primordium develops between
the cortex and the cork layer as shown in text-figure i. The
overlying cork layer is smooth in appearance. Specimens were
placed in moist chambers and within twenty-four hours exuda-
tions of spermatia appeared. The cork becomes irregularly
cracked so that the spermatia exude in yellowish droplets. There is
no special aperture through the bark for the escape of the spermatia ;
they ooze out as sticky exudations through cracks naturally
formed by the growth of the gall. On removing the overlying
cork a yellowish crust-like layer of spermatiophores is exposed.

258 Semi-centennial of Torrey Botanical Club

This layer is continuous over the gall except where interrupted
by small irregular patches or strips of cork that can not be re-
moved easily. A burl with portions of the cork removed is shown
in PLATE 4, FIG. I. The irregular patches of thin cork are out-
lined to bring out by contrast the smooth, shining spermatiophore

surface. A continuous area half
an inch square can frequently be
uncovered, disclosing the crust-
like spermatial surface. Micro-
tome sections three quarters of
an inch long have been made,
showing a continuous palisade
layer. A small portion of a sec-
tion through a matured sper-
matial layer is shown in text-
figure 2. We have not ob-
served in our sections that this
effused palisade of spermatio-
phores is limited by a definite
marginal system of sterile cells.
6permatial primordia frequently
extend from the margin of ma-
tured areas as a plectenchyma
of hyphae between the cork and
cortical parenchyma as illus-
trated in TEXT-FIGURE I. The
cortical cells immediately below
the layer of spermatiophores are
not spread apart by the hyphae
as conspicuously as are those be-
low the aecidium. We have not
seen in any instance spermatial
hyphae developing in the tissue overlying that in which the
aecidia are being formed. Cross sections of the Virginia material
developing both spermatial and aecidial fructifications on the
same gall show that there is no sharp line of demarcation be-
tween the two. In one burl there was a space of only 700 ^
separating them. The nature of the spermogonia of P. Cerebrum

Fig. 2. Section through the cortex
showing the spermatial layer. 5, scleren-
chyma; c, cork; sm, spermatia; sp, sper-
matiophores; b, basal tissue; cp, cortical
parenchyma.

Dodge and Adams: Peridermium Cerebrum 259

has been noted by Arthur and Kern
that the spermogonia are not definitely
deHmited units. The spermatia are de-
veloped from an extensive palisade layer
of spermatiophores spreading out in-
definitely over the surface of the galh
thus producing a typical caeoma-like
structure. Our conception of the mean-
ing of the terms spermogonium and
pycnium must be broadened if we are to
use either of them in describing this
structure.

In the New Jersey material it has
not been possible to determine with ac-
curacy very long in advance those swell-
ings which will develop aecidia. The
tissue in which aecidia are developed is
usually sloughed off by the following
spring. It appears as a dry corky layer,
the surface of which possesses the cere-
broid outline, due to the aecidial scars.
The aecidium primordium has been ob-
served in cross sections of material as
early as April 29. At this time it. ap-
pears as an extensive, deep-seated yel-
lowish layer in the cortex, where it can
be easily recognized. The cells of the
cortex in the region of the primordium
are conspicuously separated by the
abundant development of the vegeta-
tive mycelium. The outer two or three
rows of cortical parenchyma cells are
pushed outward by the primordium.
The relation of the primordium to the
host tissue may best be understood by
referring to text-figure 3. This figure
is drawn from a section of the cortex in

* and others. We find

Fig. 3. Section through
aecidium primordium from ma-
terial collected at Bedford, Vir-
ginia, s, sclerenchyma; c, cork;
cp, cortical parenchyma ; v,
vegetative hyphae; p, pseudo-
parenchyma; g, gametophoric
hyphae; /, fusion cells; b, basal
tissue.

* Arthur, J. C, & Kern, F. D. North American
Mycologia 6: 109-138. 1914.

species of Peridermium on pine.

260 Semi-centennial of Torrey Botanical Club

which the primordium is developed. Only two rows of overlying
cortical parenchyma cells are shown in this section. The number,
however, varies and there may be as many as four or five layers.
Above these outer parenchyma cells {cp) there are from one to
four layers of flat, thin-walled cells (c) and beyond these one to
four layers of large sclerenchyma cells {s). At the base of the
primordium there is a compact mass of interwoven hyphae (h)
from which parallel rows of cells originate. These are the gameto-
phoric hyphae {g) which are eight or more cells in length. The
fusion cells (/) in these chains are recognized by their being deeply
stained in the preparations. Beyond the gametophoric hyphae, in
this stage, we find a considerable development of pseudoparen-
chyma {p). Above the pseudoparenchyma the vegetative hyphae
{v) are shown pushing in between the cells of the outer layers of
cortical parenchyma {cp). The aecidium has its origin slightly
deeper than the spermogonium. In the spring of 1 916 at Lake-
hurst, N. J., the matured aecidia were first observed on May 21.

In no instance have we discovered spermogonia and aecidia
following each other on identical areas of the same gall.
Certain galls were found developing only aecidia, others only
spermatia. In the Virginia material it was found in several
instances that both developed on different parts of the same
gall. This would indicate there is an alternation of the
aecidium and spermogonium as reported by Hedgcock and
Long* and others. In the large canker-like swellings of
the New Jersey material we have not found galls bearing
both aecidiospore and spermatia galls. We have found a few
cases of infection on P. rigida in New Jersey with the swellings
still bearing the rough, scaly bark showing plainly aecidial scars
in April. When this was removed we found directly beneath,
separated from it by a few layers of new cork, aecidium primordia.
This may have been due to the possibihty that the old cork
layers were not shed the previous year, that is, at the time
spermatia were developed.

Seedling oaks of Quercus ilicifolia and Q. marilandica were
found near Lakehurst, New Jersey, with mature uredosori as early

* Hedgcock, G. G., & Long, W. H. Identity of Peridermium fusiforme with
Peridermium Cerebrum. Jour. Agr. Research 2: 247-249. 1914.

Mem. Torrey Club

Volume 17, plate 5

PERIDERM I UM CEREBRUM Peck

Mem. Torrey Club

Volume 17, plate

PERIDERMIUM CEREBRUM Peck

Dodge and Adams: Peridermium Cerebrum

261

as June 3. These seedlings were located within two feet of an infec-
tion at the base of a tree. The first collection of teleutosori was
made on July 4 at Toms River. We have conducted infection
experiments with this form of P. Cerebrum in the pine barrens of
New Jersey and have obtained infections on these two species
of oak and on Q. heterophylla.

explanation of plates 4-6

Plate 4

Fig. I. Peridermium Cerebrum on Pinus virginiana, Bedford, Virginia, April 9,
191 7. Portions of the thin cork layer have been removed at the center and above
to show the extensive spermatial crust (portions outlined with ink). Natural size.
About half of the surface shown has the spermatial layer exposed. The wavy out-
lines marking the boundaries of the exposed areas do not necessarily indicate the
limits of the spermatia-bearing region. The very remarkable extent of the fertile
layer is brought out strikingly. The small unpeeled patches at the center and
above were at this time, at least, sterile regions, the significance of which is not
very clear. Morphologically we should probably interpret the whole surface of the
gall as one continuous spermogonial crust. The sterility of certain regions is
doubtless accidental.

Fig. 2. From a photograph of a tree at Lakehurst, New Jersey, May 21, 1916.
This shows how the fungus spreads by a series of migrations, giving rise to a number
of associated galls. The original point of infection is shown by the oldest dead gall
at the center. Three other dead galls are seen adjacent to it, above, at the right,
and below. About the periphery of the canker are six or seven living galls. The
galls were outlined with ink on the photograph. Tree 18 inches in diameter.

Plate 5 .

Fig. I. Peridermium Cerebrum on Pinus rigida, Toms River, New Jersey, 1916.
The prevailing type of infection in this region causes a characteristic bending of the
trunk. Infection beginning at the left has spread peripherally to the right. The
galls at the left and center are dead. The one at the extreme right is living. Tree
10 inches in diameter at this point. (Cross-section of this specimen is shown in
PLATE 6.)

Fig. 2. Peridermium Cerebrum on Pinus rigida, Toms River, New Jersey,
June, 1916. Infection at base of tree about 12 inches in diameter. The infection
spread peripherally. The dark area just at the right of the center is dead, but
the bark still adheres. The other lobes of the gall are producing aecidiospores.

Plate 6

Cross-section of the infected region of a trunk ten inches in diameter (a
surface view of the same specimen is shown in plate 4, fig. i). Infection beginning
at about third year, as shown by death of the wood in the right central part, has
spread peripherally in both directions. The first gall has entirely disappeared owing
to decay. On either side can be seen the remains of other galls that were formed
with the first migrations of the fungus. At the left are two living galls.

THE VEGETATION OF THE HEMPSTEAD

PLAINS

By Roland M. Harper

College Point, New York

WITH PLATE 7

Contents

Introduction 262

Environment 264

Area and topography 264

Geology and soil , 265

Climate 267

Vegetation 268

Habitats 268

Aspects 268

Influence of fire 270

Plant census 272

Methods of investigation and treatment 272

Upland vegetation 273

List of plants 273-275

Dissemination, etc 275

Rate of growth 275

Retrograde succession? 276

Valley or meadow vegetation 276

List of plants 277

Dissemination, etc 278

Weeds 278

Comparisons with other regions 279

Miscellaneous phytogeographical data 282

Destructive influences 283

Introduction

The Hempstead Plains, in the central part of Nassau County,
Long Island, is a bit of prairie similar in aspect to parts of the
Great Plains, and appearing quite out of place on the Atlantic
seaboard. Its general geographical features were described by
the writer a few years ago,* with a very superficial account of the

* Bull. Am. Geog. Soc. 43: 351-360./. 1-3. May, 1911. Reprinted in abridged
form, with a different set of illustrations, in Torreya 12: 277-287. /. j-7. Dec.
1912. These contain references to some earlier publications which do not need to
be cited again here. See also New Internat. Encyc, ed. 2, 11: 133. 1915.

262

Harper: Vegetation of the Hempstead Plains 263

vegetation. The present communication describes the vegetation
more fully, but does not attempt a complete enumeration of the
flora, which could very well constitute a separate paper of con-
siderable length. Facts previously published will not be repeated
here except where necessary for the continuity of the discussion,
for the earlier papers are quite accessible.

Although this unique eastern prairie was mentioned in a few
early histories and books of travel, and was well known to several
local botanists a generation ago as a good place to collect certain
species of plants, it was overlooked by all students of vegetation
(as distinguished from flora)* until a very late date, when at least
three fourths of it had already been obliterated. In a sketch of
the fauna and flora of the neighborhood of Cold Spring Harbor
by Dr. C. B. Davenport (the flora part contributed by Dr. D. S.
Johnson), published in Science for Nov. i8, 1898, for the purpose
of showing the attractions of that locality for botanists and
zoologists, there is no hint of the existence of a natural prairie,
with its many interesting ecological problems, within five miles
of the Biological Laboratory (and plainly visible to any one
coming out there by train from New York). And for nearly
ten years after that none of the botanists or ecologists who
attended the summer school at Cold Spring Harbor as instructors
or students seem to have known of this prairie, although some of
them had lived or studied in Chicago and should have had some
acquaintance with prairies. f

In Jelliffe's Flora of Long Island, 1899, there is no mention of

* See Torreya 17: i. 1917.

t There is much to be said in extenuation, however, and my own recognition of
the unique character of this area was almost as tardy. I had read about the Hemp-
stead Plains in the government soil survey report on western Long Island, by J. A.
Bonsteel, in the spring of 1905, and visited Cold Spring Harbor once that year and
twice the next, and walked a few times along the western and southern edges of the
Plains as mapped in the soil survey report, without noticing anything unusual, until
on July 3, 1907, I happened to cross the middle of the area on the way from the
Merrick cedar swamp to Hicksville; and the facts were then irresistible. On my
previous walks I had passed through only those parts where the original vegetation
had been completely destroyed, and the portions visible from the railroad I had
probably mistaken for abandoned fields, never having seen a real prairie before.
Since 19 13 I have obtained much valuable information about this area from Mr.
Henry Hicks of Westbury, as did Dr. Bonsteel ten years before, and several subse-
quent explorers.

264 Semi-centennial of Torrey Botanical Club

the Hempstead Plains, but Hicksville, which is in the heart of the
area, is cited as a locality for about twenty species, collected by Dr.
G. D. Hulst. (A few of these are introduced, but the majority
are typical prairie plants.) The first specific mention of the
Plains in botanical literature that has come to the writer's notice
is a rather indefinite one in a short paper by William L. Fisher
on Long Island violets in the Plant World (3: 91-92) for June^
1900. More explicit is a paper by James Kirby on "Some plants
of Hempstead Plains" in the American Botanist (7: no) for
December, 1904 (pubHshed in May, 1905), which enumerates 14
species; about one third of which, however, do not properly
belong to the prairie flora. In Torreya (6: 213) for October,.
1906, a few species found in the same area by the Torrey Club
excursionists on Sept. I are mentioned. In Dr. Harshberger's
Phytogeographic Survey of North America (1911), page 421, is
probably the most complete list of Hempstead Plains plants,
published up to that time, based on a walk of several miles through
the area with the writer on Aug. 25, 1909.* By 1913 this prairie
was sufficiently well known to plant sociologists to be featured
as one of the attractions for the International Phytogeographic
Excursion, most of the members of which visited it on July 27
of that year. Since then it has been on the regular field program of
the summer classes in botany at Cold Spring Harbor. Taylor's.
Flora of the Vicinity of New York (191 5) devotes nearly a page
(29-30) to this area, and farther on, in the catalogue, eight species
are recorded from the Hempstead Plains, besides a few weeds,
from Hempstead.

Environment

Area and topography. The area originally treeless was about
fifty square miles, corresponding approximately with the Nassau
County portions of the "Hempstead loam" and "Hempstead
gravelly loam" as mapped in the government soil survey. (There
seems to be no evidence that the areas of "Hempstead loam" in
Kings and Suffolk counties were ever prairie.) By 1907 the area
of natural vegetation had been reduced to about ten square miles,
and probably at least a tenth of that has been destroyed since.

*A more extended account appears on pages 170-171 of "The vegetation of
the New Jersey pine-barrens," by the same author (1916).

Harper: Vegetation of the Hempstead Plains 265

Although the soil is not particularly fertile, the proximity of New-
York City makes truck farming more or less profitable under
adverse soil conditions, and also causes large areas to be used for
residential purposes irrespective of soil.

The topography is nearly flat, as in many other prairies, but
the surface has a southward slope of about 15 feet per mile,
which is rather steep for a prairie, though almost imperceptible
to the eye. Several shallow valleys traverse the area in a general
north and south direction, and a few of these are long and deep
enough to have small permanent streams in them. The western
slopes of the valleys are nearly always steeper than the eastern,
possibly on account of the deflective effect of the earth's rotation,*
though the amount of erosion since the glacial period must be
very small.

Geology and soil. The whole area is underlaid by a mixture
of coarse sand and siliceous pebbles, supposed to represent a
glacial outwash deposit, the terminal moraine being just to the
north. At any rate, it is very recent geologically. The ground-
water level averages perhaps 30 feet below the general level of the
uplands, which explains the dryness of most of the valleys and
some of the peculiarities of the vegetation.

The soil proper is very characteristic, consisting of brownish
silty loam covering the gravel to a depth of about a foot, except
in the valleys, where it is thin or wanting. A mechanical analysis
of a sample representing the uppermost 10 inches, from two miles
northeast of Hicksville, is reported in the government soil survey
as follows :t

Per cent

Gravel (2-1 mm.) 2.70

Coarse sand (1-.5 mm.) 8.06

Medium sand (.5-.25 mm.) 3.96

Fine sand (.25-1 mm.) 4.88

Very fine sand (.1-.05 mm.) 8.96

Silt (.05-005 mm.) 49.20

*See G. K. Gilbert, Am. Jour. Sci. 127: 431-432. 1884. Collier Cobb, Jour.
Elisha Mitchell Sci. Soc, 10: 26-32. 1893. C. F. Brooks, School Sci. & Math.
17: 517-521. 1917-

t All particles exceeding 2 mm. in diameter are discarded in these mechanical
analyses, which probably does not make much difference in this particular case, but
would make a great difference in the corresponding subsoil. For this reason the
subsoil analysis given at the same place is not worth copying.

266

Semi-centennial of Torrey Botanical Club

Clay (.005-.00001 mm.)
Organic matter

22.20

8.26

Total

108.22

This has a higher percentage of organic matter than any other
soil thus analyzed in the same report,* but this may mean merely
that most of the other samples were taken from cultivated land,
where the humus was long ago exhausted, for the virgin forests in
the northwestern part of the island certainly have plenty of
humus. Curiously enough, of all the mechanical analyses pub-
lished for Long Island soils in the work mentioned, the one that
matches this most closely is that of the "Galveston clay" (salt
marsh) from two miles northeast of Far Rockaway. In fact the
two analyses do not differ any more than two different ones of
the same type of soil might be expected to. Whether or not this
indicates that our prairie was once a salt marsh it is impossible to
say; but, if it was, the surface must have undergone considerable
tilting since, to give the Plains a southerly slope of one in 350;
and it would not be very easy to explain why the prairie is sepa-
rated from the present salt marshes by several miles of forest.
It is possible also that some if not most of the soil has accumulated
as dust in the course of centuries; but if that were the case it
would be difficult to account for the absence of a dust layer in
the surrounding forests, whose topography is very similar, and in
many other level regions. Although the origin of the soil is not
a botanical problem, this particular type of soil is so closely
correlated with the prairie vegetation that one cannot help puzzling
over it. No satisfactory explanation is available at the present
writing, however.

A partial chemical analysis was reported in the first paper
cited herein, and no additional information on that point has been
obtained since. The amount of potash, one of the most important
constituents, is entirely unknown. As elsewhere in the western
half of Long Island, the soil fertility seems to increase a little
toward the west, if the vegetation is a safe guide.

The small areas of bare ground between the tufts of herbage

* Dr. Hilgard found only i per cent of humus in a sample carefully selected by
the writer about a mile southeast of Hicksville.

Harper: Vegetation of the Hempstead Plains 267

are generally covered with minute lichens* and an occasional
patch of moss, which presumably indicates that earthworms are
rare or absent, for if they were at all common the earth brought
up by them and deposited on the surface would tend to bury
these very slow-growing plants. The lack of earthworms is
probably due to the fact that the loam layer is shallower than the
depth to which the ground freezes in winter, and the worms if
present would hardly descend into the gravel to hibernate. There
are a few ants, but their hills are not numerous enough to interfere
seriously with the ground lichens.

On account of the porosity of the subsoil every heavy rain
must carry down into the ground some of the soluble salts, thus
making the soil progressively poorer. And this tendency cannot
be counteracted to any considerable extent by capillarity, on
account of the depth of the water-table, or by the soil fauna
(as it seems to be in some other placesf), for the ants, etc., prob-
ably do not go down into the gravel much.

Climate. The climate is cool-temperate, but with a long
growing season on account of the proximity of Long Island Sound
and the Atlantic Ocean, neither of which is more than ten miles
away. There are no weather stations on or very near the Plains,
but if we take the average of the data for New York City, Setauket,
and Brookhaven we will probably be not far wrong. The New
York records were taken from 1826 to 1864 at Jamaica, which is
within ten miles of the west end of the Plains, and since then on
Manhattan Island, at a gradually increasing altitude as the build-
ings became taller. The Brookhaven records cover the period
from 1864 to 1882, and those for Setauket from 1886 to 1909.
The data given are the average temperature, in degrees Fahren-
heit, and precipitation, in inches, for each month and for the whole

year.

Months Temperature Precipitation

January 30.4 3.70

February 30.0 3.81

March 37.3 4.31

April 47.9 3-59

* Mostly Cladonia symphycarpa epiphylla, according to Mr. R. S. Williams,
who identified a specimen for me recently. The moss is mostly Polytrichiim juui-
perinum.

t See Ann, Rep. Fla. Geol. Surv. 7: 147. 1915.

268 Semi-centennial of Torrey Botanical Club

May 58.7 3.69

June 67.8 3.02

July 73.3 4.04

August 71.7 4.14

September 65.5 3.47

October 54.8 3.84

November 44.1 3.85

December 34.5 3.83

Annual 51.3 45.28

The average growing season, or period free from killing frost,
is from April 10 to November 7, 211 days. In this respect our
area compares favorably with some places five hundred miles
farther south, say in Georgia and Alabama. June is the driest
month by a small margin, but it would be hard to find a place
with a more evenly distributed precipitation. There are no accu-
rate data on wind, sunshine, evaporation, humidity, or snowfall;
but the average annual amount of the last is probably something
like two or three feet.

Vegetation

Habitats. The natural vegetation may be divided into two
habitat groups: that of uplands and that along watercourses.
In a more detailed study the gravelly slopes of the valleys with
the beds of the dry ones might make a third group, but the vegeta-
tion of such places, though differing a little in composition, is so
similar in aspect to that of the level uplands that it is hardly
worth while to separate it. There is also a characteristic weed
vegetation along roads and in abandoned fields, which will be
discussed briefly farther on. The upland vegetation is by far
the most extensive, but that of the valleys is (or was) a little
richer in species. The next few pages will deal with the natural
upland vegetation exclusively, unless otherwise indicated.

Aspects. The prevailing aspect of the vegetation is a moder-
ately dense growth of coarse grasses and other herbs, averaging
about two feet tall, with a sprinkling of shrubs of about the same
height, and a few trees, either solitary or in small open groves.
All the woody plants are most abundant eastward, except the
commonest shrub, which is pretty uniformly distributed. There
are no stout broad-leaved herbs like the Silphiums of the Middle
Western prairies, but on the other hand there are few evergreens

Harper: Vegetation of the Hempstead Plains 269

or succulents. The prevailing color of the herbage in summer is
grayish green, on account of the prevalence of glaucous and
canescent leaves,* but the scene is brightened by flowers of various
colors, changing from month to month as in many other grass-
lands and the southeastern pine-barrens. In the fall, the color
gradually changes to light brown, with a strong shading of gray
from the plumose spikelets of the prevailing grass (which grows
on nearly every square foot of upland, and makes up something
like three fourths of the total herbaceous vegetation).

Fig. I. Typical prairie scene about miles northeast of Hicksville, looking
south. Andropogon and Bapthia in foreground, farm-house and a few shade-trees
in middle distance, and edge of forest barely visible on the horizon. 2:05 p.m.,
Oct. 20, 1907.

On account of the small size of this prairie one could stand
at any point on the upland and see the surrounding forests in
every direction if buildings and shade-trees did not interfere;
but in the western part one can step down into one of the shallow
valleys and get an absolutely treeless horizon in some directions,
indistinguishable in a photograph from some places far out on
the Great Plains. f

* In this connection see Wiegand, Bot. Gaz. 49: 430-444. 1910.
t See PLATE 7, and compare this with a scene in western Kansas published in
Bull. Am. Geog. Soc. 40: 338. June, 1908.

270 Semi-centennial of Torrey Botanical Club

Fire. Fire seems to be a normal environmental factor in this
prairie, as in all others more than a few acres in extent,* but just
what its normal frequency may have been in prehistoric times
it is impossible now to determine. Nearly all the herbs and shrubs
have thick or matted subterranean rootstocks, so that they sprout
up again readily after a fire. Plants with barbed fruits (which
are most effective in dissemination if they remain on the plant
for several months) seem to be entirely absent, and shrubs with
nuts or berries generally grow in clumps, whose centers are thus

Fig. 2. Edge of oak grove near the railroad, about two miles north of Hicksville,
looking S.S.E., showing Quercus Marylandica, Q. stellala, and the herbaceous vegeta-
tion characteristic of dry prairies. 12:20 p.m., Oct. 20, 1907. The trees evidently
have, and need, little or no protection from fire running through the grass. (The
view in Torreya 12: 282 was taken from the same point.)

protected from fires of moderate intensity. Woody vines (all
of which seem to be sensitive to fire) are scarce, and chiefly
confined to the clumps of bushes in the eastern part and to the
valleys. The commonest tree on the Plains, the gray birch, is
often partly protected by a cluster of shoots around its base, and
it is rather short-lived anyway, so that young trees spring up as

* See Plant World 20: 60. "Feb." 1917.

Harper: Vegetation of the Hempstead Plains 271

fast as the older ones are killed by fire. The pines and oaks
scattered over the eastern part of the Plains are of species not
very sensitive to fire, so that they do not need to grow in dense
groves for protection, like the prairie groves of the Mississippi
valley.* The original boundary between prairie and forest here
has been almost entirely obliterated by cultivation, but it was
probably rather sharp in most places, for the regular forest trees
of Nassau County, both north and south of the Plains, are mostly
of species not very tolerant of fire, and the fires probably stopped

Fig. 3. Portion of pine grove ("Island of Trees") in prairie about a mile
southwest of Central Park, showing especially Pinus rigida and Baptisia tinctoria.
2:47 p.m., Aug. 25, 1909. In the absence of shrubby undergrowth this place differs
notably from the pine-barrens of Long Island and New Jersey, and resembles some
of those in the southeastern states. (For another view of the same grove see Bull.
Am. Geog. Soc. 43= 359-)

abruptly at the edge of the forest, where the shade kept the humus
too damp to burn readily.

It seems rather strange for the ground to be covered with
lichens and sprinkled with mosses in an area subject to ground
fires, for these plants are supposed to be very sensitive to fire;
but probably any one spot on the uplands does not get burned
over more than once in two years, on the average. And the
commonest lichen is so minute and close to the ground that fire

* See Gleason, Bot. Gaz, 53: 38-49. 1912; Torreya 13: 173-181. 1913.

272 Semi-centennial of Torrey Botanical Club

jumping from one tuft of grass to another may pass over it without
doing much injury, and the mosses and fruticose lichens are
mostly in gravelly places, where the vegetation is too sparse to
make much of a blaze.

The vegetation of the wet valleys seems to be practically
exempt from fire.

Plant census. The approximate relative abundance of the
species has been ascertained by a rapid reconnoissance method
which is a crude modification of Clements's quadrant method.
I have traversed the area on foot repeatedly in every direction
(mostly in the .summers and falls of 1907 to 1909, with a few
additional observations made in passing through in 1916 and 1917),
and in so doing have stopped every few yards or rods and jotted
down the name of every plant in sight, indicating relative abund-
ance by a somewhat arbitrary scale. When hundreds of such
little lists are combined they ought to give the relative abundance
(combined with size and duration) of the species pretty accurately,
for the largest and most abundant species of course are noted
oftenest. Herbs which are recognizable only during a brief period
when they bloom naturally do not figure as largely in the returns
as the more lasting ones, but that is all right, for the ephemeral
species do not take as much water, etc., from the soil and make as
much hay as the others.

On account of the difficulty of making proper allowance for
plants of different sizes, ranging all the way from lichens to trees,
and the great preponderance of one species among the herbs,
I have not ventured to assign percentages to the several species.
But when the percentages are finally worked out and arranged in
numerical order they will probably make something like a geo-
metrical progression, for in all areas of natural vegetation that are
large and homogeneous enough there seem to be many more small
and rare species than large and abundant ones; just as in human
society there are always more insignificant people than celebrities,
more poor men than millionaires, or more small towns than large
cities.

In the following lists trees, shrubs, herbs, and cellular crypto-
gams are separated, and arranged in order of abundance in each
group, as usual. A few of those seen least often are omitted, on

Harper: Vegetation of the Hempstead Plains 273

account of the considerable possibility that they may have been
introduced, or wrongly identified. The names of evergreens are
in heavier type, and those of a few species of weedy tendencies,
which may not have been in the prairie in prehistoric times, are
put in parentheses.

The nomenclature is in most cases identical with that in
Taylor's Flora of the vicinity of New York (191 5); and where it
differs from that it conforms with other easily accessible works.
After the name of each species is put the numbers of the months
in which it normally blooms, the prevailing color of its flowers
(replaced by a dash in the case of wind-pollinated species which
have no organs for attracting insects), and a symbol indicating
the mode of dissemination, when known. Wind-disseminated
species (including tumble-weeds) are indicated by Y, tonoboles*
(i. e., plants with capsules or firm cup-like calyces borne on stiff
stems which stand up through the winter) by T, berries and nuts
by O, and pods which discharge their seeds by elastic force by E.
One could go still farther and have symbols or abbreviations for
annuals and perennials, the Raunkiaerian growth-forms, various
types of leaf, etc.,t but it is just as well not to undertake too much
at one time, and some of these matters — as well as the authors*
names, common names, phaenological curves, etc. — can very well
be deferred to a more exhaustive study of the flora.

The first list is for uplands and dry valleys.

Trees

Betula populifolia 5 Y

Quercus marylandica 5 Q

Quercus stellata 5 Q

Pinus rigida 5 Y

Shrubs

Pieris Mariana 5-7 white T

Salix tristis (?)t 4 ■ Y

Quercus prinoides 5 Q

(Populus tremuloides)% 4 Y

* See Clements, Bot. Surv. Neb. 7: 47. 1904.

tSee Ann. N. Y. Acad. Sci. 17: 36-38 (1906) for a more elaborate method of
treating plant association lists.

X This could just about as well be S. humilis. No one seems to have succeeded
in drawing a sharp line between the two forms.

§ This is normally a small tree, but on the Hempstead Plains it seldom gets
more than three or four feet tall, perhaps on account of the frequent fires.

274

Semi-centennial of Torrey Botanical Club

Gaylussacia haccata 5 pink O

Comptonia peregrina 5

Rhus copallina 7-8 yellow O

Myrica carolinensis 5 Q

Quercus ilicifoUa 5 O

Corylus americana 4 O

(Rubus cuneifolius) 5-7 white O

Herbs

Andropogon scoparius 8-9 Y

lonactis linariifolius 9-10 blue Y

Baptisia tinctoria 6-9 yellow Y

Aster dumosus strictior 9-10 white Y

Crocanthemum sp.* 5-8 yellow

Cracca virginiana 6 cream and purple E

Aletris farinosa 6-7 white T

Viola pedata 5 blue E

SoUdago puberula{?)f 9-1 1 yellow Y

Antennaria neglecta 4-5 white Y

AgaUnis acutaX 9 pink-purple T

Lespedeza capitata sericea 8-9 cream 5

Juncus Greenei 6 green T

Lechea villosa 7-8 dark purple T

(Euthamia tenuifolia) 9-10 yellow Y

Scleria pauciflora 6-7

Sorghastrum nutans 9 •

Sisyrinchium sp 5-6 blue

Linum intercursum% 7-8 yellow T

Poly gala Nuttallii 7-9 pink

Hypoxis hirsuta 5-6 yellow

Lechea maritima 7-8 dark purple T

(Potentilla canadensis) 5 yellow

Sericocarpus linifolius 7-8 white Y

(Agrostis alba?) 5-6 Y

Poly gala viridescens 8-9 pink

Carex pennsylvanica 4-5

Lespedeza angustifolia 8-9 cream T

Eupatorium hyssopifolium 8-10 white Y

Viola fimbriatula 4-5 blue E

Cirsium discolor^) 6-7 pink-purple Y

P oly gala poly gama 6-8 pink-purple

Andropogon furcatus 9 Y

* There may be more than one Crocanthemum (long known as Helianthemum) .
in which case both would take a lower rank in the list. See Bicknell, Bull. Torrey
Club 40: 613-615. 1913. Fernald. Rhodora 19: 58-60. 1917.

t Some 5. nemoralis may have been included with this.

% Described since Taylor's Flora, in Bull. Torrey Club 42: 338-340. June, 1915.
Formerly referred to Gerardia decemloba, a species of more southerly range.

§ Bicknell, Bull. Torrey Club 39: 418. 1912. Previously confused with L.
medium or L. floridanum. See also C. A. Weatherby, Rhodora 18: 224. 1916.

Harper: Vegetation of the Hempstead Plains 275

Bartonia virginica

Viola lanceolata

Houstonia longifolia

(Sarothra gentianoides) . . .
Antennaria plantaginifolia

Comandra umbellata

Ihidium gracile

8-9 cream

5 blue E

5-6 pink-purple T

8-9 yellow

4-5 white Y

5-6 white
8-9 white

Cryptogams

Cladonia symphycarpa epiphylla (and others)
Polytrichum juniperinum

Boletus sp.

As in many other parts of the country, the trees all have vernal
wind-pollinated flowers, and the same is true of most of the shrubs.
Among the herbs the commonest species is wind-pollinated, but
most of the others have yellow, white, or blue flowers (with little
or no odor). There are more herbaceous flowers in spring than
in midsummer, and more in fall than in spring, at least if we con-
sider species regardless of their relative abundance. Plumose
seeds or fruits prevail among the herbs, but "tonoboles" are quite
common also. None of the herbs seem to have fleshy fruits, but
the nut-like fruits of Comandra may be eaten by small mammals.
Some of the smaller herbs, particularly the Polygalas, have ap-
pendaged seeds which are thought to be adapted to transportation
by ants.

Some dynamic studies of the upland vegetation were made in
1916. On Oct. 27, about a mile S.S.E. of Westbury Station, a
typical sample of herbaceous vegetation, consisting chiefly of
Andropogon scoparius (which constitutes the bulk of the herbage
of the Plains) was cut close to the ground from a small measured
area, so as to get the total annual growth per unit area, exclusive
of a small amount of stubble and roots. It weighed 8,220 pounds
per acre at the time, but then growth had probably ceased and
the drying out begun, so that if it had been cut a month earlier
the weight might have been greater. The same vegetation when
air-dry weighed 5,975 pounds per acre, which is probably less
than the average annual increment of vegetation in the eastern
United States, though much higher than figures obtained by
Shantz for somewhat similar vegetation in eastern Colorado.*

*U. S. Bur. Plant Industry Bull. 201: 81. 191 1.

276 Semi-centennial of Torre y Botanical Club

The ash weighed 265 pounds per acre, or between 4 and 5 per cent
of the air-dry weight.

According to some of the old inhabitants, the Plains vegetation
formerly grew taller than it does now. This probably does not
mean that the Andropogon scoparius was any taller, but that the
taller grasses, such as A. furcatus and Sorghastrum (which are
said to be more characteristic of the fertile prairies of the West) ,
were more abundant. If that is true the annual growth per unit
area must be decreasing, which is consistent with the suggestion
on a preceding page about the progressive impoverishment of
the soil. And the fact that the groves of pines at Island of Trees
are composed of rather small trees appears to indicate a com-
paratively recent invasion, which would be in harmony with the
same tendency, for Pinus rigida, like most other pines, flourishes
in very poor soils. But one would hardly suppose that the soil
could deteriorate so rapidly that the difference in vegetation would
be noticeable in a lifetime, and there may be some entirely different
explanation for the supposed change in vegetation.

The vegetation characteristic of the wet valleys is very limited
in extent. The largest stream on the Plains is East Meadow
Brook, which rises about three miles east of Garden City and flows
south about a mile before passing into the forest region. Most
of its vegetation within the prairie area has been destroyed in the
last few years, unfortunately, and the brook itself is nearly dry
now, but pretty full notes were taken there in 1907-1909. Hemp-
stead Brook, which flows through the eastern part of the village
of Hempstead, is next in importance, and there is a smaller brook
about two miles farther west which still has a trace of its original
vegetation.

Along the streams there are no trees except a few small
specimens of Acer rubrum and Nyssa, scarcely rising above the
shrubbery, but the shrubs are considerably larger than those of
the uplands, many of them being higher than a man's head. Fire
seems to be a negligible factor in the environment.

The meadow plants are divided into small trees and shrubs,
vines and undershrubs, herbs and mosses. Otherwise the treat-
ment is the same as that of the upland vegetation, the rarer
species being omitted, for the reasons already given.

Harper: Vegetation of the Hempstead Plains 277

Small trees and shrubs

Rhus Vernix 6-7 cream O

Myrica caroUnensis 5 O

Viburnum dentatum 5-6 white O

Rosa palustris 6 pink-purple

Cholisma ligustrina 6-7 white

Spiraea latifolia 6-7 white

Aronia nigra 5 white O

Spiraea tomentosa 7 pink-purple

Pieris Mariana 5"? white

Samhucus canadensis 6-7 white O

Vaccinium corymbosuml 5 white O

Acer rubrum 4 red Y

Vines and undershrubs

Rubus hispidus 6-7 white O

Rhus radicans 5-6 cream O

Oxycoccus macrocarpus 6-8 pink O

Herbs

Dulichium arundinaceum 6-8

Lycopus sp 8-9 white T

Hypericum adpressum 7 yellow T

Vernonia noveboracensis 7-9 purple Y

Osmunda cinnamomea

Eupatorium perfoliatum 8-9 white Y

Panicum virgatum 7-8

Juncus canadensis! 7-8 green T

Dryopteris Thelypteris

Lysimachia terrestris 6-8 yellow

Kneiffia linearis? 6-8 yellow*

Eriophorum gracile? . . 5 Y

Rhynchospora alba 7-8 white

Poly gala cruciata 7-9 pink

Eriocaulon septangulare 7-9 white

Rhexia virginica 7-9 pink-purple T

Polygonum sagittatum 6-10 white

Triadenum virginicum 7-9 pink-purple T

Gentiana Saponaria 9-1 1 blue T

Sparganium sp 7-8

Linum striatum 7-8 yellow

Viola primulifolia 5 white E

Juncus acuminatus? 6-8 green T

Helianthus angustifolius 8-9 yellow

Onoclea sensibilis
Osmunda regalis

* See Plant World 8: 301-303. 1906. In that paper the false common name
"evening primrose," which belongs to the related genus Oenothera but decidedly not
to Kneiffia, was inserted by the editors without the writer's knowledge or consent.
The figures are four times natural size.

278 Semi-centennial of Torrey Botanical Club

The majority of the shrubs and vines bloom in early summer
and have white flowers and fleshy fruits. Among the herbs the
proportion of aestival pink-purple flowers, and of tonoboles, is
noticeably larger than on the uplands, and there are few or no
fleshy fruits. There are, however, many species whose mode
of dissemination is not certainly known.* Some of these doubt-
less have seeds that float downstream, and are carried in other
directions on the feet of aquatic birds. The shrubs that bear
capsules are perhaps to be classed as tonoboles.

Weeds. Many of the roads across the Plains are entirely
unimproved, mere wheel-tracks, which are shifted a little from
time to time as the ruts become too deep, in precisely the same
manner as some of those in eastern Colorado described recently by
Shantz.f Along almost every such road can be found Euthamia

* See Torreya 8: 159. 1908.

t Jour. Ecology $: 19-42,/. 1-23. March, 191 7. Several of the illustrations in
that paper could be matched very closely on the Hempstead Plains, and much of
the text would apply very well too, except for the names of the plants discussed.

Agalinis purpurea

Viola lanceolata

Gratiola aurea

Xyris sp

Ludwigia alternifolia . . ,

Car ex sterilis?

Eriophorum virginicum .
Eupatorium purpureum .

Castalia odorata

Asclepias pulchra ,

Rhynchospora glomerata .
Euthamia tenuifolia . . . .

Drosera intermedia

Potamogeton sp

Hypericum canadense . .

Carex lurida

Solidago rugosa

Sagittaria latifolia

Eleocharis melanocarpa .

Aster salicifolius?

Lycopodium adpressum

Lobelia Nuttallii

Ibidium cernuum

Aletris farinosa

Juncus Greenei

8-10 pink-purple T

5 white E

6-7 yellow

7-9 yellow T

6-8 yellow T

5

7 Y

.6-8 pink-purple Y

6-9 white

. 7-8 pink-purple Y

6-8

8-10 yellow Y

. . 6-8 pink-purple

7-9 yellow T

5

8-10 yellow Y

8-9 white

6-7

. , 9 blue Y

9-10 white Y
. . 6-7 white T
. . . .6 green T

7-9 blue

Mosses

Sphagnum sp. (perhaps more than one)

Harper: Vegetation of the Hempstead Plains 279

tenuifolia, which may be native in some parts of the eastern
United States, but nearly always grows in places whose natural-
ness is not above suspicion, all the way from here to Florida.
Agrostis alba is very common also along roads, and most of the
other species whose names are in parentheses in the upland
vegetation list grow in similar places, where the original vegeta-
tion has been damaged without much disturbance of the soil.
None of them seem to invade undisturbed vegetation, however.

Where the soil has once been plowed up and cultivated many
additional weeds, such as Oenothera biennis, Ambrosia artemisii-
folia, Persicaria sp., Linaria vulgaris, Daucus Carota, Syntherisma
sanguinalis, and Aster ericoides, come in, and these seem able to
hold the ground indefinitely against a re-invasion by native
species. Very little attention has been paid to this particular
phenomenon as yet, but there will be time enough for it after the
natural vegetation, which needs more immediate attention, is all
gone.

Comparisons with other regions
There is no precisely similar vegetation anywhere else, as far
as known, but there are many places near and remote with vegeta-
tion somewhat similar in aspect or composition, or both. Among
the nearer places are the so-called heaths of Nantucket, described
by Harshberger,* and Block Island, Montauk Point, and various
other places near the coast of southern New England, if we may
judge by the few photographs and fragmentary floristic descrip-
tions that have been published, though in some of these cases the
treelessness is said to be the result of deforestation within historic
times.

The "hilltop barren formation" of eastern Massachusetts,
described by Blankinship,t has quite a number of species in
common with the area under consideration. In the government
soil survey of Rhode Island by F. E. Bonsteel and E. P. Carr,
published in 1905, there is described a "Miami silt loam,"|
occurring principally in the township of South Kingstown, in the

* Bull. Geog. Soc. Phila. 12: 73-76. 1914.
tRhodoraS: 128. May, 1903.

t In a subsequent publication of the Bureau of Soils this was changed to " Merri-
mac silt loam," a type of soil not reported outside of Rhode Island, and classed as a
glacial lake deposit.

280 Semi-centennial of Torrey Botanical Club

southern part of the state, which must be very similar to the
'Hempstead loam," the principal differences brought out in the
description being that the sand and gravel begin about three feet
below the surface instead of one, and the ground-water level is
much nearer the surface, sometimes rising above it in rainy weather
or when snow is melting rapidly. The soil survey report says
little or nothing about the vegetation, but in Rhodora (9: 117-
122) for July, 1907, there is a paper on The flora of the Great
Swamp of Rhode Island, by E. S. Reynolds, which throws some
light on the subject. The Great Swamp, which borders Worden's
Pond, is immediately south of the "Miami silt loam" areas, and
Reynolds's list includes quite a number of species which are cer-
tainly not swamp plants, and may have come from the silt loam
area, though habitats and localities (and abundance) are not
indicated. Species previously collected by others in the same
neighborhood are excluded from his list, and the reader is given
no intimation of what those might be, except that they are about
as numerous as those listed. Under the circumstances, therefore,
it is interesting to find in Reynolds's list the following which are
characteristic of the upland vegetation of the Hempstead Plains
(taking them in the same order in which they appear in the
present paper) : Rhus copallina, Baptisia tinctoria, Viola pedata,
Solidago puberulay Lespedeza capitata, Hypoxis, Bartonia virginica,
Ibidium (Spiranthes) gracile. And it is reasonably certain that
among the species collected by others and therefore ignored by
Reynolds there are other typical Hempstead Plains plants*. The
similarity of his list to our meadow vegetation is of course much
closer, as he was ostensibly dealing with swamp plants only.

The sand-plains of North Haven, Connecticut, described by
W. E. Britton,t also have many of the same species as our area,
and Dr. G. E. Nichols has sent me photographs of parts of those
sand-plains where the herbage was denser than any figured by Dr.
Britton, and appeared much like that of the Hempstead Plains.
The sand barrens of southern Staten Island, according to S. H.
Burnham.t are likewise characterized by some of the same species.

*For example, Fernald, in Rhodora 19: 58, reports Crocanthemum dumosum
from South Kingstown.

t Bull. Torrey Club 30: 571-620. pi. 23-28. 1903.
} Torreya 13: 249-255. Nov. 1913.

Harper: Vegetation of the Hempstead Plains 281

Other marked similarities to our area, floristic or vegetational,
are found in the serpentine barrens of Pennsylvania and Maryland,
discussed by Harshberger,* Pennell,t and Shreve,t and the sand
areas of Illinois, described by Gleason and others.§ (The typical
lUinois prairies, however, have much richer soil and more luxuriant
vegetation, with more broad-leaved herbs and almost no shrubs.)

Still farther west we can find a number of resemblances in the
sand-hills of western Nebraska, described by Rydberg|| and Pool.H
The less typical sand-hills of northeastern Colorado, visited by
the writer under the guidance of Dr. Shantz in August, 1915, are
probably more like the Hempstead Plains than are those of
Nebraska, for the vegetative covering is more continuous. The
dominant grass on the Colorado sand-hills is the same as on the
Hempstead Plains (or at least taxonomists have not yet separated
them). The regular short-grass prairie in the same neighbor-
hood also has some features in common with that under discussion.
Instead of our Baptisia tinctoria there is another leguminous plant
of much the same aspect, namely, Psoralea tenuiflora.'^'^ The simi-
larity of roadside conditions there and on Long Island has already
been mentioned under the head of weeds.

The gravelly prairies south of Puget Sound, described by
Piper,tt resemble ours in being level and grassy, with scattered
oaks, and even have a species of Sericocarpus, the only member of
the genus that grows outside of the eastern United States.

Considering briefly the southeastern states, the dry fields of
Middle Georgia have several of the same plants as the Hempstead
Plains uplands, and the prairie meadow plants are pretty well

♦Science 11. i8: 339-343- Sept. ii, 1903.

t Proc. Acad. Nat. Sci. Phila. 62: 541-584. 1911; 64: 520-539. 1913.

% Plant Life of Md. 213-215. pi. 20. 1910. The writer had opportunity to
visit some of the well-known serpentine barrens near Baltimore and Philadelphia
in June, 191 7. An early description, primarily mineralogical, of those nearest
Baltimore is by Dr. H. H. Hayden in Am. Jour. Sci. 24: 349-360. 1833.

§ Bull. 111. State Lab. Nat. Hist. 7: 149-194. Jan. 1907; 9: 23-174. pi. 1-20.
1910.

II Contr. U. S. Nat. Herb. 3: 133-200. pi. i, 2. 1895.

^ Minn. Bot. Stud. 4: 189-312. pi. 26-40. 1914. Reviewed in Bull. Am. Geog.
Soc. 47: 873-874. Nov. 1915.

**For desciiptions of prairie vegetation in Colorado, see Shantz, U. S. Bur. PI.
Ind. Bull. 201, 1911; particularly, plate 3- fig- i> and plate 4, fig. i.

tt Contr. U. S. Nat. Herb. 11: 42-44. pi. q, 10. 1906.

282 Semi-centennial of Torrey Botanical Club

represented in the meadows of the same neighborhood, and also
in western North CaroHna, though these southern grass-lands
may have once been wooded.*

Lastly, by way of contrast, we may consider the pine-barrens
of Long Island, which begin immediately east of the Plains. The
flora of the two regions has much in common, but the vegetation
is very different. f There are also some differences between
representatives of the same species in the two areas, which may
possibly hereafter be made the basis of subspecific distinctions.
For example, Quercus prinoides on the prairie has nearly simple
stems in large clumps, with broader, thinner, and paler leaves
than in the pine-barrens; and Sericocarpus linifolius has broader
and more numerous leaves on the prairie, presumably indicating
better soil, notwithstanding the greater exposure to sun and wind,
which ought to have just the opposite effect on leaves, if other
factors were equal. Of the two shrubby oaks characteristic of
both places, Quercus ilicifolia outnumbers Q. prinoides at least
ten to one in the pine-barrens, while on the Plains the latter is at
least twice as abundant as the former.

A discussion of the geographical affinities of the flora, the
families and genera most numerously represented or conspicuous
by their absence, etc., belongs more properly to a floristic paper.
But it may be noted in passing that most of the upland species
are widely distributed in sunny places, on rather poor soils, in
the northeastern United States south of the boreal conifer region,
and almost none of them are found in Europe or near the Pacific
coast. Few if any are confined to the coastal plain. The two
arborescent oaks, one of which reaches its northeastern limit here,
while the other extends to Massachusetts, have been seen by the
writer, or are reported by others to occur, usually together, on
the coast of New Jersey, on the edges of the serpentine barrens of
Pennsylvania and Maryland and dry prairies in Illinois, Missouri,
and Arkansas, around flat rock outcrops in Georgia and Alabama,
in the prairies, flatwoods, and barrens of Alabama and Mississippi,
and lastly in the "cross-timbers" of Oklahoma and Texas, where
they are said to constitute the bulk of the forest. Both reach

* See Bull. Torrey Club 27: 322, 327. 1900; Torreya 10: 63. 1910.
t See Torreya 8: 1-9. 1908.

Harper: Vegetation of the Hempstead Plains 283

their southern Hmits in the northern edge of Florida, as do some
of the other plants under consideration.

Among the most widely distributed members of the upland
vegetation, besides the two oaks just mentioned, are Rhus copal-
lina, Andropogon scoparius, Cracca virginiana, Aletris farinosaj
Viola pedata, Sorghastrum nutans, Hypoxis hirsuta, Car ex penn-
sylvanica, Poly gala polygama, Andropogon fur catus, and Antennaria
plantaginifolia; while some of the most local are Aster dumosus
strictior, Crocanthemum dumosum, Agalinis acuta, Juncus Greenei,
and Linum intercursum. (These last are all reported also from Nan-
tucket Island.) The species which are probably more abundant
here than in any other equal area in the world, in addition to the
five last named, are Pieris Mariana, Quercus prinoides, Andro-
pogon scoparius, lonactis, Baptisia tinctoria, Viola pedata, Anten-
naria neglecta, Lespedeza capitata sericea, Scleria pauciflora, Serico-
carpus linifolius, Lespedeza angustifolia, and Eupatorium hyssopi-
folium.*

The meadow plants are more widely distributed, on the whole,
than those of the uplands, most of them ranging from Canada to
Georgia at least.

Destructive influences

Finally the influences tending to destroy the prairie vegetation
may be reviewed briefly. Whei;i the neighborhood was first
settled, in the seventeenth century, it was soon discovered that
the Hempstead Plains was not very well adapted to agriculture,
and for a century or two the greater part of it was treated as
public property or free pasture, much as unfenced land in the
more thinly settled states is today. Grazing has continued in a
small way down to the present time, but the area has probably
never been overgrazed sufficiently to w^eaken the native vegetation
and allow weeds to enter, except in enclosures near dwellings.

The absence of trees and rocks and hills made cultivation
very easy, however, so that some farms were established on the
Plains at an early date in spite of the povery of the soil (which
could be counteracted to some extent with manure, etc.), to supply

* Future species-splitting may require modification of some of the statements
in this paragraph, but apart from that they can hardly be challenged until the
vegetation of a number of other areas is studied quantitatively.

284 Semi-centennial of Torrey Botanical Club

the surrounding villages, before easy railroad communication with
the interior of the country made it more economical to bring part
of the food supply from the more fertile lands of the West. Agri-
culture on the Plains, as elsewhere on Long Island, probably reached
its maximum extension two or three generations ago. Census
statistics for Nassau County go back only to 1900, for it was not
separated from Queens until 1899, but the number of farms in
the county decreased from 1,658 in 1900 to 1,017 in 1910, and the
acreage of farm land decreased during the same period from a
little over half the total area to less than a third, and is doubtless
still less now.* But prairie land once cultivated and afterwards
abandoned seems never to revert to the original vegetation, as
pointed out under the head of weeds; or at least if it does the
process is so slow that no evidences of it have yet been discovered.

At the present time more of the land is used for residential
purposes than for agriculture, the proximity of New York City
and the ease of communication causing many people to settle on
and around the Plains quite independently of soil conditions.
Nassau County had 202 inhabitants per square mile in 1900 and
303 in 1910, and probably has about 400 now. Although this
causes considerable encroachment on the prairie and may be the
ultimate means of obliterating it, in a way it tends to protect it
from agricultural exploitation, for it makes some of the land too
valuable for farmers to touch, just as there is said to be more
natural prairie now inside the city of Chicago than for a consider-
able distance outside, for a similar reason.

A more serious menace at the present time is the appropriation
of considerable areas for pleasure purposes, such as polo and golf ;
the latter having brought about the destruction of most of the
Meadow Brook vegetation a few years ago, causing keen regret to
nature-lovers. For there are innumerable suitable sites for golf

* There was less than half as much farm land and only a little more than half
as much cultivated land, on Long Island in 1910 as in 1850, when such statistics
were first included in the census. The farmers are evidently being crowded out by
commuters and millionaires. In the last few years the Garden City Company,
perhaps unmindful of the rapid decline of farming on Long Island and the futility
of trying to counteract it, has plowed up several hundred acres of virgin prairie by
machinery and offered to lease the land to farmers (see news item at bottom of
page 113 of Torreya for June, 1914), but apparently without much success.

Harper: Vegetation of the Hempstead Plains 285

links, but only one Meadow Brook. (The name East Meadow
Brook does not necessarily imply that there were two, but was
probably applied originally simply to designate the brook flowing
through the East Meadow: i. e., east of Hempstead.)

During the Spanish-American war in 1898 some of our soldiers
were encamped on the Plains near Mineola (Camp Black), and
in 1917 a much larger encampment (Camp Mills) was located east
of Garden City and Hempstead, causing the trampling down or
otherwise injuring of about a square mile of vegetation, which
will probably never fully recover.*

When the aeroplane became an accomplished fact, in 1909, the
Hempstead Plains was very soon selected as an ideal place to
experiment with the new means of locomotion, on account of the
large flat area comparatively free from obstructions, and its
proximity to our largest city. The necessary buildings and regular
alighting places have encroached on the prairie a little near
Mineola, but otherwise this industry has done little damage;
and it probably deserves the good wishes of botanists, for it is
decidedly to the interest of the aviators that no more of the Plains
should be cultivated or built upon. There was indeed once a little
complaint from them that the surface was a little too rough (from
the tufts of grass, no doubt), and should be smoothed, but they
have apparently become reconciled to that.

Let us hope that the State or the federal government or some
public-spirited organization will soon take steps to preserve the
rest of this unique and easily accessible prairie permanently from
further encroachments, for the benefit of aviators, even if the
interests of plant sociologists and other nature-lovers are not con-
sidered at all in these days of commercialism. But if it comes to
the worst the southeastern corner of the Plains, which is remotest
from settlements, will probably have some of its natural vegeta-
tion still in condition for study two or three generations hence,
and some important ecological principles may yet be discovered
there.

* a news item sent from Camp Mills to the daily papers on Oct. 21, 1917, men-
tioned the menacing of the camp by a prairie fire; something that some of the western
soldiers may have been familiar with, but could hardly have expected to see so
near the metropolis.

286 Semi-centennial of Torrey Botanical Club

At least four other chapters should be written about this

unique area by competent persons before it is too late : one on its

geology, with special reference to the origin of the soil; one on its

flora, with attention to the points mentioned on pages 273 and 282,

and others easily called to mind; one on its fauna; and one on its

agricultural and economic history.

Explanation of plate 7

Looking southeast in dry valley at head of Hempstead Brook, about i H miles
east of Garden City, showing treeless horizon about }/i mile away. The view
embraces a ' horizontal angle of about 40°. (There was no house within a
mile of this spot, but if the camera — which was about four feet from
the ground — had been raised a few feet, some buildings and planted trees would
have appeared in the view.) Taken in a gentle rain at 4 p.m., Oct. 27, 1907.
(For other views taken near the same place see Bull. Am. Geog. Soc. 43: 352;
Torreya 12: 279. The most conspicuous plant in the foreground of the former,
not there designated, is Eupatorium hyssopifoUum.) Ten years later this place was
included in the site of Camp Mills, and its appearance greatly altered; so that
there will probably never again be an opportunity to take such a photograph on
Long Island.

Addenda (March, 1918)

P. 276. The tradition is, Mr. Henry Hicks tells me, that a century or more
ago a man crossing the Plains on horseback on a dewy morning would be wet to
his waist; which seems to indicate that much of the grass was five or six feet tall.
This is not at all improbable, for a news item in a Kansas City paper of Oct. 27, 191 5,
reports blue-stem grass (Andropogon furcatus) growing to a height of 93^ feet in
Chase County, Kansas.

Pp. 276, 284. Both Hempstead Brook and East Meadow Brook within the
prairie area are dry most of the time now, a considerable change having taken
place in that respect within the writer's recollection, perhaps on account of the
lowering of the ground-water about their sources by pumping from deep wells to
supply the rapidly growing villages.

A STUDY OF SOME FACTORS INFLUENCING
THE STIMULATIVE ACTION OF ZINC SUL-
PHATE ON THE GROWTH OF ASPER-
GILLUS NIGER. I. THE EFFECT OF
THE PRESENCE OF ZINC IN
THE CULTURAL FLASKS

By R. a. Steinberg

Columbia University

The increased growth of Aspergillus niger resulting from the
introduction of zinc salts into the nutrient solution was first
observed by Raulin (9). This observation was confirmed by
Richards (10) in 1897, who noted also that other elements (Co,
Ni, F, Fe, etc.) exercised an influence similar to that of zinc.

The studies of these two investigators, as well as those of
Ono (8), of Richter (11), and of others agree in that there is the
greater growth in the presence of zinc, although the percentage
increase the various authors obtained is not identical.

In 1903, however, a publication by Coupin (2) appeared, in
which this author came to the conclusion that the increased dry
weights formed in cultures of Aspergillus niger through the addi-
tion of zinc salts takes place only in impure cultures. The addi-
tion of zinc is effective, in this investigator's opinion, only in those
cases in which there is an opportunity for the suppression of the
activities of other organisms, whose presence in the Aspergillus
niger cultures prevents the full development of this fungus. In
pure cultures the maximum dry weight is obtained without the
addition of zinc.

Another interpretation of Coupin 's results has been suggested
by Javillier (3) — namely, the introduction of zinc into the nutrient
liquid through solution of constituents of the glass of the cultural
flask. Javillier, indeed, states that with cultures in Jena glass
(flasks of the same kind of glass were made use of by Coupin) the
addition of zinc is unnecessary inasmuch as the maximum growth

287

288 Semi-centennial of Torrey Botanical Club

(about I gram per 50 c.c. nutrient solution) takes place. Simul-
taneous cultures in Kavalier Bohemian and quartz flasks did not
exceed a yield of about 0.3 grams per 50 c.c. nutrient solution
unless zinc was added.

According to Lepierre (5, 6), on the other hand, the increased
yields obtained by Coupin should be ascribed to the excessive
volume of nutrient solution employed as compared to the volume
of the containing flask. The increased depth of the solution, he
assumes, results in decreased aeration of the cultures and increased
growth ensues. Zinc, it is stated by Lepierre (6), could be found
neither in the flasks (whether Jena or Kavalier is not stated) or
compounds used nor in the fungal membranes.

Javillier (4), in repeating Lepierre's experiments in Kavalier
glass, did not succeed in obtaining an essential variation in the
yield by varying the ratio of surface to volume.

It is interesting to note in this connection that the presence
of zinc in Jena glass has frequently been reported in the literature,
more recently through the analyses of Nicolardot (7) and of
Walker and Smither (12). Moreover, Kavalier Bohemian glass
according to the same authors is free from zinc.

That the composition of the cultural flasks is of importance
and that both the rate of growth and the fructification of Asper-
gillus niger can be influenced by solution of components of the
glass has been claimed by Benecke (i).

To obtain, if possible, additional evidence as to whether
cultures of A. niger to which no zinc has been intentionally added
attain a greater growth in Jena glass than in Kavalier Bohemian,
I have grown parallel cultures in these two glasses, and in addition,
a third glass, Pyrex. Zinc according to Walker and Smither (12}
does not enter into the composition of Pyrex glass.

In the experiments whose description follows the Pfeffer nutrient
solution has been used (10).

Pfeffer solution

Grams

Water ' looo.o

Cane sugar 50.0

Ammonium nitrate 10. o

Mono-potassium phosphate 5.0

Magnesium sulphate 2.5

Iron sulphate trace

Steinberg: Zinc sulphate and Aspergillus nigkr 289

The compounds employed in the preparation of this solution were
water redistilled through glass; Merck's "Reagent" ammonium
nitrate and magnesium sulphate; Kahlbaum's "Zur anal."
magnesium sulphate; and Baker's "Analyzed" potassium phos-
phate, ferric sulphate, and zinc sulphate. The cane sugar used
is that sold under the proprietary name of "Crystal Domino"
sugar. This solution was prepared as needed.

The flasks used were: 200 c.c. Jena and 150 c.c. Pyrex Erlen-
meyer's; and 250 c.c. Kavalier Bohemian Florence flasks. They
were cleaned by rinsing with concentrated sulphuric acid, tap-
water, lastly distilled water, and inverted to drain dry. The pre-
caution was taken of reserving part of the flasks for zinc-free
cultures and part for zinc cultures only, though this performance
is unnecessary, as the following indicates:

I. On the efficiency of the method for cleaning the cul-
tural FLASKS

Pfeffer solution: water redistilled through glass; "Crystal
Domino" sucrose; NH4NO3 (Merck) ;• KH2PO4 (Baker) ; MgS04.-
7H2O (Kahlbaum). Temperature 30-31° C. Period of growth
7 days. No zinc added. Pyrex flasks.

Flasks reserved Flasks reserved

for zinc-free cultures for zinc cultures
0.924 grams* • 0.283 grams

0.336 " 0.265
0.321 " 0.313
0.328 " 0.341
0.35s " 0.352

0.335 0.311

The flasks in the second column were previous to this experi-
ment used three consecutive times for cultures containing 10
mg. Zn/L. and were cleaned as usual.

Zinc was added where indicated to the entire solution used
in the preparation of cultures having the same concentration of
this heavy metal and not to the individual flasks. The stock solu-

* High yield, due probably to the accidental introduction of zinc. This value
not included in the average. In at least one case it was noted that the same Pyrex
flask consistently gave cultures having a high yield, though the addition of zinc
was omitted.

290 Semi-centennial of Torrey Botanical Club

tion contained 2.5 mg. zinc per cubic centimeter (i. e., 11. o mg.
ZnS047.H20 per c.c).

The flasks, each containing 50 c.c. of nutrient solution measured
in a 50 c.c. graduate, were sterilized at 14^2 lbs. for 20 minutes.

Inoculations were from stock bread-cultures grown at room
temperature (18-25° C). Enough spores were added with a
platinum loop to make a visible and apparently almost continuous
layer on the solution surface. The A. niger culture used in these
experiments was obtained originally from the " International-
stelle fiir Pilz-Kulturen, Amsterdam." Immediately after inocu-
lation the flasks were placed in a dark cupboard at room tempera-
ture (18-23° C), or in the thermostat at 30-31 ° C.

When harvested, the membrane, together with the solution,
was thrown on a washed and weighed filter, washed with distilled
water and dried at 103-105° C. for four days.

The yields while given to the third place are probably reliable
to two places only.

2. The effect of adding zinc to the culture medium
Pfeffer solution: water redistilled through glass; "Crystal
Domino" sucrose; NH4NO3 (Merck) ; KH,P04 (Baker); MgS04.-
7H2O (Merck). Room temperature (18-23° C). Period of
growth ten days.

Kavalier Bohemian

Pyrex

No zinc

ID mg. Zn/L

No zinc

ID mg. Zn/L

0.317 grams

0.330 "
0.341 "
0.302
0.314 "

0.888 grams
0.905 "
0.886 "
0.880 "
0.924 "

0.387 grams
0.328 "
0.306 "
0.286
0.325 "

0.864 grams
0.855
0.878 "
0.868 "
0.903 "

0.321

0.897 "

0.326

0.874 "

Here we see that in the Kavalier and Pyrex flasks the addition
of zinc to the nutrient solution results in an increased dry weight,
the increase being almost three-fold.

3. The influence of a zinc glass (Jena) on the yield
Pfeffer solution: water redistilled through glass; ''Crj^stal
Domino" sucrose; NH4NO3 (Merck); KH2PO4 (Baker) ; MgS04.-

Steinberg: Zinc sulphate and Aspergillus niger 291

7H2O (Kahlbaum). Room temperature (18-23° C). Period of
growth ten days.

Jena

Kavalier Bohemian

Pyrex

No zinc

10 mg. Zn/L

No zinc

10 mg. Zn/L

No zinc

10 mg. Zn/L

0.989 gr.
0.958 "
0.919 "
0.933 "
0.953 "

0.980 gr.
0.940 "
1.005 "
0.988 "
1.022 "

0.270 gr.

0.299 "
0.285 "
0.300 "
0.351 "

0.924 gr.

0.943 "
0.886 "

0.947 "
1. 017 "

0.319 gr.
0.248 "
0.306 "
0.252 "
0.309 "

0.940 gr.
0.980 "
0.917 "
0.952 "
0.997 "

0.950 "

0.987 "

0.301 "

0.943 "

0.287 "

0.957 "

With nutrient solutions to which no zinc has been added a
distinct difference is displayed, as concerns the yield, between
the Jena flasks, on the one hand, and the Kavalier and Pyrex
on the other. The low yields characteristic of the zinc-free culture
are obtained only in the latter two glasses.

Cultures grown in the presence of 10 mg. zinc (calculated as
metal) per liter attained a dry weight of approximately one gram
irrespective of the kind of flask used.

Additional experiments bringing out the increased growth
taking place in a culture medium to which no zinc has been inten-
tionally added when in Jena glass are as follows :

4. The influence of a zinc glass (Jena) on the yield
Pfeffer solution: water redistilled through glass; "Crystal
Domino" sucrose; NH4NO3 (Merck) ; KH2PO4 (Baker) ; MgS04.-
7H2O (Kahlbaum). Temperature 30-31° C. Period of growth
seven days. No zinc added.

Pyrex Jena
0.247 grams 0.948 grams

0.232
0.327
0.279
0.257

0.268

0.896
0.893
0.904
0.904

0.909

Conditions as in the preceding experiment

Pyrex Jena
0.336 grams 0.915 grams

0.316
0.376
0.332
0.296

0.849
0.881
0.872
,0.864

0.331

0.876

292 Semi-centennial of Torrey Botanical Club

Here also we see that while addition of zinc is necessary to
bring about increased growth in Pyrex flasks, this increase in
growth occurs in the cultures in Jena glass to which no zinc has
been intentionally added.

A comparison of the values obtained in these experiments will
show the agreement in the dry weights of individual duplicate
cultures, and of the mean values (average of five duplicate cultures)
obtained in the different experiments. On the average the devia-
tion from the mean does not exceed 0.040-0.050 gram, in excep-
tional instances deviations of as much as 0.10-0.15 gram being
encountered. This means, therefore, a variation of about 5 per
cent for the zinc cultures and of 15 per cent for the zinc-free cul-
tures, since the mean weight of the former is about 0.95 grams, of
the latter 0.30 grams.

We are therefore justified in concluding that while the addition
of zinc serves to bring about an increased formation of dry weight
in Kavalier Bohemian and Pyrex flasks, such increased formation
of dry weight takes place in Jena flasks to which no zinc is inten-
tionally added. That the increased growth that occurs in the
Jena flasks is due to the solution of small amounts of zinc from
the flasks is also highly probable in view of the presence of zinc
in this glass.

LIST OF PAPERS REFERRED TO IN TEXT

1. Benecke, W. Die Bedeutung des Kaliums und des Magnesiums

fiir Entwickelung und Wachstum des Aspergillus niger v. Th.
sowie einiger anderer Pilzformen. Hot. Zeit. 54: 97. 1896.

2. Coupin, H. Sur la nutrition du Sterigmatocystis nigra. Compt.

Rend. 136: 392. 1903.

3. Javillier, M. Une cause d'erreur dans I'etude de Taction biologique

des elements chimiques: la presence de traces de zinc dans le
verre. Compt. Rend. 158: 140. 1914.

4. Javillier, M. Utilite du zinc pour le developpement de r^5/)ergz7/w5

niger cultive sur milieux profonds. Bull. Soc. Chim. IV.
15-16: 568. 1914.

5. Lepierre, C. Inutilite du zinc pour la culture de V Aspergillus

niger. Compt. Rend. 157: 876. 191 3.

6. Lepierre, C. Zinc et Aspergillus. Les experiences de M. Coupin

et de M. Javillier. Compt. Rend. 158: 66. 1914.

Steinberg: Zinc sulphate and Aspergillus niger 293

7. Nicolardot, P. Sur I'attaque des verres de France, de Boheme et

d'Allemagne. Compt. rend. 163: 355. 1916.

8. Ono, N. Ueber die Wachsthumsbeschleunigung einiger Algen

und Pilze durch chemische Reize. Jour. Coll. Sci. Imp. Univ.
Tokyo 13: 142. 1900.

9. Raulin, J. £tudes chimique sur la vegetation. Ann. Sci. Nat.

Bot. V. 11: 93. 1869.

10. Richards, H. M. Die Beeinflussung des Wachsthums einiger Pilze

durch chemische Reize. Jahrb. Wiss. Bot. 30: 665. 1897.

11. Richter, A. Zur Frage der chemischen Reizmittel. Die Rolle

des Zn und Cu bei der Ernahrung von Aspergillus niger.
Centralbl. Bakt. II. 7: 417. 1901.

12. Walker, P. H. & Smither, F. W. Comparative tests of chemical

glassware. Journ. Ind. & Eng. Chem. 9: 1090. 1917.

TWO MONTHS IN THE SOUTHERN CATSKILLS

By Oliver P. Medsger

Lincoln High School, Jersey City

During the summer of 1914, the writer spent ten days at
Woodland, New York, as the guest of Mr. H. W. Little, the director
of "Camp Wake Robin," one of the oldest and best known of the
boys' camps in the East. Much of the time was spent in getting
acquainted with the flora of the region. This year (191 7) he
accepted the position as councilor and director of nature study in the
camp and spent from June 30 to August 30 there. Although some
attention was given to birds and insects, yet ample time was
available for a further study of the flora.

The writer was not the first Torreyite to collect and study in
this region. We find in the register of the Roxmor hotel at
Woodland a record of a field meeting of the Torrey Botanical
Club, June 2, 1901. Members who stopped at the hotel were
Fanny A. Mulford, Heloise G. Esterly, N. L. Britton, Mrs.
Britton, Anna Murray Vail, Alexandrina Taylor, E. P. Bicknell,
and C. L. Pollard. It is not often that so many distinguished
botanists attend a field meeting of the Torrey or any other botan-
ical club. Mr. Edward Miller, the proprietor of the hotel at
Woodland, is a former member of the Torrey Club and took
much interest in our botanical work. Mr. Bicknell must be very
familiar with the region, for it was in 1880, on Slide Mountain,
that he discovered the thrush that bears his name. To me it
was a great pleasure while camping on Slide Mountain to Hsten to
the fine flute-like strains of this bird.

Woodland is situated in Ulster County, about four miles south
of Phoenicia, in the heart of the southern Catskills. It has an
altitude of about 1,000 feet. Many of the highest peaks are in
full view. In the narrow valley flows Woodland Brook, one of the
finest and most beautiful of all the streams in the Catskill region.
The eminent naturalist, John Burroughs, on returning from a

294

Medsger: The southern Catskills

295

trip to the summit of Wittenberg, remarked : "The trail to which
we had committed ourselves led us down into Woodland valley,
a retreat which so took my eye by its fine trout brook, its superb
mountain scenery and its sweet seclusion, that I marked it for
my own." Elsewhere, Mr. Burroughs speaks of this immediate
locality: "Of all the retreats I have found amid the Catskills,
there is no other that possesses quite so many charms for me as
this valley; it is so wild, so quiet, and has such superb mountain
views."

The rocks of the region are practically all of sandstone of
the Devonian period. The tops of the mountains are capped
with a coarse conglomerate which apparently disintegrates easily;
thus the soil, as a general thing, is sandy or gravelly. Loose
stones abound almost everywhere, either the result of glacial
action or of weathering. The mountains are steeper and rougher
and the valleys narrower and deeper than they are in the northern
Catskills.

Every week we took a hike with the camp boys to some
interesting locality. We spent on each trip from one to three
days. The region around Woodland for a radius of nearly ten
miles was fairly well covered. Among the mountains visited
were Terrace, Wittenberg, Cornell, Slide, The Giant's Ledge, Cross
Mountain, and Mt. Pleasant. We also collected plants about
Winnesook, along the Panther Kill, at Diamond Notch, in the
West Kill Valley, at the Broad Hollow Notch, and about Shanda-
kin. Slide Mountain is the highest peak of the Catskills and we
found it a most interesting region botanically.

Almost the entire area is covered with forests. Years ago,
some of the best timber was cut away and a surprising waste in
wood took place. Hemlock trees by the thousands were felled
for the bark, which is extensively used for tanning, and the logs
were allowed to decay on the ground. This was probably the
most abundant tree of the neighborhood, now it is scarcely com-
mon. It was about the time of the Civil War that the greatest
destruction of this tree occurred. Where trees were cut for lumber,
very wasteful methods were used. Sometimes, as on Cornell
Mountain, one may find a virgin forest of rare beauty. It is
always a delight these days to find such a forest in the East. All

296 Semi-centennial of Torrey Botanical Club

the higher mountains of this region are in the State Forest Re-
serve. This reservation includes practically everything above an
altitude of 1,500 feet.

About forty-five species of native trees were found. Un-
doubtedly the most abundant tree of the neighborhood is the
yellow birch, Betula lutea. To a great extent it is taking the
place of the hemlock. In some localities one sees scarcely any-
thing else. A few beautiful specimens of the paper birch were
observed. Their chalky whiteness could be seen from afar. The
sweet or black birch, Betula lenta, is common. In July, when the
American chestnut was in bloom, many fine trees were seen about
Woodland. This tree extended up the mountain side to an
altitude of about 1,500 feet. The chestnut-tree blight, so destruc-
tive in many localities, reached that part of the Catskills this
year for the first time. During the latter part of August, its
ravages could be plainly seen.

Oaks are scarce in that locality. A few good trees of the red
oak were observed in the vicinity of Woodland. The black oak is
the only other Quercus that we came across. Hickories are also
rather scarce, only a few trees being found. Among the more
common trees growing there are the American beech, American
hornbeam {Carpinus caroliniana) , American aspen (Populus
tremuloides) , large-toothed aspen {Populus grandidentata) , Amer-
ican linden, white ash, sugar maple, mountain maple, red maple,
witch hazel, American elm, and slippery elm. The serviceberry
{Amelanchier canadensis) is quite common and has finer fruits
than it has in any other locality where I have found it. A few
fine specimens of the butternut {Juglans cinerea) are growing in
the lowlands about Woodland. At an altitude of about 2,000
feet, the striped or goosefoot maple is quite common. Some of
the trees are forty feet high with trunk diameters of six to eight
inches.

The white pine is a common tree about Woodland and
Phoenicia. Many of the smaller trees are attacked by the white-
pine weevil, Pissodes Strohi Peck. The attack in almost every
case is made on the central axis about three feet from the top,
causing it to die. We found no evidence of the white-pine blister
rust.

Medsger: The southern Catskills 297

On the summits of the mountains, evergreens predominate.
On Wittenberg and Slide mountains, the balsam fir (Abies hal-
samea) is the chief tree. Near the summit of the former mountain
are almost impenetrable forests of this tree. On Wittenberg are
also to be found the black spruce and a few small trees of the red
spruce. Cornell Mountain, which is one of the highest peaks
of the Catskills, reaching an elevation of 3,900 feet, has its top
and west side covered with a heavy virgin forest of red spruce
(Picea rubra). Viewed from the summit of Slide Mountain, this
beautiful forest appears to cover several square miles. Scarcely
any other species grow in this area and its boundaries are very
definite. It is almost inaccessible to the lumbermen. The trees
are tall and straight, many of them attaining a diameter of two
feet or more.

The mountain ash is a common tree on Slide and some of the
other mountains. This year its large fruit-clusters were well
developed. Judging from the variation in the leaves, both Sorbus
americana and Sorbus sambucifolia probably exist in this locality.
This will be a question for further investigation.

Thirty species of ferns were collected. The most abundant
were the hay-scented fern, Dennstaedtia punctilobula, and the
spinulose shield-fern, Dryopteris spinulosa. A coarse variety of
the latter fern is found near the summit of Slide Mountain. The
intermediate variety is especially abundant along the northeast
base of Slide, where it grows with Lycopodium lucidulum. The
two often nearly cover the ground. Nowhere else in the country
have we found this club moss so plentiful. Among the rarer ferns
observed were Botrychium lanceolatum, Camptosorus rhizophyllus,
Matteuccia Struthiopteris (of which only one station is known in
the locality we studied), and Dryopteris Braunii. The last-named
fern was of especial interest to us, for the specimens were the first
we had seen outside of the herbarium. We first found it on the
east side of Mount Pleasant, where it was growing with Dryopteris
Goldieana. We afterward found many plants of it growing along
the trail to Wittenberg at an altitude of about 1,500 feet. It is
graceful, distinct, and certainly one of the most beautiful of all
our eastern species.

Inasmuch as nine tenths of the region studied is covered

298 Semi-centennial of Torrey Botanical Club

with woodland, weeds and foreign plants generally are not
plentiful. However, two of these are especially abundant.
One is Echium vulgare, commonly known as blueweed or
viper's bugloss. Its bright blue flowers were conspicuous
all along the roadside from Phoenicia to Woodland. The other
plant referred to is Origanum vulgare or wild marjoram, also
a native of Europe. It grows in great abundance in almost all
the cleared or waste land about Woodland, often so plentiful as
to exclude other plant life. As it grew everywhere about the
hotel, its purple bracts, conspicuous blossoms and strong mint odor
brought forth many inquiries concerning it.

It is always interesting to note the succession of plants and
trees as one ascends a mountain. This we especially observed in
going to the summits of Wittenberg, Cornell, and Slide. The
trail to Wittenberg first leads up the northeast side of Terrace
Mountain, which is covered with a dense forest, in which the
most abundant tree is yellow birch. The chestnut, beech, and
other trees of low elevation are soon left behind. The shrub
most common is hobble-bush. Viburnum alnifolium. It seems
rather strange to find this plant in such abundance, for one
rarely sees it fruit in these dense woodlands. At about i,8oo
feet Acer pennsyhanicum appears. The dry glaciated top of
Terrace is reached at an altitude of 2,300 feet. Here amid
the rocks and boulders, the blueberries of two or three species
grow in great abundance. There are also a few small scat-
tered trees of balsam fir and black spruce. The mountain ash
also appears for the first time. Probably the most conspicuous
plant on Terrace, is Clintonia horealis. It grew everywhere along
the trail, about the rocks and in fact in any place where it could
get a chance to grow. What a pity its large blue berries are not
edible, for gallons of them could have been gathered. Clintonia
umbellulata, which is common in Woodland valley, is also occa-
sionally found here.

As we leave Terrace to continue the trail to the summit of
Wittenberg, we begin to find Trillium undulatum, w'th its bright
red berries filled with seeds. This plant is most numerous at an
altitude of about 2,500 feet. At 3,000 feet the ground hemlock
or American yew, Taxus minor, becomes plentiful. This small

Medsger: The southern Catskills 299

conifer sometimes nearly covers the ground. We next came across
the creeping snowberry, Chiogenes hispidula, with its snow-white
berries and strong odor of wintergreen. Deciduous trees are
getting fewer, while the fir and the spruce are becoming more
abundant. From 3,000 to 3,700 feet the mountain is steep and
the trail is difficult to follow. Soft moss generally covers the
ground and often the trunks and branches of the fir trees are
covered with moss to their very tops. The sun shines but a few
hours a day on this, the north side of the mountain. Nowhere
else have we seen the common polypody fern grow in such perfec-
tion as on Wittenberg and Cornell mountains.

Probably the most conspicuous plant on the summit of Witten-
berg, which has an elevation of about 3,900 feet, is Aralia hispida.
Most of the plants had just finished blooming and small green
berries were forming. On the morning of that same day, August
25, we found a cluster of these plants in Woodland valley, where
the berries were dark purple, ripe, and many of them had either
fallen off or had been eaten by the birds. This plant looks much
more like an umbellifer than do the other species of the genus and
the immature plant is apt to be mistaken for one. On the very
summit of Wittenberg, the most interesting shrub to us is Ilicioides
mucronata or mountain holly. Heretofore we have seen this
shrub only in mountain swamps, and were surprised to find it on
the dry summit of Wittenberg. Growing in the open, the shrub
developed a fine globular form with somewhat pendent branches
containing numerous red berries. It would be a splendid thing in
cultivation if it could be induced to grow on moderately dry soil.

We spent the night on Wittenberg and had the great pleasure
of seeing the aurora b.orealis as it is generally pictured in books.
Great luminous streams flared up in the northern sky extending
well toward the zenith. They were ever changing yet always
beautiful. The phenomena lasted for nearly an hour. From
Wittenberg, we went to the summit of Cornell which is just a
little less than 4,000 feet in altitude. This mountain is covered
with trees. We went down the west side of Cornell through this
beautiful virgin forest of red spruce, crossed a flat strewn with
boulders and at an elevation of probably 3,000 feet began to
ascend the eastern slope of Slide. Here against the sunny incline,

300 Semi-centennial of Torrey Botanical Club

the wild red raspberry, Rubus strigosus, grows with much larger
stalks and finer, larger berries than in Woodland valley, where it
is so plentiful. The eastern slope of Slide is very vSteep and
difficult to climb, but we finally reached the summit. This is the
highest of all the Catskill range, with an altitude of about 4,250
feet. The trees are mostly balsam fir. On the very highest part
of Slide, the ground is often entirely covered with Cornus cana-
densis, the dwarf cornel or bunchberry. Its beautiful green leaves
bedecked with bunches of bright red berries made one of the most
pleasing botanical sights we have ever witnessed.

It is surprising how many bog plants and plants that grow in
cool, damp places are to be found on the dry summits of these
mountains. The probable causes are that there is more precipi-
tation on mountain tops, that it is cooler, the warm season is
shorter, and evaporation is less, so that during the greater part
of the year the soil is quite moist.

I have presented but a few of the botanical conditions of this
interesting region and hope that the future will give opportunity
for a more exhaustive study.

A PRELIMINARY REPORT ON THE RUSSULAE
OF LONG ISLAND

By Gertrude S. Burlingham

Eastern District High School, Brooklyn

Field work on the genus Russula on Long Island has been
limited to a region on the northern shore reaching from Cold Spring
Harbor to Port Jefferson. The first collections of which we have
published record were made by the state botanist, Dr. Charles H.
Peck, near Port Jefferson. As a result of this work he described
three new species of Russula in the State Museum Bulletin, number
50, pubHshed in 1897. In August, 1902, Dr. Peck, in company
with Professor F. S. Earle, continued the search for fleshy fungi
in the vicinity of Port Jefferson and Smithtown. From these
collections two new species of Russula were described by Dr. Peck
the following year in the sixty-seventh Bulletin of the State
Museum. In 1909 Professor C. H. Kauffman described one new
species, Russula sphagnophila, from Cold Spring Harbor. In the
summer of 1912 I spent July and the early part of August at
Cold Spring Harbor studying the ^ussulae and Lactariae of the
locality. Although the season was unusually dry, I was able to
secure 23 different species of Russula, four of which were un-
described.

In all, thirty-six species of Russula have been identified from
Long Island, of which fourteen are European species and twenty-
two American species. Nine of the latter have their type locality
near Port Jefferson or Cold Spring Harbor. The European species
reported from this region have a distribution in the United States
both to the north and the south of Long Island. Of the American
species eight, R. albida Peck, R. compacta Peck, R. crustosa Peck,
R. flavida Frost, R. Mariae Peck, R. suhvelutina Peck, R. uncialis
Peck, R. variata Bann. & Peck, have a distribution from New
England to Virginia, North Carolina, or Alabama. On the other
hand, the distribution of three species, R. alhella Peck, R. Earlei

301

302 Semi-centennial of Torrey Botanical Club

Peck, and R. pusilla Peck, seems to extend toward the south only ;
while R. hetulina Burl., R. flaviceps Peck, and R. serissima Peck
have been reported from Long Island to the northward only. Of
the type species from Long Island, R. anomala Peck, R. magnifica
Peck, and R. sphagnophila Kauff. have been found only in the
type locality. Although R. blanda Burl, has been found only at
Gold Spring Harbor, several collections have been made in different
localities and during different seasons.

One of the most abundant species of Russula found at Cold
Spring Harbor was R. Mariae Peck. It grew in woods, by wooded
roadsides, or even in the middle of sandy unfrequented roads.
R. pectinata Peck, although not so widely distributed as R.
Mariae Peck, occurred in abundance wherever found. One
species resembling R. decolorans Fr. in some respects but with the
wounded flesh turning red then gray is identical with an un
published new species which Professor H. C. Beardslee has in
manuscript.

The species referred to R. ohscura Romell is very probably R.
ruhescens Beards. I am more inclined to this opinion because this
autumn I have collected this on Staten Island and have seen fresh
specimens of it from White Plains, N. Y. In fact since I began
critical examination of the wounds of specimens resembling R.
ohscura Romell, I have seen none the wounds of which did not
turn red as in R. ruhescens Beards. When this was described in
1 9 14 it was known only from the type locality. But since then
it has been found in abundance around Boston, and in Newfane,
Vermont.

The region included between Port Jefiferson and Cold Spring
Harbor lies in the part of Long Island covered with ice at the
Ronkonkoma stage, and the soil is of a sandy or stony loam, and
gravel structure. The woods are composed of mixed hardwoods
with an abundance of oaks intermingled with chestnuts. It is
possible that the seemingly limited distribution of certain species,
and the apparent southern or northern distribution of others is
due to lack of extensive field wOrk or the rare occurrence of these
species. On the other hand the character of the soil and the
forest types may determine the southern or northern limit of
distribution. Except as temperature afifects the forest type it is

Burlingham: Russulae of Long Island 303

not probable that it affects the distribution of the Russulae on
Long Island.

The work on this genus on Long Island has been only begun.
It is probable that double the number of species found may yet be
discovered when the work of collecting has been extended to the
outwash plains toward the south and to the pine-bearing regions.
The work thus far done has shown that we may expect to find
not only those species ranging from New England to Virginia
and southward, but some southern species which reach their
northern limit in the latitude of New York, and certain northern
species which extend their southern limit to Long Island. And it
is possible that some species may prove to be found exclusively on
Long Island.

American species of Russula occurring on Long Island

1. Russula albella Peck, Ann. Rep. N. Y. State Mus. 50: loi.

1897.

Port Jefferson, Long Island, Peck & Earle 812, in herb. N. Y.
Bot. Garden. This is the type locality of the species. It occurs
in dry soil in deciduous woods. Its distribution in the United
States extends south as far as Mississippi.

2. Russula anomala Peck, Ann. Rep. N. Y. State Mus. 50: 99.

1897.

Only the type material of this has been found. It was collected
at Port Jefferson, on damp ground under trees.

3. Russula albida Peck, Bull. N. Y. State Mus. i-: 10. 1888.
Suffolk County, Long Island, Peck. The specimens are in the

herbarium of the State Museum at Albany.

4. Russula betulina Burl. N. Am. Fl. 9: 227. 1915.

Port Jefferson, Peck & Earle 805, in herb. N. Y. Bot. Garden.
The collection was made Aug. 5, 1902.

5. Russula blanda Burl. N. Am. Fl. 9: 213. 1915.

Cold Spring Harbor, type, 24, 1912, in herb. Burl., extype, herb.
N. Y. Bot. Garden and the Brooklyn Bot. Garden.

6. Russula compacta Frost & Peck; Peck, Ann. Rep. N. Y. State

Mus. 32: 32. 1880.
The specimens of this were collected by Peck in Suffolk County
and reported in the N. Y. State Mus. Bull. 116: 72. 1906.

304 Semi-centennial of Torrey Botanical Club

7. RussuLA CRUSTOSA Peck, Ann. Rep. N. Y. State Mus. 39: 41.

1887.

Port Jefferson, Peck & Earle 821, in herb. N. Y. Bot. Garden;
Cold Spring Harbor, Burlingham.

8. RussuLA Earlei Peck, Bull. N. Y. State Mus. 67: 24. 1903.
Port Jefferson, Peck Earle 843, type material in herb. N. Y.

Bot. Garden, col. Aug. 6, 1902; also 8y8, in herb. N. Y. Bot. Gar-
den, col. Aug. 2, 1902; Smithtown, Peck & Earle gi4, herb. N. Y.
Bot. Garden, col. Aug. 8, 1902.

9. RussuLA flaviceps Peck, Ann. Rep. N. Y. State Mus. 53 : 843.

1900.

Cold Spring Harbor, Burlingham 45, 1912.

10. RussuLA FLAVIDA Frost & Peck; Peck, Ann. Rep. N. Y. State

Mus. 32: 32. 1880.
Port Jefferson, Peck & Earle 884, in herb. N. Y. Bot. Garden,
col. Aug. 2, 1902.

11. RussuLA HUMiDicoLA Burl. N. Am. Fl. 9: 230. 1915.

Cold Spring Harbor, Burlingham 20, 191 2, type. Abundant in
thoroughly moist soil in woods. This has also been found in
Massachusetts.

12. RussuLA magnifica Peck, Bull. N. Y. State Mus. 67: 24.

1903.

Port Jefferson, Peck & Earle 841, in herb. N. Y. Bot. Garden,
col. Aug. 6, 1902. Thus far this species has not been reported
from any other region.

13. RussuLA Mariae Peck, Ann. Rep. N. Y. State Mus. 24: 74.

1872.

Cold Spring Harbor, Burlingham. Very abundant.

14. RussuLA modesta Peck, Bull. N. Y. State Mus. 116: 78.

1907.

Cold Spring Harbor, Burlingham 52, 1912.

15. RussuLA PUSiLLA Peck, Ann. Rep. N. Y. State Mus. 50: 99.

1897.

Suffolk County, Pecky type material; Cold Spring Harbor,
Burlingham ^o, 1912.

16. RussuLA SERissiMA Peck, Bull. N. Y. State Mus. 139: 44.

1910.

Cold Spring Harbor, Burlingham 100, 1912, August 3, 1912.

Burlingham: Russulae of Long Island 305

17. RussuLA SPHAGNOPHILA Kauffman, Rep. Mich. Acad. Sc\ 11 :

86. 1909.

Cold Spring Harbor, type locality, collected by C. H. Kauff-
man. Distribution limited to the type locality.

18. RussuLA SUBVELUTINA Peck, Bull. Torrey Club 33: 215.

1906.

Port Jefferson, Peck & Earle 871, in herb. N. Y. Bot. Garden,
col. Aug. 6, 1902.

19. RussuLA UNCiALis Peck, Bull. N. Y. State Mus. i^: 10. 1888.
Cold Spring Harbor, Burlingham gi, 1912. In moist oak and

chestnut woods, Aug. 3, 1912.

20. RussuLA VARIATA Bann. & Peck; Peck, Bull. N. Y. State Mus.

105: 41. 1906.

Port Jefferson, Peck & Earle 853, in herb. N. Y. Bot. Garden;
Cold Spring Harbor, Burlingham.

21. RussuLA viNACEA Burl. N. Am. Fl. 9: 217. 1915.

Cold Spring Harbor, Burlingham 8j, 1912, type; 86, 1912;
P7, 1912. Abundant in wet woods of oak and chestnut in early
August. This species occurs also on Staten Island and in New
Jersey.

22. RussuLA CINERASCENS Beardslcc (in manuscript).
Cold Spring Harbor, Burlingham.

European species of Russula found on Long Island

23. Russula aeruginea Lindbl.; Fries, Monog. Hymen. Suec. 2:

198. 1863.

Cold Spring Harbor, Burlingham gg, 1912, Aug. 3, 191 2.

24. Russula decolorans Fries, Epicr. Myc. 361. 1838.

Cold Spring Harbor, Burlingham. In wet woods of oak,
chestnut, and red maple, Aug. 3, 1912.

25. Russula delica Fries, Epicr. Myc. 350. 1838.
Cold Spring Harbor, Burlingham.

26. Russula densifolia (Seer.) Gill. Champ. Fr. 231. 1876.
Suffolk County, Peck, as reported in Bull. N. Y. State Mus.

116: 70. 1906.

27. Russula emetica (Schaeff.) Pers. Obs. Myc. i: 100. 1796.
Cold Spring Harbor, Burlingham. In moist woods on decaying

log.

306 Semi-centennial of Torrey Botanical Club

28. RussuLA FLAVA Lindbl. Nord. Svampb. 27. 1895.
Port Jefferson, Peck & Earle 828, Aug. 5, 1902.

29. RussuLA foetens (Pers.) Fries, Epicr. Myc. 359. 1838.
Cold Spring Harbor, Burlingham 41, 1912.

30. RussuLA FRAGiLiFORMis Burl., Mycologia 8: 312. 1916.
Russula fragilis Fries, Epicr. Myc. 359. 1838.

Cold Spring Harbor, Burlingham, July 5, 1912.

31. Russula heterophylla Fries, Epicr. Myc. 352. 1838.
Cold Spring Harbor, Burlingham.

32. Russula lepida Fries, Sv. Aetl. Svamp. 50. 1836.

Port Jefferson, Peck & Earle 854, Aug. 6, 1902, and 887, Aug.
7, 1902, in herb. N. Y. Bot. Garden.

33. Russula obscura Romell, Oefv. Sv. Vet.-Akad. Forh. 48:

179. 1891.

Cold Spring Harbor, Burlingham go, 1912, Aug. 3, 1912. This
is very possibly R. ruhescens Beardslee. It is impossible to dis-
tinguish some of the dark specimens of R. ruhescens from the
typical R. obscura. Beardslee did not describe his species until
1 914 and prior to that time no notes had been made of the flesh
of red forms of Russula changing to red when wounded. The only
way to distinguish these two positively is to observe the change
in the wounds of fresh plants.

34. Russula pectinata Fries, Epicr. Myc. 358. 1838.

Cold Spring Harbor, Burlingham 42, 191 2. Very abundant in
sandy soil. This is probably the species which Dr. Peck referred
to R. sororia Fries, Bull. N. Y. State Mus. 116: 84. 1907.

35. Russula purpurina Quel. & Schulz. ; Schulzer, Hedwigia 24:

139. 1885.

Cold Spring Harbor, Burlingham 75, 1912. Moist woods,
August.

36. Russula subolivascens Burl. N. Am. Fl. 9: 223. 191 5.
Russula olivascens Fries, Epicr. Myc. 361. 1838.

Port Jefferson, Peck & Earle 852, in herb. N. Y. Bot. Garden,
col. Aug. 6, 1902.

Doubtful species
Russula rubra Fries, Epicr. Myc. 354. 1838.

Suffolk County, Peck. Reported in Bull. N. Y. State Mus.
116: 79. 1907.

♦

ON THE OSMOTIC CONCENTRATION OF
THE TISSUE FLUIDS OF DESERT
LORANTHACEAE*

By J. Arthur Harris

Station for Experimental Evolution

1. Introductory remarks

In an earlier publication Mr. Lawrence and If discussed a
series of determinations of the osmotic concentration of the tissue
fluids of Jamaican montane rain-forest Loranthaceae parasitic
on various hosts. The purpose of this paper is to present the
results of studies of the sap concentration of the tissue fluids of
desert mistletoes and of that of their host plants for comparison
with the rain-forest series already published.

The number of determinations upon which the conclusions of
the present paper is based is not so large as that which Mr.
Lawrence and I were able to obtain in Jamaica. In explanation
I may merely say that the difficulties under which the desert
series was secured were far greater! than those which surrounded
the work in the Blue Mountains. This has necessarily restricted
the number of determinations, and notwithstanding our best

* Cooperative investigations carried out under the auspices of the Department of
Botanical Research and the Department of Experimental Evolution, of the Carnegie
Institution of Washington.

I am greatly indebted to Mr. P. C. Standley, of the United States National
Herbarium, for identifying these and other Arizona plants. Dr. Trelease, whose
splendid monograph of the genus Phoradendron has recently appeared, has kindly
gone over the parasites.

t Harris, J. Arthur, & Lawrence, John V. On the osmotic pressure of the tissue
fluids of Jamaican Loranthaceae parasitic on various hosts. Am. Jour. Bot. 3:
438-455- 1916.

X I am greatly indebted to the director and the staff of the Desert Laboratory
for every facility that could be given for these studies. The difficulties were such
as are inseparable from physiological work in camp under summer conditions in the
Southwestern deserts. For example, the torrential rains, which are characteristic of
the region during the midsummer season, more than once cut us off from supplies or
facilities necessary for our work.

307

308 Semi-centennial of Torrey Botanical Club

efforts has possibly limited somewhat the precision of the deter-
minations.*

While I hope later to secure larger, and better, series of data
on osmotic concentration in the tissue fluids of desert Loran-
thaceae, and to obtain series from mesophytic regions for com-
parison with those from the hygrophytic and xerophytic habitats
now available, there is little prospect of the completion of this
work in the near future. It seems proper, therefore, to place on
record the results so far obtained on the desert species for the
use of other physiologists.

In the present state of our knowledge, determinations of the
properties of the tissue fluids of desert Loranthaceae have a two-
fold interest.

1. In studies in the Arizonaf and the Jamaican coastal
deserts, t my associates and I have confirmed the conclusions of
Drabble and Drabble and of Fitting by showing that the osmotic
concentration of the sap of desert plants is in general far higher
than that of those of mesophytic and hygrophytic regions. It
seems a matter of considerable interest to determine whether the
same relationship holds for the Loranthaceae of these climatically
antithetical regions.

2. In our first study we found that the concentration of the
tissue fluids of the parasite is generally, but not invariably,
higher than that of the host plant. It seems desirable to test this

* Because of the rapid evaporation in the desert air, the danger of differential
water loss in the collection of the tissues of parasite and host was far greater in the
desert series. Because of the higher temperature to which the tubes of tissue were
necessarily subjected until they could be placed in the freezing mixture, the danger
of changes in the composition of the sap were necessarily greater than in the cool
climate of the Blue Mountains. The concentrations in molecules and ions of the
solutes, not the nature of the constituent compounds which may possibly be some-
what altered, is the subject under investigation. It may be doubted whether
standing even for a much longer time would seriously alter concentration. Further-
more, any changes of this kind would be quite as likely to affect the tissues of both
host and parasite, and hence to leave the relationship between them unaltered, as to
influence one alone.

t Harris, J. Arthur, & Lawrence, John V., with the cooperation of Gortner, R. A.
The cryoscopic constants of expressed vegetable saps as related to local environ-
mental conditions in the Arizona deserts. Phys. Res. 2: 1-49. 1916.

t Harris, J. Arthur, & Lawrence, John V. Cryoscopic determinations on tissue
fluids of plants of Jamaican coastal deserts. Bot. Gaz. 64: 285-305. iQi?-

Harris: Desert Loranthaceae

309

conclusion against data from as widely dissimilar environmental
conditions as possible.

II. Presentation of data

Since earlier studies have shown that in trees the concentra-
tion of leap sap is related to the height of insertion of the leaves,*
care was taken in the collection of samples of the host leaves to
secure them from as nearly as possible the same level as the
parasite. Generally they were gathered from the same branch.

Osmotic concentration was measured by the cryoscopic method.
Sap was extracted after freezingt the tissues to obviate the dif-
ferential extraction of sap, first carefully investigated by Dixon
and AtkinsI and verified by ourselves. §

The results are expressed in terms of freezing-point lowering in
degrees centigrade, corrected for undercooling! | and in atmos-
pheres pressure from a published table. 1|

The actual constants for parasite and host, together with the
habitats of the species and the dates of the determinations, are
given below.

The original collection numbers are retained. The values at
the extreme right are the constants for the parasites. Below these
are given the differences between the concentrations of parasite
and host, the positive sign indicating higher and the negative
sign indicating lower osmotic concentration in the tissue fluid ^
of the parasite.

* Harris, J. Arthur, Gortner, R. A., & Lawrence, J. V. The relationship be
tween the osmotic concentration of leaf sap and height of leaf insertion in trees'
Bull. Torrey Club 44: 267-286. 1917.

t Gortner, R. A., & Harris, J. Arthur. Notes on the technique of the deter-
mination of the depression of the freezing point. Plant World 17: 49-53- 1914-

t Dixon, H. H., & Atkins, W. R. G. Osmotic pressures in plants. I. Methods
of extracting sap from plant organs. Sci. Proc. Roy. Dublin Soc. N. S. 13: 422-433.
1913. Also in Notes from Bot. Sch. Trin. Coll. Dublin 2: 154-172. 1913.

§ Gortner, R. A., Lawrence, John V., & Harris, J. Arthur. The extraction of sap
from plant tissues by pressure. Biochem. Bull. 5: 139-142. pi. i. 1916.

II Harris, J. Arthur, & Gortner, R. A. Note on the calculation of the osmotic
pressure of expressed vegetable saps from the depression of the freezing point, with
a table for the values of P for A = .001° to A = 2.999°. Am. Jour. Bot. i: 75-78.
1914.

^ Harris, J. Arthur. An extension to 5.99° of tables to determine the osmotic
pressure of expressed vegetable saps from the depression of the freezing point.
Am. Jour. Bot. 2: 418-419. 1915.

310 Semi-centennial of Torrey Botanical Club

In several cases more than one mistletoe was taken from the
same host tree. In such cases determinations based on the sap of
the individual parasites may be compared with a single constant
for the host plant, or may be compared with the determination
based on the sample of host leaves nearest the parasite. Differ-
ences which for some reason indicated by the context are regarded
as of doubtful value are bracketed.

Phoradendron calif ornicum Nutt.
Coll. 531, on Acacia Greggii Gray. Aug. 12.

A = 2.00, P = 24.0
For host, A = 1.87, P = 22.5 +0.13, + 1.5

On sandy floor of Sabino Creek, near mouth of Sabino Canyon,
Santa Catalina Mountains.

Coll. 546, on same species, Aug. 15. A = 1.82, P = 21.8
For host, A = 1.68, P = 20.2 + 0.14, + 1.6

Small arroyo on bajada between Tucson and Sabino Canyon,
Santa Catalina Mountains.

Coll. 535, on Cercidium Torreyanum (S. Walts.) Sarg. Aug. 13.

A = 1.99, P = 23.9
For host, A = 1.55, P = 18.7 + 0.44, + 5.2

Coll. 574, on same species, Aug. 19. A = 2.45, P = 29.4
No determination for host.
The first of these collections from Cercidium was taken in the
same locality as the August 15 sample from A. Greggii. The
second was taken from a very small shrub in Sabino Canyon,
on a steep slope about two or three miles below Dam Site. The
lack of daylight precluded a collection of the very small leaves
of the host.

Coll. 267, on Prosopis velutina Wooton, July 6.

A = 2.65, P = 31.8
For host, A = 2.64, P = 31.6 + o.oi, + 0.2

Coll. 538, on same species, Aug. 14. A = 2.29, P = 27.5
For host, A = 2.40, P = 28.8 — o.ii, — 1.3

These two collections were from the same large mesquite tree,
on the edge of a small arroyo on the upper bajada, near the mouth
of Sabino Canyon. The first collection was taken before the
earlier summer rains, the second after they had fallen. The
samples of parasites were from different plants.

Harris: Desert Loranthaceae

311

Phoradendron Coryae Trel., on Quercus
Dr. Trelease notes that certain of these numbers belong in his
form stenophylla, but it has not seemed worth while for present
purposes to separate these from the more typical P. Coryae.

As far as known this species occurs exclusively on Quercus.
The leaves of these desert oaks are so hard that at the mid-summer
season when these determinations were made it was impossible
to express sufficient sap from them to make freezing-point deter-
minations.

On Quercus ohlongifolia Torr.

Coll. 282, Mistletoe i,

A

= 2.23,

P

= 26.7

Mistletoe 2,

A

= 2.09,

P

= 25.1

Coll. 283, Mistletoe i,

A

= 2.01,

P

= 24.2

Mistletoe 2,

A

- 1.88,

P

= 22.6

All four determinations were made on samples collected July
10 on two trees in the Basin, Santa Catalina Mountains.

Coll. 292, on Quercus Emoryi Torr. A = 2.15, P = 25.9

The Basin, Santa Catalina Mountains, July 11.
Coll. 528, on Quercus hypoleuca Engelm. August 10.

A = 1.73, P = 20.8

Coll. 570, August 19. A = 2.26, P = 27.1

Both of these collections were taken from a very large Phora-
dendron on a small oak, growing among the boulders on the edge
of Sabino Creek, in the Basin, Santa Catalina Mountains. The
first determination seemed suspiciously low, and the second sample
was taken from the same individual plant on the 19th to check
the results.

Coll. 355, on Quercus arizonica Sarg. July 19.

A = 2.63, P = 31.5

Near Mud Springs, Santa Catalina Mountains.
Coll. 356, on same species July 20. A = 2.43, P = 29.1

On Mount Lemmon trail, between the Basin and Mud Springs.

Phoradendron macrophyllum (Engelm.) Cockerell
on Fraxinus attenuata Jones
For the host and parasite I am able to give determinations
made in the early spring of 1914 and in the summer of 1916.
The results secured on the desert mistletoe in the earl}' spring of

312 Semi-centennial of Torrey Botanical Club

1914 suggested the desirability of the later studies in Jamaica
and Arizona.

The spring determinations were made from trees in the sandy
(generally dry) bed of Agua Verde Creek, Tanque Verde Moun-
tains.

Coll. 94, Mar. 24, 1914. A = 2.56, P = 30.7

For host, A = 1.39, P = 16.7 + 1.17, + 14.0

The leaves of the host had not yet attained their full size.
The leaves of the parasite, which was in flower, were of course old.
A second visit to this locality was made later to determine whether
the striking difference in the sap of the two sets of leaves might
be due merely to differences in maturity. On April 8 the still
not fully matured leaves of the host gave A = 1.44, P = 17.3,
agreeing well with the values obtained on our first visit.

Coll. 171, April 9, 1914. A = 2.28, P = 27.4

For host, A = 1.63, P = 19.6 + 0.65, + 7.8

These values, from a second large tree in the same locality,
are in excellent agreement with those cited above.

The following determinations were made on samples collected
along Sabino Creek, at the mouth of Sabino Canyon, Santa Cata-
lina Mountains.

Coll. 268, July 7, 1916. A = 3.29, P = 39.5

For host, A = 2.08, P = 24.9 + 1.21, + 14.6

Coll. 270, July 7, 1916. A = 2.46, P = 29.5

For host, A = 1.90, P = 22.8 + 0.56, + 6.7

The first of these samples was taken from a single large mistle-
toe, the second from a number of small plants on another tree.
Coll. 576, August 19, 1916.

Sample B, A = 1.96, P = 23.6

- .29, - 34
Sample C, A = 2.33, P = 28.0

+ 0.08, -f i.o
Sample D, A = 2.53, P = 30.4

+ 0.28, -f 3.4
Sample E, A = 2.69, P = 32.3

-f .44, H- 5.3
All these were taken from the same small host tree, with injured
trunk and several dead limbs, but with the living parts apparently

Harris: Desert Loranthaceae

313

in a perfecf y healthy and vigorous condition*. The sap from the
leaves of the host gave A = 2.25, P = 27.0.

I was able to find only two trees of Fraxinus with Phora-
dendron in the Basin, Santa Catalina Mountains. The results
are:

Coll. 518, Aug. 9. A = 1.97, P = 23.7

For host, A = 1.58, P = 19.0 + 0.39, + 4-7

Coll. 520, Aug. 9. A = 2.15, P = 25.8

For host [A = 3.23, P = 38.7]

For host, Aug. 19, A = 2.15, P = 25.9

[=t o, - O.l]

The two collections of the parasite made on August 9 are very
consistent. The determination for the host in the case of the
second is obviously erroneous, presumably because of contamina-
tion with salt. On August 19 another trip was made to the
Basin for the specific purpose of obtaining another collection from
this tree. The determination based on a sample of that date
gives a distinctly lower value than the first determination from
this tree, but a higher value than the one obtained August 9 from
the first tree. Unfortunately there was not enough of the parasite
for a second collection.

Phoradendron macro phyllum Jonesii Trel. on
Fraxinus attenuata Jones
The following determinations were made from collections along
Sabino Creek, at the mouth of Sabino Canyon, Santa Catalina
Mountains.

Coll. 542, August 14, First Mistletoe A = 1.99, P = 23.9
For host, A = 2.17, P = 26.1 — 0.18, — 2.2

Second Mistletoe A = 1.94, P = 23.4
Same host determination — 0.23, — 2.7

My notes state in regard to the first of these parasites that the
leaves are very yellow. Note that in both cases the determina-
tion for the parasite is lower than is usually the case.

Coll. 575, August 19, 1916. A = 2.51, P = 30.1

For host, A = 1.79, P = 21.5 + .72, + 8.6

August 19, 1916. A = 2.54, P = 30.5

Same host tree, A = 1.76, P = 21.1 + 0.78, + 9.4

314

Semi-centennial of Torrey Botanical Club

Two collections of leaves from the same host tree were made.
The results are in excellent agreement.

Phoradendron macrophyllum on Populus
Four determinations on Populus Wislizenii (S. Wats.) Sarg.
were secured from two large trees, 4-5 feet in diameter, on the
banks of the Rio RiUito, August 12.

Coll. 532, Parasite i, A = 1.96, P = 23.6

+ 0.35, + 4.3
Parasite 2, A = 1.91, P = 22.9

+ 0.30, + 3.6

For host, A = 1.61, P = 19.3
The leaves of the host were taken as near as possible to
Parasite i.

Coll. 533, Parasite i, A = 1.94, P = 23.3

For host, A = 1.79, P = 21.5 + 0.15, + 1.8

Parasite 2, A = 2.13, P = 25.6

For host, A = 1.76, P = 21.2 + 0.37, + 4.4

The two samples of the leaves of the host were taken as near
as possible to the parasites with which they are compared.

III. Discussion of constants

Taking the results for the broad-leaved Phoradendron, P.
macrophyllum and P. macrophyllum Jonesii, on Fraxinus, the data
show that in both of the two determinations made in the early
spring the osmotic concentration of the tissue fluids of the parasite
was considerably higher than that of its host plant.

The average of the two determinations is 29.05 atmospheres
for the parasite as compared with 18.15 atmospheres for the host,
or an average difference of 10.90 atmospheres.

In the summer collections, the data show that there are eight
cases in which the osmotic concentration of the parasite is higher
than that of the host as compared with three cases in which it is
lower.*

* The results for one determination (Coll. 520) are not included because the only-
trustworthy value for the host was obtained on a different date. Two of these
determinations (Coll. 542) are based on two different parasites on the same tree,
with only one determination of the best plant for comparison. In one case the leaves
of the parasite were definitely noted as old. The third exception (Coll. 576) came in
a series in which four samples were taken from a partly dead tree. Possibly this one
of the four plants was not in a normal condition.

Harris: Desert Loranthaceae

315

The average freezing-point lowering of the tissue fluids of the
parasite is 2.383° as compared with 2.041° in the fluids of the host,
or a difference of + -342°.

In terms of osmotic pressure, the average for the sap of the
parasite is 28.63 atmospheres, that of the host is 24.50 atmospheres,
and the average difference between them is +4.13 atmospheres.

The four determinations of P. macro phyllum on Populus are
consistent in indicating higher osmotic concentration in the
parasite. The excess ranges, roughly, from i to 5 atmospheres.
The average for the four comparisons is 23.85 atmospheres for the
parasite and 20.33 atmospheres for the host, a difference of 3.52
atmospheres.

Thus there are for the leafy desert Loranthaceae 14 deter-
minations in which the concentration of the parasite exceeds that
of the host against three cases in which the reverse is true.

Of the five determinations based on P. californicum in which
comparison with the host is possible, four show an excess for the
parasite, but the difference is extremely slight in one case. Far
more work must be done before any final conclusions can be
drawn concerning the relationship of the sap concentration of
parasite and host in the leafless forms.

IV. Recapitulation

This paper presents data toward the solution of the problem
of the water relationships of the desert Loranthaceae.

Three species of the genus Phoradendron, the leafless P. cali-
fornicum and the leafy P. Coryae and P. macrophyllum, have been
investigated on a number of hosts.

The osmotic concentration of the tissue fluids of the Loran-
thaceae of the Arizona deserts is, roughly speaking, twice as great
as demonstrated by similar methods in the tissue fluids of the
species investigated in the montane rain-forest of the Jamaican
Blue Mountains.

In desert Loranthaceae, as in those of the montane rain-forest,
the osmotic concentration of the tissue fluids of the parasite is
generally, but not invariably, higher than that of the host.

These studies will be continued.

INHERITANCE STUDIES IN PISUM. III. THE
INHERITANCE OF HEIGHT IN PEAS*

By Orland E. White

Brooklyn Botanic Garden

As regards height, varieties of peas were classified by Mendel
(*66) as either tall or dwarf. When these two types were crossed,
the Fi generation was either as tall or taller than the tall parent.
The F2 generation consisted of approximately 3 tall : i dwarf
(actual proportions 787 tall : 277 dwarf, or 2.84 : i). All of the
dwarf plants and approximately one third of the tall (28 out of
100) bred true in F3, while the remaining 72 F2 tails gave both
tails and dwarfs in the ratio of 3 : i. These results were inter-
preted as demonstrating a one-factor difference between tall and
dwarf varieties of peas.

However, Bateson (^05), Keeble and Pellew ('10), Lock (^05),
and others found the inheritance of height in certain cases to be
more complex than indicated by Mendel. A class more or less
intermediate in height between dwarfs such as Nott's Excelsior
and Little Marvel and such tails as Scotch Beauty, Champion of
England, and Spate Gold was recognized, to which the name half-
dwarf was applied. Such varieties as First of All, Velocity, and
Express are excellent examples of this class. Bateson ('05) says
half-dwarfs are easily distinguished from tails, and lays particular
emphasis on the zigzag growth of the stem as a marked char-
acteristic of this group. If all varieties ranging in height from
3.5 to 4 or 4.5 feet are to be regarded as half-dwarfs, the above
statement regarding zigzag growth, in the writer's experience,
is only true of one group of half-dwarfs and these have short
internodes, as noted later. Evidently Bateson's half-dwarfs are

* Brooklyn Botanic Garden Contributions No. 20. These studies on the genetics
of Pisum are carried on in collaboration with the Ofifice of Forage Crop Investigations
and the Office of Horticultural and Pomological Investigations, U. S. Department of
Agriculture. For other titles of this series, see "Literature Cited."

316

White: Inheritance of height in peas 317

all of this type, but those of Keeble and Pellew ('lo) and others
belong to at least two types.

Crosses between tails and half-dwarfs gave all tall in Fi and
approximately 3 talis : i half-dwarf in F2 in some cases (Tscher-
mak. *02), while in other cases talis, half-dwarfs, and true dwarfs
have appeared in F2 (Bateson, '05, '09, p. 19). Half-dwarfs with
long internodes crossed with half-dwarfs with short internodes
gave talis with long internodes in Fi and approximately 9 tails : 3
half-dwarfs (short internodes) : 3 half-dwarfs (long internodes) : i
dwarf in F2. The results actually obtained by Keeble and Pellew
(*io) were 114 : 33 : 32 : 13 — expected 108 : 36 : 36 : 12.

Notes taken by the writer for several years on the height,
internode length, and internode number (per plant) of over 200
varieties of peas, grown under similar soil and climatic conditions,
indicate a still greater complexity as regards the inheritance and
classification of height in varieties of this genus.

For example, tall varieties (over 4.5 feet) may be divided into at
least three distinct groups. One type of tall pea has 40 to 60 long
internodes (Scotch Beauty, Spate Gold). Another type has 20 to
40 long internodes (Mammut, Goldkonig, White-Eyed Marrowfat).
The internodes of each of these types average twice or more the
length of the short internodes of the dwarfs. The third type of tall
pea varieties is illustrated by Haage and Schmidt's "Graue Reisen
Schnabel" which has 21 to 30 very long internodes. Variation in
height and internode number among plants of the same true breed-
ing variety is due largely to differences in environment. The
absence of the factor for normal stem (Fa), causing fasciation, also
brings about a shortening of the internodes. Other types of talis
doubtless exist, but the writer's studies have not been detailed
enough as yet to recognize them.

Crosses between the three types of talis, so far as they have
been made, give in Fi and F2 all talis, but tails of different types.
Large number of internodes is usually dominant over the low-
number types. Sufficient data, as yet, are not available to deter-
mine the relation of these types in terms of factorial differences.
Each type of tall undoubtedly represents a separate and distinct
mutation.

As in the case of the tall-growing varieties of peas, the so-called

318 Semi-centennial of Torrey Botanical Club

"half-dwarfs" can be separated into at least two genetic types.
One of these is illustrated by the examples of half-dwarf already
mentioned — Velocity, First of All, and Express. These have
long internodes, similar to the 40-60 and 20-40 long internode
tails, but fewer in number, ranging between 10-20. The other
type of half-dwarf has short internodes, similar to the short
internodes of the true dwarfs. This type is illustrated by the
variety Dwarf Gray Sugar, with internodes ranging in number
from 20 to 40. The variety Autocrat, as studied by Keeble and
Pellew ('10), probably also belongs in this category.

Half-dwarfs with long internodes crossed with short internode
half-dwarfs give long internode tails in Fi and long internode tails,
long internode half-dwarfs, short internode half-dwarfs, and true
dwarfs, approximating a 9 : 3 : 3 : i ratio in F2. Similar results
from such a cross, so far as the writer can judge, have been ob-
tained by Keeble and Pellew ('10), although they have given them
a somewhat different interpretation (see also Lock, '05). Half-
dwarfs of each type crossed with similar half-dwarfs, as expected,
breed true in Fi, F2 and later generations. Half-dwarfs of the
short internode type crossed with the 20-40 long internode tall
type give in Fi tails with long internodes, which in F2 produce a
population approximating 3 talis with long internodes : i half-
dwarf with short internodes. The difference in height between
such tails and such half-dwarfs is due largely to internode length,
the number of internodes in each type being approximately the
same. Half-dwarfs with long internodes crossed with talis with
long internodes (20-40 type) give talis in Fi and populations
approximating 3 tails : i half-dwarf in F2, both, of course, with
long internodes, the difference between them in this case being
due to internode number.

True dwarfs (6 inches to 3.5 feet high) in peas all have short
internodes, ranging f om 8 to 20 in number. Laxtonian, Nott's
Excelsior, and several French varieties obtained through the
courtesy of Phillipe Vilmorin are excellent examples. When
crossed with the various types of tails, the Fi generation always
consists of tails with long internodes, although large number of
internodes may not be completely dominant over small number
of internodes in certain cases, e. g., Pois nain a chassis, tres hatif

White: Inheritance of height in peas

319

crossed with Wachs Schwert (40-60 internodes tall). In such
crosses, the Fi generation is to be regarded as intermediate, so
far as the gross character height is concerned, although this inter-
mediate condition is brought about through incomplete dominance
of high over low internode number. Tschermak ('01) also men-
tions obtaining intermediates in Fi, as regards height, from crosses
of tails and dwarfs, though he says nothing about internode

Fig. I. generation tall and dwarf segregates from cross of tall X dwarf.
Photo from cultures of E. M. East.

number or length. The F2 generation from tall X dwarf or its
reciprocal consists of four classes — talis with long internOdes, half-
dwarfs with either long or short internodes, and true dwarfs.
These approximate a 9:3:3:1 ratio as in the case of short
internode half-dwarf X long internode half-dwaif. Laxton ('06)
obtained practically these same results. This is the cross usually

320 Semi-centennial of Torrey Botanical Club

made to illustrate the inheritance of height and probably the one
made by Mendel for the same object. What is usually meant by
geneticists, so far as the writer can learn, when discussing inherit-
ance of height in peas, is really the inheritance of difference in
internode length. Hence all short internode varieties or segre-
gates, irrespective of actual height or number of internodes, are
classed as dwarfs, while all long internode varieties or segregates,
irrespective of number of internodes, are classed as tails. Classi-
fied in such a manner, the population from such a cross as de-
scribed above would have approximately 3 tall (long internodes) : i
dwarf (all short internodes) . The long internode half-dwarfs would
be called talis, while the half-dwarfs with short internodes would
be called dwarfs (see Bateson et al., '05; Lock, '05, '08; Laxton,
'06).

Crosses between half-dwarfs with long internodes and true
dwarfs gave half-dwarfs in Fi and approximately 3 half-dwarf
(long internodes) : i dwarf (short internodes) in F2 in the writer's
experiments. Bateson ('05) commonly obtained intermediates in
Fi from crosses between half-dwarfs (presumably short internodes)
and dwarfs.

The simplest interpretation of the above data involves the
presence and absence of at least five genetic factors for height,
two of which primarMy determine the differences in internode
length and three of which are largely responsible for the hereditary
differences in number of internodes. These with their expression
may be represented as follows:

Le = long internodes

Lei = very long internodes

T = 20-40 internodes

Ti = 40-60 internodes

T2 = 20-30 internodes

Absences
le = short internodes
t = 10-20 internodes
Le and T have been referred to in previous numbers of this
series (White, '17 a and b), Le being the factor isolated by Mendel
and confirmed by many later workers. T is referred to by Keeble

White: Inheritance of height in peas

321

and Pellew (*io) as the factor for robust stems, but in the writer's
interpretation of their results, it determines the difference in
internode number.

On the above interpretation, the factorial composition of the
three classes of talis would be:

(1) LeT = 20-40 long internodes.

(2) LeTi = 40-60 long internodes.

(3) LeiT2 = 20-30 very long internodes.
The factorial composition of the half-dwarfs would be:

(4) Let = 10-20 long internodes.

(5) leT = 20-40 short internodes.

The true dwarfs on this scheme would represent the absences of
Le and T or (6) let.

Sufficient data have not yet been accumulated to determine
in any detail the relations of these factors to each other except in
the case of Le and T. Varieties with formula (i) crossed with
(4) should and do give all long internode tails in Fi and talis and
half-dwarfs (long internode) in F2. Combination (i) X (5) gives
long internode tails in Fi and approximately 3 talis (long inter-
nodes) : I half-dwarf (short internodes) in F2. Combination
(i) X (6) gives all long internode talis in Fi and an F2 population
approximating 9 tall (l.i.) : 3 hd. (Li.) : 3 hd. (s.i.) : i dwarf (s.i.).
The two half-dwarf types, (4), (5) crossed with each other give
all long internode tails in Fi, but an F2 population similar to (i) X
(6). Half- dwarf varieties (5) X dwarfs give intermediates m r 1
in some cases. The writer has no data on this cross as yet. Half-
dwarf varieties with the formula Let (4) crossed with dwarfs (6)
give all long internode half-dwarfs in Fi, and approximately 3 half-
dwarfs : I dwarf in F2.

Critics of Mendelian methods and conceptions will say again,
as the above results are noted, "another unit-character has been
split up." But the writer wishes to emphasize that with the
same genetic pea material that Mendel and others have used to
obtain the F2 ratio of 3 talis : i dwarf, the same results will still
be secured. The difference in interpretation has come from more
detailed studies and the inheritance of height in peas has become
complex only because of studies on new or distinctly different
material, the characters of which, there is reason to believe, are
due to distinct mutations.

322 Semi-centennial of Torrey Botanical Club

A large series of crosses involving height is in pt ogress, and
the data from these will be published in detail.

LITERATURE CITED
Bateson, W., Saunders, E. R., Punnett, R. G., Hurst, C. C, & Killby,

Miss. Reports to the Evolution Committee of the Royal Society.

1902-1906. See Rep. 2: 55-80. 1905, for peas.
Bateson, W. Mendel's principles of heredity, ix + 396. /. 1-37. pi.

1-6. Cambridge (Eng.) Univ. Press. 1909.
Keeble, F., & Pellew, Caroline. The mode of inheritance of stature

and of time of flowering in Pisum sativum. Jour. Genet, i: 47-56.

1910.

Laxton, W. The cross-breeding and hybridization of peas and of
hardy fruits. Rep. 3d Internat. Conf. on Genetics, London,
468-473. 1906.

Lock, R. H. Studies in plant-breeding in the tropics. II. Ann. Roy.
Bot. Gard. Peradeniya 2: 357-414. 1905.

Lock, R. H. The present state of knowledge of heredity in Pisum.
Ann. Roy. Bot. Gard. Peradeniya 4: 93-111. 1908.

Mendel, G. Versuche iiber Pflanzenhybriden. Verh. naturf. Ver. in
Briinn, 4: 3-47. 1866. See also Bateson (1909) for English trans-
lation.

Tschermak, E. Weitere Beitrage iiber Verschiedenwertigkeit der
Merjcmale bei Kreuzung von Erbsen und Bohnen. Ber. Deut.
Bot. Gesells. 19: 35-51. 1901. (For peas, see 35-45.) Same
paper in Zeitschr. Landw. Versuchsw. in Oesterr. 4.

Tschermak, E. Ueber die gesetzmassige Gestaltungsweise der Misch-
linge. Fortgesetzte Studien an Erbsen und Bohnen. Zeitschr.
Landw. Versuchsw. in Oesterr. 5: 781-861. 1902. (For peas, see
789-819.)

White, O. E. Inheritance studies in Pisum I. Inheritance of coty-
ledon color. Am. Nat. 50: 530-547. 1916.

White, O. E. Idem. II. The present state of knowledge of heredity
and variation in peas. Proc. Am. Phil. Soc. 56: 487-588. 1917a.

White. O. E. Idem. IV. Interrelation of the genetic factors of
Pisum. Journ. Agr. Research 11: 167-190. tables I-IV + append,
tables 1-27. 191 76.

CENTROSOMES DURING EARLY FERTILIZA-
TION STAGES IN PREISSIA QUADRATA

By Margaret Graham

Hunter College
(with plate 8)

The problem as to the behavior of the centrosome during cell
and nuclear fusion cannot be said to be settled for either plant or
animal cells and from this standpoint I am engaged in an investiga-
tion of the processes connected with fertilization in Preissia
quadrata (Scop.) Nees.

My material was collected from gorges around Ithaca and
prepared for sectioning at Cornell University. I am indebted to
Professor G. F. Atkinson for the privileges of the botanical
laboratory there, where for several years I experimented with
methods of handling the plants and with various killing reagents.
I also acknowledge the privileges of the laboratory in the Cornell
Medical College at Ithaca. The material was killed in the field
in a modified Flemming solution. The sections were stained and
studied at the botanical laboratory at Columbia University and
were examined by Professor R. A. Harper, to whom I am indebted
for a critical examination of my stained preparations.

Dr. Osvaldo Kruch ('90) observed many eggs of Riella Clau-
sonis with one antherozoid in the cytoplasm. He also observed
that both nuclei before fusion were approximately of the same size.
Fusion of the pronuclei was not observed, nor were astral rays and
centrosomes. Since this article by Kruch no other has appeared
on fertilization in the liverworts.

I shall describe here only the stages after the egg has been
penetrated by the antherozoid and when the pronuclei are already
near together. During this stage the cytoplasm of the egg of
Preissia quadrata is plainly made up of two zones. The inner zone
is granular with rounded bodies forming a dense aggregate that
lies in masses around the pronuclei and among the rays of the

323

324 Semi-centennial of Torrey Botanical Club

centrospheres. A small amount of the same material also clings
to the cytoplasmic fibers at the periphery of the cell. This dense
cytoplasm may appear more or less alveolar at this stage. Prior
to fertilization the whole cytoplasm of the egg has this consistency.
The outer zone of the cytoplasm is coarsely vacuolar. The films
between the vacuoles are very thin and delicate. A few larger
and quite dense homogeneous granules are scattered through both
zones of the cytoplasm. The egg nucleus and the nucleus of the
antherozoid are plainly differentiated by their size and in figures
I, 2, and 3 are shown lying in the central part of the cell. In
figures I and 3 their nuclear membranes are in contact.

In the dense cytoplasm at the opposite poles of the egg nucleus
astral rays are seen converging upon small dense rounded bodies,
the centrosomes (figs, i and 2). These rays extend long distances
through the cytoplasmic ground substance. They may pass close
to the nuclear membrane or may touch it. A fiber radiating from
the centrosome at the upper part of figure i touches the nuclear
membrane of the antherozoid; another radiating from the same
centrosome touches the outer membrane of the egg nucleus. The
astral rays make up an open aster with few rays; but they are
very definite fibers easily distinguishable from other cytoplasmic
structures. Peripherally they end rather abruptly and have no
conspicuous physical connection with any of the other elements of
the dense cytoplasm. It is quite possible that there are other
shorter and more delicate rays, but I have drawn only those which
are plainly differentiated. All these fibers center on a centrosome
which seems to be a single body.

Figure 3 shows a slightly earlier stage of fertilization. A
centrosphere lies above the egg nucleus, the rays extending to its
membrane. Figure 4 is another section of the same egg and shows
a second centrosphere.

One or more large dense bodies lie among the astral rays
a short distance from the center, but they do not constitute a part
of the region on which the rays center (figures 1-4).

Centrosomes and asters have been demonstrated in the vegeta-
tive cells of liverworts by Farmer and Reeves ('94) in the germinat-
ing spore of Pellia epiphylla and by Van Hook (*oo) in the cells
of the stalk of the archegoniophore of Marchantia polymorpha.

Mem. Torrey Club

Volume 17, plate 8

GRAHAM: CENTROSOMES IN PREISSIA

Graham: Centrosomes in Preissia

325

Centrosomes have been observed in the divisions just preceding
spermatogenesis and as blepharoplasts, during the metamorphosis
of the antherozoid mother cell. From my studies it is evident
that they are also present in the fertilized egg at the time when
the pronuclei are paired.

LITERATURE CITED

Farmer, J. Bretland, & Reeves, Jesse. On the occurrence of centro-
spheres in Pellia epiphylla, Nees. Ann. Bot. 8: 219-224. 1894.

Kruch, Osvaldo. Appunti sullo sviluppo degli organi sessuali e sulla
fecondazione della "Riella Claiisonis Let." Malpighia 4: 403-430.
1890.

Van Hook, J. M. Notes on the division of the cell and nucleus in
liverworts. Bot. Gaz. 30: 394-398. 1900.

explanation of plate 8

The figures were drawn with the aid of a Bausch & Lomb camera-lucida, the
drawing being at the level of the base of the miscrocope; Zeiss 1.8 mm. oil-immersion
objective, 1.25 N. A., and oc. 4. Magnification about 1431 diameters.

Figures i and 2 are sketches of drawings made for another article.

Fig. I. A section of the egg cell. The membranes of the egg and antherozoid
nuclei are in contact. Centrospheres lie at the poles of the egg nucleus.

Fig. 2. A section of the egg cell in which the nucleus of the antherozoid lies in
the cytoplasm near the egg nucleus. A centrosphere appears at the poles of the
egg nucleus.

Fig. 3. A slightly younger stage of fertilization. One centrosphere is shown
above the egg nucleus.

Fig. 4. Polar view of a centrosphere lying in the cytoplasm of another section
of the egg shown in figure 3.

ORIGIN AND DEVELOPMENT OF THE
LAMELLAE IN SCHIZOPHYLLUM
COMMUNE

By J. F. Adams

Columbia University
(with plate 9)

Fries ('15) in establishing the genus Schizophyllus which he
later (*2i) changed to Schizophyllum says that it is to be dis-
tinguished from Agaricus and Merulius by the longitudinal
splitting of the gills with the resulting halves becoming revolute.

Hoffman (*6o) describes the carpophores as consisting of
lamellar systems which correspond to the crenatures of BuUer
('09). The smaller secondary gills are often but not always
divided. He believes that the upper surface of the carpophores
should be regarded as a "pellicula" comparable to the velum.

Winter ('84) and de Bary ('87) add nothing of consequence
to the description by Hoffman.

Fayod ('89) includes Schizophyllum in the tribe "Panoides"
with Panus as representative of the gymnocarpus type. He noted
that the young lamellae are entire and held that the splitting was
hygroscopic.

Buller ('09) is quite explicit as to the origin of the gills. He
says in substance:

"The under layer of the pileus is produced downwards to form
the gills. Whilst a pileus is extending by marginal growth, the
interlamellar spaces gradually widen. When a space has attained
a certain width, it becomes divided into two down the middle,
owing to the formation within it of a new gill which arises as a
short median downgrowth from the pileus flesh. The splitting of
the gills permits the hymenial surfaces being protected during
periods of drought. The recurving of the gill plates may be partly
explained when a fruit body dries up by the cell walls of the hy-

326

Adams: Origin of lamellae in Schizophyllum 327

menial and sub-hymenial layers contracting much more strongly
in the vertical direction than those of the tramal layer."

There is no clear account of the early stages of gill formation
in the literature and I have undertaken to fill this gap by the
present study. The nature of the whole carpophore also becomes
much clearer with a knowledge of the method of origin of the gill.

Materials

Carpophores of Schizophyllum commune for this study were
collected in the field and also grown on agar. Lima-bean agar,
prune agar, and dung agar were favorable media. The cultures
were grown in flasks of 50 c.c. and 250 c.c. capacity. Cultures
were started from immature carpophores collected in the field,
which were washed in distilled water before being transferred to
the flask.

Within five days after the carpophores are planted on it the
agar surface is covered with a dense white growth of mycelia.
The hyphae in cultures are conspicuously branched and micro-
scopical examination shows numerous clamp connections. The
hyphae average 3 m in diameter.

Dense aggregations of buttons often develop over the agar
surface. From the cut end of the carpophores and from their
margins small buttons are first formed. A number of the buttons
when they are in dense clusters fail to mature. Numerous ab-
normal forms appear, as has been reported also by Miss Wakefield
('09). In some cases an immature carpophore that is transferred
to agar will continue its normal development by marginal growth
and form a large normal fruit-body (pl. i, fig. i).

Various stages in the development of the carpophore were
fixed for study in Flemming's medium mixture, dehydrated, and
embedded in 52 parafiin. Sections were cut from 6 /x to 12 u
in thickness.

The young carpophore

The young button appears as a dense, globular mass of inter-
twined hyphae raised above the substratum. As noted, they
often appear in clusters. These carpophore primordia later be-
come differentiated into cylindrical outgrowths which enlarge

328 Semi-centennial of Torrey Botanical Club

radially at their outer end. They vary in length from about i to
10 mm. and up to 5 mm. in width at the tip.

They are clothed with a loose outgrowth of hyphae from their
earliest appearance. These hyphae are thick-walled and no
branching was observed. Clamp connections or cross-walls were
not found in these superficial hyphae.

Transverse vertical sections of young buttons show a dense
homogeneous mass of intertwined hyphae developing parallel
with the elongating axis of the primordium. The hyphae show
conspicuous clamp connections. The carpophore primordia are of
leathery consistency and show no evident differentiation until the
appearance of the hymenium primordium.

The hymenium

In young buttons the first appearance of the fruiting layer is
just below the upper end. It consists of a growth of densely
staining hyphae. Horizontal sections show this first structure or
plectenchyma of hyphae centrally located. With further growth
of the carpophore primordia they become somewhat elliptical in
outline. The individual hyphae now become oriented, their free
ends converging towards the center of the mass.

The first gill cavity is formed in this hymenium primordium.
The oldest primordia are in the central part and the youngest
towards the sides, showing the order of their formation (text-
fig. A, NO. 2).

The whole carpophore enlarges and as result of growth ten-
sions the gill cavities appear in the center of these primordia of the
hymenium. The gill cavities are lined from the first by a palisade
layer. The palisade layer is increased by the intercalary addition
of new elements with the further increase in size of the gill cavity.
The elongated cells composing the palisade layer stain deeply and
appear to arise as a system of short branches from the subjacent
hyphae which are to form the trama.

The gill cavities thus represent from the first the space between
the adjacent halves of two lamellae. The hymenia are oldest
toward the base and less undifferentiated toward the tip of the
carpophore.

In two instances gill cavities were observed developing in the

Adams: Origin of lamellae in Schizophyllum 329

trama of the matured lamellae. It was observed several times
that the inner surface of the gill cavity before maturity was not
completely covered by the palisade layer. This condition is found
usually along the lower edge of the gill cavity, but the condition
becomes normal with the maturity of the gill cavity (pl. i, fig. 4).
The wall between two adjacent gill cavities is occasionally quite
weakly developed owing to the close proximity in their origin.

Fig. a. Diagrams of sections illustrating different stages of development of
the early carpophore, i. Transverse vertical section of a button in which no
differentiation has appeared. 2. Section cut as above showing later stage with four
endogenous and separate gill cavities, the lateral ones younger. 3. A longitudinal
median section through a young carpophore showing one of several gill cavities which
have opened longitudinally below. a, Region of extension of the hymenium pri-
mordium. 4. Transverse vertical sections thrpugh the apical growing region of the
carpophore, showing the origin of the crenatures, two of which are already separated
by clefts extending to the dorsal surface of the carpophore. The middle and outer
crenatures at right show gill cavities comparable to those in Fig. 4, PI. i. In the
case of the others, similar gill cavities appear in sections nearer the base of the
pileus. 5. Transverse vertical section like those of Buller through the median
portions of a carpophore, showing a series of young lamellae, the gill cavities open
below except in the case of the one at the right. The gills already show more or
less of the characteristic splitting.

In such instances the separating wall, which would normally
become a lamella, gradually thins out and disappears. Instead
of two gill cavities normally maturing they become one by the
abortion of the separating wall.

The lamellae

The gill cavities split along their lower edge and lamellae are
thus completed. They consist, as noted, of the adjacent walls of

330 Semi-centennial of Torrey Botanical Club

two gill cavities which originate endogenously as tubes in the
substance of the carpophore.

The tissue above the gill cavities already formed increases by
intercalary growth and in general the gill cavities lie much nearer
the ventral than the dorsal surface of the carpophore. The tissue
below the gill cavity appears gradually to become looser in texture
as the gill cavity gets larger and this favors the splitting by which
the edges of the gills are set free.

Owing to the method of their origin the margins of the lamellae
are never entire, but appear irregular and frayed out. The final
splitting of the lamella is apparently a hygroscopic phenomenon
as described by Buller Cog). The trama is continuous with the
tissue of the pileus above the gill. In transverse sections of the
lamellae the split is seen to be parallel with the elongated hyphae
of the trama, as seen in plate i, figure 7.

Further development of the carpophore

The growth of the carpophore is marginal and the lamellae
are extended in length by the elongation of the gill cavities and the
palisade layer in the interior pari passu with the development of
the carpophore.

In the young immature carpophores after several lamellae
have been formed the margin becomes divided into the crenatures.
These crenatures are due to the development of a cleavage lamella
' splitting in certain cases clear through the dorsal surface of a pileus
(text- FIG. A, NO. 4). They not only include the primary gills but
allow for the origin of new gills on either side in the usual way.
The later-formed lamellae are narrower and thus we get the lamellar
systems of Fries and the fasciculi of gills of Buller. The pileus
enlarges by the continued growth of the primary lamellae and the
successive development of additional lamellae, all of which have
their origin from gill cavities in the manner described. The more
central crenatures are the oldest and the younger are towards
the sides of the pileus.

In mature carpophores it is often observed that a short gill
appears isolated between the right and left halves of the adjacent
lamellae (text-fig. B, c). At the point where such a lamella
originated the crenature was increasing in width. The new

Adams: Origin of lamellae in Schizophyllum 331

lamella was continued from its point of origin so long as the
gro\Vth of the crenature allowed for its development. After
attaining a certain width the crenatures gradually become nar-
rower and thin out at the margin. Under such conditions of
limited growth some of the enclosed lamellae in the crenature can
no longer be continued and do not reach the margin of the crena-
ture. In the mature carpophore the margin is always thinned
out and the gills thus become lanceolate in form.

Discussion

In recent years considerable advances have been made in our
. knowledge as to the origin and development of the lamellae in

different members of the Agaricaceae. The endogenous origin
of the lamellae has been firmly established for a number of forms.

In Schizophyllum the lamellae originate by the formation of
endogeneous gill cavities in a fashion similar in principle to that
which Levine (*I4) finds in Coprinus micaceus. They are de-
veloped simultaneously in C. micaceus, while in Schizophyllum
they are successively formed. In both cases a series of gill cavities
are produced which represent the space between the adjacent
sides of a pair of lamellae. There is no general gill cavity into
which the lamellae grow downward. In Schizophyllum, owing
to its habit of growth, this method of origin of the gill cavities as

332 Semi-centennial of Torrey Botanical Club

independent tubes is diagrammatic in its simplicity and whether
Schizophyllum is a progressive or a reduced type we have in it the
evidence that gills in their essential nature are hymenium-bearing
plates between independently originating endogenous gill cavities.

To Professor R. A. Harper and Dr. M. Levine I wish to
express my sincere appreciation of their kindly criticisms and
helpful suggestions.

LITERATURE

Bary, A. de. Comparative morphology and biology of the fungi,

mycetozoa and bacteria, 302. Oxford, 1887.
BuUer, A. H. R. Researches on fungi, 113-119. London, 1909.
Fayod, M. V. Prodrome d'une histoire naturelle des Agaricinees. '

Ann. Sci. Nat. Bot. VIL 9: 331-333. 1889.
Fries, E. Observationes mycologicae, 103. 1815.
Fries, E. Systema mycologicum, i : 330. 1821.

Hoffman, H. Beitrage zur Entwickelungsgeschichte und Anatomic der

Agaricinen. Bot. Zeit. 18: 389-397. i860.
Levine, M. The origin and development of the lamellae in Coprinus

micaceus. Am. Jour. Bot. i: 303-322. 1914.
Wakefield, E. M. Ueber die Bedingungen der Fruchtkorperbildung

sowie das Auftreten fertiler und steriler Stamme bei Hymeno-

myceten. Naturw. Zeitschr. Forst.-u. Landwirtsch. 7: 521-551.

1909.

Winter, G. Die Pilze Deutschlands, Oesterreichs und der Schweiz. In
Rabenhorst, L., Krypt.-Fl. i: 493. 1884.

EXPLANATION OF PLATE 9

Fig. I. A carpophore thirty-four days old, grown from a piece of Schizophyllum
commune transferred from a branch of birch. 6/13 of the natural size.

Figs. 2 and 3. Transverse vertical sections through three crenatures of a carpo-
phore with gill cavities in different stages of development. Fig. 3 is further back
from growing region than Fig. 2. The gill cavities are seen to lie nearer the ventral
than dorsal surface. Photomicrograph X 26.

Fig. 4. Transverse vertical section through the growing region of a carpophore,
showing only three crenatures as indicated by the two ventral furrows. The first
crenature to the right shows undifferentiated tissues. In the middle crenature
appears an imperfectly developed gill cavity and a closed hymenium primordium in
which a gill cavity will appear. The crenature on the left shows two mature gill
cavities. Photomicrograph X 43.

Figs. 5 and 6. Sections cut as above, slightly oblique through young carpophore
in which gill cavities are open, showing three gills that have split along their lower

Mem. Torrey Club

Volume 17, plate 9

SCHIZOPHVLLUM COMMUNE Fr.

Adams: Origin of lamellae in Schizophyllum 333

edge. The irregular and frayed out margins of the lamellae are already evident.
Fig. 6 is further back from the growing region than Fig. 5 and a small closed hymenium
priraordium appears in the lower lateral margin to the right. Photomicrographs.
Fig. 5 X 35; Fig. 6 X 26.

Fig. 7. Section cut as above through the median portion of a carpophore
showing a series of lamellae. The gill cavities are opened, forming the lamellae
which are here seen split. Photomicrograph X 38.

STATISTICAL STUDIES OF FLOWER NUMBER
PER HEAD IN CICHORIUM INTYBUS: KINDS
OF VARIABILITY, HEREDITY, AND
EFFECTS OF SELECTION

By a. B. Stout and Helene M. Boas

The New York Botanical Garden
(with plates 10-13)
Table of contents

Introduction 335

Review of literature pertaining to number of flowers per head in

the compositae 337

I. Ludwig's evidence that flower-number is a specific character 337

II. Theories regarding the evohition and development of flower number

per head 339

III. Special criticisms of facts and theories 342

IV. Evidence of intraseasonal variability 344

V. Observations and experiments on the influence of environment .... 345

VI. Studies of interannual variability 347

VII. Evidence that position is a factor in partial variability 348

VIII. Special views regarding heredity, differentiation, and symmetry,

held by Pearson and by Bateson 351

IX. Evidence of hereditary variations 354

X. Previous study of flower number in Cichorium Intybus 356

XI. Summary 356

The PROBLEMS IN Cichorium Intybus 357

Material and methods 358

Presentation of data for flower number in chicory 362

I. General survey of the kinds of variability present 362

1. Partial variability 362

A. Intraseasonal 362

B. Interannual 366

2. Individual variability 379

II. Statistical treatment of data 380

III. Detailed presentation of data bearing on:

1. Relation of length of blooming period to rate of change 385

2. Significance of the range of variability in flower number per head. 391

3. Relation of flower number per head to position of heads 393

A. Descriptive studies regarding position 393

B. Statistical studies regarding position 395

C. The phenomenon of intermittent annual growth 407

334

Stout & Boas: Statistical studies in Cichorium 335

4. Individual variations from the usual performance as to seasonal

decrease 410

5. Variation in partial variability with the age of a plant 412

6. Variation in range and distribution of modal numbers 419

IV. Characteristics of flower number in the variety red-leaved Treviso. . . 421

V. Characteristics of flower number in a hybrid generation of a cross

between plants of red-leaved Treviso and a wild plant (A) 427

VI. Selection and heredity; characteristics of different progenies and
various lines of descent derived from a cross between a wild
plant (A) and plants of the variety Barbe de Capucin 428

1. General comparison of progeny with immediate parents as to

values of a and b 428

2. Detailed analysis of six races as to inheritance of flower number

per head 432

A. A semi-dwarf race (race i) 432

B. A dwarf race (race 2) 435

C. Race with intermittent growth (race 3) 436

D. A brittle-stemmed race (race 4) 437

E. A semi-dwarf bushy race (race 5) 439

F. A tall-growing race (race 6) 441

Discussion and conclusion 446

Bibliography ; 454

Explanations of plates 458

INTRODUCTION

Much study and speculation have been directed to the so-
called fluctuating variations in the effort to determine their sig-
nificance in development, in heredity, and in evolution. One of
the principal reasons why such investigations have not been more
conclusive undoubtedly rests in the difficulty of properly grading
each of the individuals that comprise the species, the population,
the generations, or the lines of descent to be analyzed.

This difficulty is frequently present in the more readily meas-
urable or quantitative characters, especially when the estimate
of the individual involves measurements of homologous organs
among which there is also fluctuating variability. There are in
such cases two grades of fluctuating variability : one is an individual
variability (using this term as defined by de Vries, '01, p. 37, to
refer to differences between individuals as" such), the other is a
partial variability (de Vries, '01, p. 37) which is a variability
within the individual. The latter exists within the former and
when present the estimate of the individual and of individual
variability involves partial variability and is only adequate to
the degree that the determination of partial variability is adequate.

336 Semi-centennial of Torrey Botanical Club

It will also readily be recognized that partial variability may
be complicated further by the development of homologous organs
or parts of the organs at different times : the variations may involve
organs which mature over a somewhat extended period either in
different years or in a single season. A marked example of the
latter is to be seen in perennials which produce a new crop of
leaves, flowers, and fruit in consecutive years, giving inter-year
variations which may involve the age of the individual. Further-
more, annuals, biennials, and perennials most frequently show a
more or less extended period of development in a single season,
during which homologous organs come into maturity with quite
different positions on a plant and at quite different times.

All these aspects or factors of partial variability complicate
the adequate determiniation of the expression of the capacities of
an organism, and, it would seem, are factors that must be consid-
ered in an analysis of individual variability upon which any dis-
cussion of variation, heredity, or selection is based.

Furthermore, partial variability (as well as individual vari-
ability) may involve different kinds or grades of characters such
as: (i) size, as of seeds, fruit, etc.; (2) chemical properties, as
color, sugar-content, etc.; (3) number of homologous organs
grouped together in a specialized structure such as the number of
flowers in an inflorescence or the number of petals or sepals in a
flower. Bateson has used the term "meristic" to apply to varia-
tion in number, as distinguished from substantive (such as color)
variations which are more qualitative. The term "meristic" is,
however, not applied solely to partial variability.

Further, partial variability may involve elements of differen-
tiation which may not be suspected from random observations
and collections of data, but which may constitute a source of
error. The relation of fluctuating variability to differentiation
is by no means clear, as the discussions of Pearson ('01 and '02a)
and Bateson ('03) indicate. Very generally, however, the element
of differentiation has been entirely disregarded in statistical
studies of characters exhibiting wide partial variability.

The development of varying numbers of flowers in the different
heads produced on a plant of a species of Compositae illustrates a
type of partial meristic variation in the total number of flowers

Stout & Boas: Statistical studies in Cichorium 337

per head. If there be a noted differentiation among the flowers
in a head, the number of both ray- or disk-flowers may vary.
The compact inflorescence terminating a branch is itself considered
as a unit structure in the study of partial variability upon which
the value of the individual is to be based. The study of flower
number in the Compositae, particularly the variation in the num-
ber of ray-flowers, has received much attention. The biometrical
treatment inaugurated by Quetelet and Galton has been utilized in
the descriptions and analyses of the performance of a species as a
whole and of various populations and lines within a species. These
studies have not always recognized the extent and kinds of individ-
ual and partial variability that may exist, or the degree to which
differentiation may be recognized as operative in what appear as
chance variations. The reasons for this will be obvious from the
following considerations of the points of view and aims which
influenced and guided different investigators in their studies.

REVIEW OF LITERATURE PERTAINING TO NUMBER
OF FLOWERS PER HEAD IN THE COMPOSITAE

I. LUDWIG'S EVIDENCE THAT FLOWER NUMBER IS A SPECIFIC

CHARACTER

Ludwig was one of the first to investigate intensively botanical
subjects by biometric methods and his studies were especially
directed to problems relating to number of flowers per head in
various Compositae. He also investigated species of Umbellif-
erae as to the number of rays per umbel. His interest at first
centered on the analysis in Galtonian terms of flower number
for species as such. In the Compositae he studied those species
which have both ray- and disk-flowers, but confined his obser-
vations almost exclusively to ray-flowers.

Ludwig did not realize the importance of the behavior of the
individual plant, hence his method involved counts of heads col-
lected indiscriminately. From such data curves were constructed
to determine the behavior of the particular species, and especially
to determine the highest mode or the maximum frequency of
number per head. In the collection of data therefore, for the
most part, no special recognition was given to individual variability
and, of course, partial variability was ignored.

338 Semi-centennial of Torrey Botanical Club

Ludwig ('95) published a summary of all his studies, which
involve twenty-six genera and over sixty species of Compositae.
The maxima determined for the number of ray-flowers per head
are given for the different species. Often the counts were made
from only a few flower heads so that the maximum is frequently
not to be considered as established, as Ludwig himself fully recog-
nized.

The largest number of counts made of any composite was of
Chrysanthemum Leucanthemum. Data for 17,000 heads were col-
lected in Europe at different places and at different times of the
year over a period of several years. Ludwig concludes that 21
ray-flowers per head is the number characteristic of this species.
In no instance does he give data showing that a race with a dif-
ferent maximum was found growing in isolation, but he considered
that such races were indicated by secondary maxima obtained
from data of mixed populations.

Data for ray-flower number of approximately 12,000 heads of
Bellis perennis (Ludwig, '98) were presented to show that maxima
may occur on the Nebenzahlen " of the Fibonacci series, i. e.,
39, 42, 55, 63, which were considered as "duplica" or "triplica"
of various numbers of the main series. The only maxima that
can be considered as at all established for Bellis perennis are 34
(one of the main series) and 42 (twice 21). Study was also made
of total flower-number per head in B. perennis: 860 heads were
counted and for these the range was from 63 to 233, a very large
range of variability.

For Chrysanthemum inodorum (Ludwig, '95), data for 1,000
heads reveal a range in ray-flower number from 10 to 32 and. a
pronounced maximum at 21.

The data for Chrysanthemum segetum (Ludwig, '04) are of
special interest as this is the species in which de Vries later made
selection experiments. Ludwig reports 750 counts made in one
locality and 250 made in another. Of the 750 from the one lo-
cality, 150 were made on July 29 and 600 on August 30. The 250
counts in the other locality were made on August 16. This is
the only instance where Ludwig reports the dates of collections.
However, the two series show no significant differences either in
maxima or in range of variability; for the 750 counts the maximum

Stout & Boas: Statistical studies in Cichorium 339

is at 13 and the range from 11 to 23; for the 250 counts the
maximum is 13 and the range from 10 to 23. Ludwig thus
failed to find any evidence that here differences in flower number
may appear according to age of a plant or to different stages in
the period of bloom, such as will be reported in this paper for
Cichorium.

For Anthemis arvensis Ludwig ('95) reports that for 1,802
flower heads the number of ray-flowers ranged from 2 to 21 with a
maximum at 8 and some indications of a secondary one at 12-13.

In only two cases did Ludwig ('87) make a study of individual
plants. In Achillea Ptarmica the modes for number of ray-
flowers were determined for each of 79 plants; 30 plants gave a
mode at 8; for 17 the mode was at 13; and for the other 32 the
modes were about evenly distributed at numbers ranging from
9 to 12 inclusive. The total number of flower heads counted was

I, 048 and the number of ray-flowers ranged from 6 to 15, with a
maximum at 8. For Senecio Jacohaea the distribution of ray-
flower number for 5 individual plants was studied. The range
here was small, being from 12 to 14 and the total number of heads
counted was 109.

In only a few cases does Ludwig consider the number of all the
flowers produced per head. Such a study for Bellis perennis has
been noted above. For Senecio nemorensis he presents ('96) data
for 500 heads, which show that the total flower number per head
ranged from 15 to 26 with maxima at 18 and 21, and the maximum
for the ray-flowers in 357 heads of these was 5, with a very small
variability. In these studies, however, the behavior of individual
plants was not considered. Studies of total flower number were
made in Senecio Fuchsii ('96), Centaurea Cyanus ('96), C. Jacea
('96), and Solidago Virgaitrea ('96), but the number of observa-
tions is too small for the determination of the specific character-
istics, which was the object of his interest and studies.

II. THEORIES REGARDING THE EVOLUTION AND DEVELOP-

MENT OF FLOWER NUMBER PER HEAD

A most important conclusion which Ludwig reaches from his
studies is that the maxima for the number of ray-flowers and the
total number of flowers in the Compositae and the number of rays

340 Semi-centennial of Torrey Botanical Club

per umbel in the Umbelliferae all follow the series of Fibonacci
(i» 2, 3, 5, 8, 13, 21, etc.), the maxima differing for different species,
or for different races of the same species. He sought to relate
deviations from this series to the " Nebenzahlen " of the series,
i- 39> 42, 55, 63, which may be considered as "duplica" or
"triplica" of certain of the main series ('97a).

However, in discussing various views regarding phyllotaxy,
Ludwig ( '97b) suggests the development of other series than the
Fibonacci (1/2, 1/3, 2/5, 3/8, 5/13, etc.). The Trientalis, for
example, differs in giving the series 1/3, 1/4, 2/7, 3/11, 5/18, etc.

Evolution in regard to flower number was thus held by Ludwig
to be discontinuous, so that the various species in a phylogeny
represent a series of discontinuous variations with values for
flower number which depend on that of the original species.

The number of flowers realized in ontogeny was considered to
be determined first by the divisions initiated in the mother organ
("Mutterorgan"), and, second, by the processes that determine
phyllotaxy. The suggestion is made that the development of a
flower head or of the number of rays involves one complete turn
of the spiral. Ludwig, however, does not attempt to correlate
flower number with the phyllotaxy of the species, and in the Com-
positae he does not find maxima that correspond to any other than
the Fibonacci series or duplica or triplica of its various numbers.

Ludwig's later ('95) theoretical conceptions of the morpho-
genetic processes involved in the development of the different
numbers of ray-flowers are based chiefly on the observations of
Otto Miiller ('83) on Melosira. In this diatom the individual
cells remain attached, forming filaments. The development of
the filamentous colony Miiller claims to be as follows. Cell
division, as always in diatoms, occurs in such a way that of the
two daughter cells one is larger. The larger then divides while
the smaller one rests. Then the latter divides simultaneously
with the larger of the newer pair. Thus one cell divides, giving
two; of these, one divides making three cells in the filament; two
of these next divide, making five in all ; three of these divide next,
making eight in all, etc. It is thus claimed that there are rhythmic
and periodic divisions in which all the older cells divide together
with one half of the newer cells. As a result, the number of cells

Stout & Boas: Statistical studies in Cichorium 341

in the filament increases according to the Fibonacci series, i, 3, 5,
8, 13, 21, etc.

On the basis of these observations, Ludwig explained the occur-
rence of maxima for ray-flowers, which he considered to correspond
to the series of Fibonacci, on the hypothesis that in the develop-
ment of such organs as ray-flowers, one part is like the mother
organ and another is like the offspring. The mother organ forms
new parts in rhythmic succession, the offspring goes through a
ripening period and then divides. In Ludwig's own words, "Das
Mutterorgan grenzt fortgesetzt in rhythmischer Wiederholung
neue Telle ab, der Sprossteil dagegen immer erst in der folgenden
Teilungsperiode, nachdem derselbe herangewachsen ist." (Quo-
tation from Vogler, '12.) The term "mother organ" is used
vaguely. Whether it is considered as a fertilized egg or as the
apical growing-point in the main stem or in lateral stems is not
stated. What significance partial variability may have is not
discussed.

Ludwig later ('04) brings this conception forward in support
of the mutation theory which had then recently been announced
by de Vries. The emphasis was laid on the discontinuous increase
in the number of cells and organs involved in such rhythmic
divisions. In the Compositae, for example, the various species
are considered to represent different steps in a serie^?, the number
of cell divisions are assumed to stop at certain points and become
hereditary. However, in referring to de Vries's studies, later to
be discussed here, Ludwig ('04) considers that it is possible to
change by cultivation the stage or step which has been reached by
a species.

As a theory of morphogenesis, the conception is interesting
and suggestive. It is not indicated, however, how a series of
rhythmic linear divisions, such as may occur discontinuously in
filamentous forms, is to be applied to complex growing-points
involving various histogens where various groups of cells are con-
cerned in the production of an organ, as in the case of the formation
of a composite flower head, and especially in the application to
the number of differentiated ray-flowers constituting only a part
of the head. Furthermore, the correlation of characteristic num-
bers for a species with the phyllotaxy is not attempted.

342 Semi-centennial of Torrey Botanical Club

Weisse ('97) attempts to refer the position and number of
flowers in the head to the mechanism of phyllotaxy, as conceived
by Schwendener ('78, '85). According to this view, the arrange-
ment of lateral organs, as leaves, is determined by the pressure
they exert upon one another when in the embryonic condition.
The arrangement and relative position of matured organs is hence
not definitely related to divisions taking place in the growing
point, but is dependent on the mutual pressures between organs
already present and above which the new ones are being formed.
Weisse 's observations were made on 141 main flower heads of
Helianthus annuus. The most definite facts revealed are that
as the ray-flower number increases more disk-flowers are laid down
before two are in contact and that there is much variation in the
phyllotaxy immediately below the flower head. There seems to be
an increase in ray-flower number with higher values of phyllotaxy,
but the great range of variability (13 to 82) in ray-flower number
and the small number of cases do not permit of a very definite
conclusion.

III. SPECIAL CRITICISMS OF FACTS AND THEORIES

Vogler ('12) discusses especially the theories held by Ludwig.
He points out that Ludwig ('87) at first related development of
flowers in a head to processes concerned with the mechanism of
leaf position, but that as early as 1888 he (Ludwig, '88) refers the
phenomenon to certain types of rhythmic cell division which were
later more fully formulated by him (Ludwig, '95 and '98). Vogler
points out that the two processes (i) the mechanism of phyl-
lotaxy and (2) periodic divisions are not necessarily exclusive, and
that leaf arrangement may be due to the same kind of periodic
cell-divisions as are involved in the production of flowers in a head.

Furthermore, Vogler ( '12) appears to contend that the develop-
ment of certain numbers as maxima rather than others does not
involve the reproduction of Anlagen according to the scheme of
Fibonacci, but results from the continuation of the phyllotaxy in
the arrangement of the number of parts of the flower head. " Diese
Bevorzugung bestimmter Zahlen ist nicht die Folge einer Ver-
mehrung der Anlagen nach dem Schema des Fibonacci, sondern
ergiebt sich aus dem gesetzmassigem Anschluss an die Spiral-

Stout & Boas: Statistical studies in Cichorium 343

stellung der Blatter." He is evidently led to this view chiefly
by his observation that partial variability is strongly in evidence.
Also in summarizing the statistical work done in the Compositae,
he ('ii) finds that only about 85 per cent, of the species exhibit
maxima that fall on main and duplica numbers of the Fibonacci
series.

Vogler's ('08, '09, '10, '11) original studies on Umbelliferae
and Compositae bear directly on the question of maxima and
their occurrence according to the law of Fibonacci. His study of
the umbellifer Astrantia major L. ('08) is of special interest, for
here he finds the maxima for the number of bracts, perfect flowers,
and male flowers to be lower for lateral umbels than for those on
the main stem. In all cases the variability was less for the parts
of the lateral umbels than for the main umbels. His conclusion
is that the maxima for the parts of the umbels of the main stem
follow in general the series of Fibonacci, while those of the lateral
stems follow another and quite dififerent series, the Trientalis.
Vogler also reports data on the number of ray-flowers in 31,000
heads of Senecio alpinus ('09, '10) collected at different localities
and at different times of the year. For these the number ranges
from 10 to 28 with a maximum at 19, which, as he points out, is
not one of the primary or secondary numbers of the Fibonacci
series.

In the case of two plants of Boltonia latisquama, Vogler ('09,
'10) studied the production of flowers in successive years. For
three successive years 500 flower heads were counted on each.
Although the weather conditions were different in the three years
each plant was quite constant in respect to ray-flower number
during the three years of observation.

Furthermore, Vogler ('10) undertook in Arnica montana and
Eupatorium album to determine various facts of partial variability,
especially the relation of position of heads to variation in the
number of ray-flowers. In counts made on ray-flowers of Arnica
montana in 1909 and 1910, he separated data for terminals and
laterals. In both years the terminals gave a higher average ray-
flower number than the laterals. In 1909, 266 terminals averaged
14.7, while 153 laterals averaged only 11. 6. In 1910 (counts were
made in a different locality from that of 1909), 314 terminals

344 Semi-centennial of Torrey Botanical Club

averaged 15.1, 149 laterals only 11. i. In both years the maximum
for terminals was 13, a number of the Fibonacci series, that for
laterals 11, a number of the Trien talis series. More intensive
studies were made of partial variability in one plant of Eupatorium
album. This species has only tubular flowers and the counts were
hence rgade of the total flower number. The plant studied had
six branches. On each branch Vogler counted the terminal heads,
and then the laterals in succession downward. The most notice-
able points brought out by his data are that the flower number is
different for the various main branches and that on the same
branch the number is different for terminals and laterals, especially
for branches near the top of the plant.

A survey of Vogler's work shows that the maxima do not always
accord with the series of Fibonacci ; they may fall on other numbers
such as certain of the Trientalis series. He has also observed and
emphasized the significance of partial variability, which he views
as evidence against the conception that rhythmic divisions give
harmony between mother and daughter organs as to position and
number.

IV. EVIDENCE OF INTRASEASONAL VARIABILITY

Some interesting observations regarding intraseasonal varia-
bility in number of flowers per head are reported by MacLeod
('99). He found that the ray- and disk-flowers of Centaurea
CyanuSj C. alba, and C. atropurpurea, whether from terminal or
from lateral heads, are more numerous per head early in the season
than later. However, as he studied heads indiscriminately, there
are no data on the individual behavior of plants. He grew some
of his plants on poor and some on rich soil and found that those
on poor soil had a lower flower number per head than those on
rich. He rightly points out that experimental breeding work of
this kind is necessary in order to determine whether differences
are due to heredity or merely to food supply and season.

Tower ('02) reported differences in flower number for heads
of Chrysanthemum Leucanthemum collected in the same location,
but at different dates (July 5 and 30) of the same season. When
grouped, such data gave a bimodal curve involving what he calls
"secular modes."

Stout & Boas: Statistical studies in Cichorium 345

Further analysis of intraseasonal variation has been made by
Shull. He ('02) made successive collections of heads of Aster
prenanthoides and counted the bracts and the ray- and disk-flowers
per head. The heads were taken from plants growing wild on a
"single small plot" and were evidently collected indiscriminately.
On the first date of collection records were made from 117 heads,
on the second from 143, on the third from 139, and on the last
from 116. The average number for all the organs counted de-
creased in number as the season advanced. Later ('04) he gives
data for twelve successive collections, made from September 12
to October 9 in 1903; the curves show low values at first, then a
sudden rise, which is followed by a gradual decrease and a sudden
rise at the end of the season. The rise at the end of the season
can hardly be considered to be significant, since it was determined
from only four heads. Shull attributes seasonal variability here
observed chiefly to individual variability rather than to partial
variability. He suggests that the low flower number seen early
in the season is due to weak or starved individuals which bloom
first and which have a lower flower number than is normal for the
species, but his data being indiscriminate do not determine the
facts regarding this point. Further, his view raises the question
of the adequacy of judging vigor by the number of flowers per
head. The data later presented in this paper for chicory will show
that partial variability may have been involved as the principal
cause of the seasonal decrease observed by Shull.

V. OBSERVATIONS AND EXPERIMENTS ON THE INFLUENCE OF

ENVIRONMENT

With the recognition that the number of parts in an inflores-
cence is often subject to considerable variability, certain experi-
mental studies were made regarding the influence of nutrition.
Some of these have a very direct bearing on the factors involved
in both individual and partial variability.

Weisse ( '97) sowed seeds of two heads of Helianthus anniius: half
of the seeds of each was sown in pots containing sand, and half
was sown in ordinary garden soil. The ray-flowers in the terminal
heads were counted. The 155 heads from plants grown in sand
had an average ray-flower number of 21 and a standard deviation

346 Semi-centennial of Torrey Botanical Club

of ±6.6. The 221 heads from plants grown in the garden had
an average ray-flower number of 37 and a standard deviation of
d=9.2. These differences observed were quite evidently due to
the effect of nutrition. A point that is of interest and that Weisse
does not call attention to is the smaller variability of the starved
plants. Weisse states further that certain plants that were heavily
manured had as many as 82 ray-flowers in a head and a maximum
at 55 ; while heads with less than 30 ray-flowers usually were found
on plants which were checked and retarded in growth by the
crowding of neighboring plants. Weisse combines the data ob-
tained from the two cultures, the one on sand and the other
on garden soil, and points out that the resulting bimodal curve,
having a maximum at 21 and one at 34, is a result of different
nutritional conditions under which the population was grown.
He believes such conditions might readily occur in nature and
that bimodal curves that have been ascribed to the mixture of
two races are often due to differences in nutrition occurring in
nature.

MacLeod ('99) published some observations on Centaurea
atropurpurea which show the effect of nutrition on the number of
ray- and disk-flowers. Here again there is a lower flower number
under conditions of poor nutrition, but the number of disk-
flowers seems to be more affected than that of ray-flowers. Mac-
Leod gives only the averages for each culture, so that the vari-
abilities of the cultures cannot be determined.

De Vries ( '01) repeatedly states that favorable environmental
conditions, such as optimum water-supply and manuring, tend to
increase the size of organs (fruits of Oenothera) and the number of
parts, as the number of rays in the umbels of Umbelliferae or the
ray-flowers in the heads of Compositae,

Danforth ('08) comparing ray-flowers of the daisy growing
in a well-drained situation with those in a drier situation, reports
a lower mean and less variability for the latter. Koriba ('08)
made successive collections of heads of Arnica unalaschensis
(also some other Composites) from two different localities, one
from a valley and the other from the slope of a mountain. Those
on the slopes were growing under the least favorable conditions and
gave uniformly lower values.

Stout & Boas: Statistical studies in Cichorium 347

Detailed studies by Burkill ('95) show that the number of
stamens in flowers of Stellaria media increases during the first two
weeks of bloom, then there is a gradual decrease to the end of the
flowering season. Reinohl ('03) working with the same plant
showed further that the plants grown in poor soil produced flowers
with a lower number of stamens. Tammes's ('05) studies of the
effect of good and poor soil conditions on various characters show
lower values for many of the characters of poorly nourished plants.
Love's researches ('11) are among the most recent that deal with
the effect of nutrition on the mean of such characters as height,
number of peas per pod, weight of seeds, etc., in such plants as
peas, buckwheat, and corn. His results agree with those men-
tioned above, that is, he finds an increase in the mean and in the
variability, as a result of better nutrition.

VI. STUDIES OF INTERANNUAL VARIABILITY

In the consideration of int^rannual variability, at least two
distinct aspects are to be recognized. First, the season of growth
may differ in a way that affects the plant, and, second, in the case
of perennial plants the age of the plant may be a factor in vari-
ability just as the period of development may influence the partial
variability seen in a single year of growth. These two factors
have rarely been distinguished and little attention has been directed
to the factor of age, for flower nurnber studies have been chiefly
indiscriminate for the population.

Haacke ('96), it seems, is the only investigator who has
attributed variation in the number of ray-flowers of composites
to the age of the plant. He suggests that older plants probably
have more ray-flowers per head. His observations on Chrysan-
themum Leucanthemum and Anthemis arvensis were not conclusive,
however, for he did not know the age of the plants studied.

Yule ( '02) gives the results of counts made in three successive
years of the number of sepals in a population of Anemone nemorosa
from one habitat. He gets differences in the means for different
years and calls attention to the importance of observing ''local
races" for several years before one can determine the character-
istics of a species. Shull ('04) made a similar study for a popu-
lation of Aster prenanthoides and obtained higher mean values for

348 Semi-centennial of Torrey Botanical Club

bracts, rays, and disk-flowers for 1900 than for 1903. He attributes
this to more favorable dimatic conditions in 1900 and points out the
importance of taking differences due to climate into account in
determining "place constants."

Clark ('10), working with timothy, calculated coefficients of
correlation for the same character in different years. Following
Tower he calls these coefficients of '* place variation." He corre-
lates height in 1905 with that in 1906, height in 1906 with that in
1907, and height in 1905 with that in 1907. He does the same
with weight. He finds considerable correlation, which means
that in spite of varying conditions, such as climate, plants high in
values one year will be high the next.

Harris ('15) has recently called attention to the different
values of interannual correlation for different characters. This
study gives a comparison of data of the same sort for successive
years by means of correlation tables, as Clark ('10) has done in
the paper just mentioned. The degree of relation is then expressed
by the coefficient of correlation. This method has been used
especially in studies on growth in man (Boas and Wissler '05),
and Harris refers to work of Pearl and Surface on egg-production
and Gavin on milk records, where interannual coefficients of cor-
relation were used. Harris published data for interannual cor-
relations for fruits of Staphylea trifolia and Hibiscus syriacus in
which he gets various degrees of correlation according to the parts
of the fruit studied; for example, for 23 fruits of Hibiscus he finds
the correlation for 1907 and 1908 to be for sepals and sepals
+0.46, for bracts and bracts +0.84, for ovules and ovules +0.94,
for seeds and seeds +0.63, etc. While these studies indicate that
the degree of correlation may be different for different characters,
they are not especially concerned with the analysis of changes in
values for a single character due to such a factor as age.

VII. EVIDENCE THAT POSITION|IS A FACTOR IN PARTIAL

VARIABILITY

With the recognition of the existence of partial variability
there developed further refinement of study which aimed to de-
termine the relation of position on a plant to difference in number
of parts. Burkill ('95) gives indiscriminate data for 102 flowers

Stout & Boas: Statistical studies in Cichorium 349

of Caltha palustris which indicate that the number of stamens and
carpels is larger in terminal flowers than in laterals.

Haacke ('96) made a detailed study of ray-flower number of
Tanacetum corymbosum, taking into account the position of the
heads on the plant. He studied 81 plants and presents data for
each plant separately. The largest number of heads for a single
plant recorded is 14. There is one flower head at the end of the
main stem, which he calls primary. There are, on the average, four
or five unbranched branches each bearing a terminal head. These
he calls secondary heads. There usually follow several branched
branches, which bear secondary and several tertiary flower heads.
The lower branches of the plant are longest, thickest, most
branched and the secondary heads on these branches are about
the same distance from the ground as the primary head. In other
words, the lower branches have the greatest similarity to the plant
as a whole. He found that the primary head had on the average
the highest number of ray-flowers, the secondary head of the first
branch the lowest. There was then an increase in the number of
ray-flowers of successive secondary heads. From the tenth branch
downwards the number of ray-flowers of the secondary heads was
the same as that of the primary head. There was a correlation
between ray-flower number in the primary head and of the secon-
dary and tertiary heads; those plants having high or low numbers
in the primary had correspondingly , high or low numbers in the
others.

MacLeod ('99) made a similar study of the flower heads of
Centaurea atropurpurea. He does not, however, keep the data
for individual plants apart and his results are not as clear as those
of Haacke. He first counted the terminal heads of the main
axes, 424 in all. The average total number of flowers (disk and
ray) in these was 47.7. He then examined the heads on the
branches. Of these the first group consisted of 524 heads in
bloom between July 10 and 12. These had an average flower
number of 39.2, considerably lower than the average for terminal
heads. He calls this a "bud -generation" (knopgeneratien) .
After seven days he cut off all the open flower heads, and his
second "bud-generation" consisted of 656 heads blooming from
July 21-25, with an average flower number of 34.4 floA^ers per

350 Semi-centennial of Torrey Botanical Club

head. He continued in this way to the end of the flowering season
and got a decrease in flower number in successive "bud-genera-
tions." The only fact revealed that clearly bears on the question
of flower number in relation to position is that there appears to be
a higher flower number for terminal heads than for those borne
on side-branches.

Schiiepp ('13) has made detailed statistical studies on
Aconitum Napellus. One of his chapters is devoted to variations
within the individual (partial variability) and his data show that
to a certain extent quantitative characters are functions of the
position of the organ on the plant. This is very apparent for a
character like leaves, which in the vegetative parts are petioled,
large, and have 40-50 points, while in the reproductive regions
they are sessile, small, and one-pointed. He also gives the number
of perianth parts, nectaries, stamens, and carpels for three regions,
base, middle, and top of plant, and in all finds a slight decrease in
the number of parts from the base upward.

Klebs ( '06) showed that there were slight differences between
lateral and terminal inflorescences in Sempervivum. Vogler
('12), whose work has been discussed earlier, presents data to show
differences according to position between number of flowers in the
inflorescences of Umbelliferae and Compositae.

Such differences as have been noted have a bearing on the
much larger question of the periodicity shown in the development
of an individual plant. Braun, Sachs, van Tieghem, J. W. von
Mohl, and de Vries have contributed much to this question.
Tammes ('03) reviews the literature on this subject and gives to
von Mohl the credit of establishing the fact that there is a period-
icity in cell division, so that the longer internodes have more as
well as longer cells than the shorter. Tammes ('03) investigated
a large number of plants and showed that there is a periodicity
in development for length and breadth of leaves, length of petioles,
and number of main veins.

These studies show that partial variability in respect to the
number of parts in a complex structure such as a flower or a
flower head is to some degree related to position on a plant in-
volving time of development, and therefore introduces an element
of differentiation. This places an emphasis on processes of devel-

Stout & Boas: Statistical studies in Cichorium 351

opment which give a periodicity or a sort of polarity. The proc-
esses assigned by such conceptions as that of Ludwig to rhythmic
cell divisions which give specific differences for species as such may
themselves undergo change, continuous or discontinuous, in the
development of successive parts of a single individual.

VIII, SPECIAL VIEWS REGARDING HEREDITY, DIFFERENTIA-
TION, AND SYMMETRY, HELD BY PEARSON AND BY BATESON

While Pearson's studies of numerical qualities in plants do not
pertain to number of flowers in any of the composites, they are of
special interest in the recognition that differentiation is a factor
in partial variability seen among organs of the same kind. They
also illustrate very well the difffculties of adequately determining
the heredity of such a character as the number of stigmatic bands
or seed chambers in fruits of poppies, of Nigella hispanica, and of
Malva rotundifolia.

Pearson's earlier report ('oi) bears on the statistical and
mathematical demonstration that "undifferentiated like organs"
or homotypes " on an individual are alike only to a certain degree.
The degree of likeness between homotypes as measured by his
methods of determining homotypic correlation, has on the average
a mean value of 0.4-0.5, which is, he considers, quite identical
w^ith the general value for fraternal correlation. Thus he con-
cludes that heredity is a phase of homotyposis and that the
sources of variability are to be sought in the individual. The
distinction between differentiated and undifferentiated like organs
is not, Pearson recognizes, always easy to make. In general, the
former class involves function, position on the individual, season
of production, etc., and is statistically discoverable by testing the
frequency distribution for heterogeneity. In contrast to this,
Pearson distinguishes variability of "undifferentiated like organs"
as due to "that combination of small causes, inherent and en-
vironmental, which leads to what is familiar in both theory and
observation as a homogeneous chance distribution" (p. 287).

We may note that when such differentiations as exist in the
poppy and in Nigella are thus treated as pure chance variations
the statistical treatment may give a high or low value for homo-
typosis. The existence of differentiation is not necessarily re-

352 Semi-centennial of Torrey Botanical Club

vealed by such treatment of data. It is to be determined only by
observation and by a refinement of methods of collecting data.

Pearson's first studies ('02b) pertaining to the heredity of the
number of stigmatic bands in capsules of the Shirley poppy are of
special interest, for here data were collected from all capsules.
These data were statistically treated by three methods :

1. The correlation of all offspring capsules with parental mean
capsule, the various progenies grown in each locality being thrown
together in a single correlation table.

2. The comparison of the average variability of an array of
offspring of a single parent plant with the variability of the off-
spring population. Here the means for individual offspring were
determined.

3. A mathematical consideration of homotypic relationship
in correcting the parental correlation determined by the first
method.

According to the first method the parental correlations for the
different crops, as a whole, range from 0.3230 to 0.1220. The
highest correlation of 0.3230 was obtained in the "most starve-
ling crop" which had few capsules per plant and the low correla-
tion of 0.1220 was obtained in the crop that was most luxuriant
in growth. Here is definite evidence that the greater vigor of
growth affects individual variability by increasing very much the
partial variability. On this account the method of collecting data
and the statistical treatment give lower parental correlation when
there is increased vigor.

In one crop of 907 plants the means were determined separately
for each plant and these were correlated with the mean of the
parents. The value was 0.1561 as compared with 0.1864 obtained
for the same crop by the first method. Here Pearson attributes
the-low parental correlation to "differentiation" and reports that
the flowers that come out "early in the season have fewer bands
than those which come later" and that "the number of capsules to
the individual plant, and the dates at which it produces them, tend
to obscure the influence of pure heredity, and make the stigmata,
however easy to count and deal with a by no means ideal char-
acter to study heredity upon" (p. 72).

The low values obtained for parental correlation were, however,

Stout & Boas: Statistical studies in Cichorium 353

not considered as correct because homotyposis was involved.
The true parental correlation, according to Pearson's conception,
was higher. By accounting for homotyposis the value was raised
from the average of 0.20 to a value lying between 0.35 and 0.40.

It would seem that much of the difficulty here experienced in
attempts to make exact determination of values, even for popu-
lations such as Pearson studied, lies in treatment of all the varia-
tions as "chance." Although Pearson definitely recognizes that
lateral flowers are differentiated from terminals, there is no attempt
to determine values for such partial variability.

In further studying heredity of number per capsule in the
poppy, Pearson ('06) sought to avoid the difficulties previously
encountered in estimating the individual when multiple observa-
tions involving partial variability were made. He attempted to
do this by "confining the attention to the first or principal flower."
In 1903 and 1904, crops were grown from seed of random samples
in 15 different localities and treated as populations. Differences
in mean and in variability were found which were attributed to
effects of environment as affecting individual variability and
which were so great as to be "not directly comparable." It was
possible to determine parental correlations for these results in
only one population; a crop grown in 1904 from parents of a 1903
crop, the two crops, however, were grown at different localities.
The raw correlation was only 0.17 17.

Pearson therefore concludes that the determination of heredity
even for such an easily measured quality as the number of stig-
matic bands in pods of the poppy is exceedingly complex and
difficult, and he now questions "whether the apical flower is as
true a measure of individuality as the totality of flowers on the
plant" ('06, p. 400).

Pearson is here concerned with population studies and in
intensity of parental correlation for rather mixed populations.
His treatment and results suggest and in fact reveal many sources
of variability. His rather uncertain results raise very definitely
the question of how to value adequately a numerical character
which exhibits elements of both chance and differential variability
for both partial and individual variations.

Bateson ('01, '03) questions the validity of Pearson's distinc-

354 Semi-centennial of Torrey Botanical Club

tion between "chance variation" and "differentiation ' in the
treatment of homotyposis and heredity. He insists that "meris-
tic" variations are discontinuous and doubts that "there is a true
material distinction between variation and differentiation as
appHed to parts of the same organism." Bateson further objects
to a comparison of "undifferentiated Hke organs" with the cor-
relation between brothers which rriay be differentiated as individ-
uals. He evidently views the partial meristic differences of
organs of the same kind actually in evidence in such plants as
Nigella, Cichorium, etc., as a differentiation of the same rank as
differentiation between individuals as such ('03, p. 23). Bateson
emphasizes the aspects of symmetry, advocates an extreme view
that tissues and organs arise somatically by "differentiant or
segregating divisions" in much the same sense as Weismann pos-
tulated, and he thus questions the adequacy of the term "chance
variations." He is perhaps strongly influenced by his earlier
studies of meristic variations in animals in which differentiation
and symmetry are in marked evidence, and by the views of segre-
gation of hereditar}^ units representing characters which may give
differentiation between individuals of the same hybrid origin.

IX. EVIDENCE OF HEREDITARY VARIATIONS

There appears to be no report of researches directed to the
study of selection and heredity involving only total flower number
per head in any of the Compositae. There are, of course, many
species in cultivation from which double-flowered varieties have
been developed, the history of which does not involve statistical
studies of total flower number. Moreover, the development of
so-called double-flowered composites does not necessarily involve
increase or decrease of total flowers per head, but a change of such
flowers as tubular disk-flowers into strap-shaped or ligulate
flowers more like the ray-flowers.

The studies of de Vries ('01) on Chrysanthemum segetum are
of interest in their bearing on selection, heredity, and evolution of
flower number. He first isolated a race having a maximum of 13
ray-flowers in the terminal heads. Then he isolated a race with
21 ray-flowers in the terminal head. In this case, however, he
considered it necessary to judge his plants not only by the ray-

Stout & Boas: Statistical studies in Cichorium 355

flower number of the terminal heads, but also by the ray-flower
number of the later heads, for he found that some plants having
21 ray-flowers in the main head gave for all flower heads curves
mth maxima lower than 21, often at 13 or 14. These plants were
discarded as not belonging to the desired race and only those
giving "partial curves" (the curves obtained from the flower
heads on<a plant) with maxima at 21 were retained. No attempts
were made to isolate races with numbers intermediate between
13 and 21, and no further studies were made of the very irregular
cases of partial variability which were in evidence.

After isolating the two races, de Vries observed variation in
the race with 21 ray-flowers in terminal heads in respect to in-
crease of ray-flowers. In a crop of 1,500 plants one- was found
with each of four lateral heads having 22 ray-flowers, a higher
number by one than was previously seen in any of the terminal
heads. Open-fertilized seed of this plant gave a progeny in 1897
of 414 plants; the number of ray-flowers in terminal heads ranged
from 14 to 34. From seed of the plant having 34 ray-flowers in
1898, 241 plants were grown ; for these the range was 19 to 48; the
modes were at 26 and 34. The average of the population was 38.
The next year 194 plants, offspring of the plant with 48 ray-
flowers, produced terminals with ray-flowers ranging from 19 to
67; the modes were scarcely pronounced at 32, 37, and 45, and the
average number of rays was 41.5. In this crop, it is stated, ligulate
flowers appeared among the disk-flowers in one head of 62 ray-
flowers. The seed of this plant gave a progeny in 1900, 31 plants
in all, ranging in ligulate flowers of terminal heads from 33 to lOi ;
there were no pronounced modes either primary or secondary;
the average number was 53.2. One plant had some heads with
only ligulate flowers.

Evidently for the first few years there was an increase in the
size of the heads and the accompanying number of ray-flowers.
Then there came also a change of disk-flowers to ray-flowers.
Throughout there had been great irregularity in heredity revealed
in range of numbers observed in individual variability both for
data of terminals alone or for terminals and laterals. Such varia-
tion de Vries assigns to mutation and attempts to show that the
increase in number follows the Fibonacci series. There can be

356 Semi-centennial of Torrey Botanical Club

no doubt that spontaneous variations here occurred, giving in-
creased variabiUty, and that selection of the extremes was effective
in giving a new race.

X. PREVIOUS STUDY OF FLOWER NUMBER IN CICHORIUM

INTYBUS

It appears that the only statistical study of flower number in
a species in which the flowers are all ligulate is that of de Helguero

('06).

His data were obtained from Cichorium Intyhus. He counted
the flowers in 1,000 heads produced by 624 individuals; the counts
were made on five different dates during August when the plants
were approaching the end of the blooming period. "Le piante
furono raccolte in 5 diverse volte durante il mese di Agosto e
percio nel periodo decrescente della fioritura." Of the total 1,000
heads, 389 were from 389 plants which had only one head each in
bloom at the time of collection: 300 heads were from 150 plants
having two heads each in bloom when the counts were made: 311
heads were from 85 plants having 3 heads open on a single date.
The data are treated almost solely to determine homotyposis.
The table of correlation for the 150 pairs (two heads for each plant
paired) gives a correlation of +0.5915. The correlation table
for all pairing of two or more per plant gave a correlation of
+ 0.6130. The results show that when a few heads (in most cases
only two) are taken from a plant on a single date at the end of the
blooming period the correlation is about +0.6. De Helguero did
not extend his observations sufficiently to determine adequate
values for partial and individual variabilities.

XI. SUMMARY

Statistical studies of such a numerical character as number of
parts in an inflorescence were initiated about twenty five years ago
by the very general collection of data from mixed populations.
The method was to study species en masse. The chief aim was
to discover specific qualities. Broad generalizations were made
(i) that flower number is specific for species, (2) that the numbers
characteristic of species fall in a series such as that of Fibonacci,
and (3) that evolution giving such differences has been discon-

Stout & Boas: Statistical studies in Cichorium 357

tinuous. The further important conclusion was reached that such
discontinuous specific differentiation arises through processes
operating in the development of the individual and these processes
were assumed to involve rhythmic cell divisions. It was also
suggested that there is a relation between flower number and
phyllotaxy.

It was early recognized, however, that the maxima for species
do not all fall in a w^ell-defined series and hence attempts were
made to show that in such cases other series were represented.
From the first, it was evident that there was often a wide range of
variation in flower number in a species, but the view was taken
that the variations within a species or a race were solely due to
chance, and that the facts could be accurately determined by
Galtonian treatment of populations.

The more recent work has been very generally directed to the
study of variation within a species. Processes operating within
the population have received attention. Studies have become
more particular and individual in scope. Various hitherto unre-
vealed sources of variability, intraseasonal, interseasonal, environ-
mental, racial, individual, and partial were thus demonstrated.

To the present time, however, these studies have been largely
dominated by the view that the variations are those of chance.
The demonstration in a few cases that such variations, especially
partial, are not purely due to chance, but may proceed in a dis-
coverable manner, has revealed a source of possible error in much
of the work done and emphasizes the desirability of combining
extensive studies with a study of the organization of individuals
as units.

It seems clear that intensive studies of individual and partial
variabilities should serve as a basis for extensive study of species
as such. Through such methods we may hope more adequately
to determine the facts which serve as a basis of judgment regarding
the processes operating in ontogeny, phylogeny, and evolution.

THE PROBLEMS IN CICHORIUM INTYBUS

The statistical studies here reported for Cichorium Intybus
were begun with the aim of determining the facts as to partial and
individual variability for such a character as the total number of

358 Semi-centennial of Torrey Botanical Club

flowers produced per head. The nature of the partial variabiUties
was found to be such as to afford special opportunities for analyses
of the intraseasonal and interannual variability and for a study of
variation among heads according to position on a plant. The
data were so collected that individual variabilities could also be
determined. As the studies progressed interest extended to a
study of heredity and the effects of selection in the different races
which appeared and which were grown in pedigreed cultures.

MATERIAL AND METHODS

Cichorium Intyhus is in many respects especially favorable for
such study. The flowers are conspicuous and with the exception
of an occasional tubular flower are all ligulate. The flowers of a
head are readily distinguished and easily counted, as all the
flowers of a head are fully expanded at the same time. A head
opens but once and is usually expanded but a few hours during
the forenoon, a behavior that somewhat limits the amount of data
that can be collected in a day, but makes the collection of data from
day to day more simple, as there is no danger of recounting the
same heads. The flower number per head is not excessively high,
which with the disposition of expanded flowers makes accurate
counts a simple matter. A considerable number of flower heads
open each day during a rather extended period of blooming. The
numbers during the greater part of the season are sufficient to give
at least ten heads per day for study. The abundant branching
and the development of a large number of heads in various posi-
tions and at various times admits of rather full development of
numerous parts of an individual and gives opportunity for the
intensive study of various aspects of partial variability. The
plants are, furthermore, hardy and easy to cultivate, so that
numerous plants can be grown under as nearly the same conditions
as is possible.

The cultures of Cichorium Intyhus studied were, for the most
part, the same plants whose behavior in respect to sterility (due
to physiological incompatibility) has already been reported (Stout,
'i6, '17); they include mainly two somewhat distinct groups of
plants, as follows:

In one group are Fi, F2, F3, and F4 generations derived by

Stout & Boas: Statistical studies in Cichorium 359

crossing plants of wild white-flowered chicory and plants of the
unimproved cultivated variety known as Barbe de Capucin.
Some study was made of a few plants of the parent stock. The
wild white-floweied parents (designated as A and C) were ob-
tained in the autumn of 1911 from the campus of the University
of Wisconsin and transplanted to the experimental plots at the
New York Botanical Garden. In 1912 a crop of plants was grown
from the open-fertilized seeds of these two plants. Plants of the
variety Barbe de Capucin were grown in 191 2 from seed obtained
from J. M. Thorburn and Company (no. 4300, catalogue of 1911).
These plants are designated the E series. All the plants of the
first crop, of both wild and cultivated strains which were experi-
mentally tested for self-fertiiity, were found to be self-sterile from
physiological incompatibility. Crosses were made involving two
plants of wild stock (A and C) and two plants of Barbe de Capucin
(£3 and £22). Some of the Fi progeny were self-fertile (Stout,
1916), and from these several self-fertilized lines of descent have
now been grown in F2, Fs, and F4 generations and utilized in
the statistical studies. The greater number of races and lines
of descent reported later are descended from a single cross between
a wild white-flowered plant (A) and one of the variety Barbe de
Capucin (£22).

Considerable data were also collected from plants of the inbred
generations of a salad variety known as improved red-leaved
Treviso, the seed of which was produced by the firm of Ernst
Benary of Erfurt, Germany, and supplied by J. M. Thorburn and
Company. One generation of 50 plants, F] hybrids between
plants of red-leaved Treviso and the wild white-flowered plant A
have also been studied. It should be stated that the seeds for all
the cultures of chicory here reported have been sown in sterilized
soil in small pans during the months of December and January,
the young seedlings were transplanted to pots all properly labeled
and grown in the greenhouse until the first of April, when they were
transferred to cold frames. By the middle of May, when planted
in the field, the plants as a rule have formed vigorous rosettes often
more than a foot in diameter. With this 'treatment nearly all
plants come into bloom in the first year of growth.

The plants of the Treviso strain, as well as such varieties as

360 Semi-centennial of Torrey Botanical Club

improved large-leaved, improved white-leaved, and improved
spotted, when treated this way come into heavy bloom and die as
annuals. The plants of the Treviso strain have therefore been
studied in only one year of bloom.

The preliminary studies of 191 2 indicated that usually there is
such a marked decrease in flower number per head as the season
advances that data which are to be considered adequate should
be taken during the entire flowering season. Hence, in the col-
lection of data here reported, the method has been to begin on
the first day of blooming and to continue throughout the season,
obtaining counts of ten heads per plant, when that number was
in bloom, on each of at least ten different dates per month at
intervals of every third day, or as close to that schedule as con-
ditions allowed. The flower heads counted were taken at random
from various branches on the plant and from various parts of the
different branches. The heads selected were usually removed,
the flowers counted, and the number recorded on sheets specially
ruled and tabulated for dates and numbers. In 1913 and 1914,
many data were also taken on the character of each flower as to
whether ligulate or tubular, as to number of teeth in the ligule and
the depth of lobing. In 1915 and 1916 data were secured for
flower number only. Special methods were used to obtain certain
data; for all plants studied in 1915 data were taken every day
during the first twenty days of blooming; and on five plants data
were taken on all flower heads (excepting those that opened on
Sundays or on days of heavy rainfall) with reference to their
position on the various branches of the plant; this was done to
determine more exactly the relation of position to time of blooming
and to the intraseasonal change in number, both of which appear
in the data as more generally collected. Data have been collected
from a few individual plants during four successive years of growth,
but the greater number of plants have been studied in the first of
bloom.

Considering all plants of all stock, in 191 3 data were collected
from 63 plants, in 1914 from no, in 1915 from 351, and in 1916
from 450. The total number of individual plants involved is 832.
The number of heads counted approximates 5,500 in 1913, 20,000
in 1914, 113,000 in 1915 (including here data collected every day

Stout & Boas: Statistical studies in Cichorium 361

for first 20 days of bloom), and 95,000 in 191 6, making a grand
total of about 233,500. The total number of individual flowers
counted during the four years is about 4,200,000.

The collection of these data has taken much time and has
involved the cooperation of several persons, as follows: In 1913,
Dr. Joseph C. Oilman and Mr. Allen C. Fraser assisted the senior
author of this paper in the collection of data. In 1914, Miss
Friedolina C. Jud and Mr. Allen C. Fraser assisted the joint
authors of this paper. In 1915, Miss Jud and Mr. R. C. Faul-
wetter assisted the writers. In 191 6, Mr. M. V. Reed assisted
during the months of July and August. In 1914, 1915, and 1916,
Mr. Charles Holste, who was gardener in charge of the experimen-
tal plots, frequently assisted, especially when the work was press-
ing. Dr. Oilman in 1913, Mr. Fraser in 1914, Miss Jud and Mr.
Faulwetter in 1915, and Mr. Reed in 1916 were recipients of
scholarship grants from the Oarden. In July, 1914, Miss Boas,
joint author in the production of this paper, became a member of
the Oarden staff and took charge of the compilation and special
methods of statistical treatment of the data. Miss Boas has
made all the computations involved in this report and has assisted
very materially in the search of literature. For the greater por-
tion of the text, for critical discussions of literature, for discussion
of results, and for opinions expressed in this paper the senior
author is responsible.

Special care has been taken to insure uniform methods in the
collection of data and the persons assisting collected from the
same plants continuously at least during the period of their co-
operation.

The somewhat laborious work of collecting such a large amount
of data has been possible only through the hearty cooperation of
the persons above mentioned. The authors wish here to express
their appreciation of this cooperation and of the support of Dr.
N. L. Britton, Director-in-Chief, in granting the scholarships as
noted.

362 Semi-centennial of Torrey Botanical Club

PRESENTATION OF DATA FOR FLOWER NUMBER

IN CHICORY

I. GENERAL SURVEY OF THE KINDS OF VARIABILITY PRESENT

I. Partial variability
A. IntrQseasonal partial variability.

The collection of random data on successive dates from individ-
ual plants reveals that, as a rule, there is a marked decrease in
flower number per head as the period of flowering advances.
Differences in number per head appear according to the stage of
development of the plant as a whole.

The seasonal performance of a plant, as shown in such tables
as I and 2, indicates that the number of flowers per head in heads
as they appear from day to day is much higher during the early
period of bloom than in the last days of bloom. This is indicated
by both the daily range and the daily average. The change,
however, is rather uniform and progressive as the season ad-
vances.

Partial variability, or variation among the apparently homol-
ogous heads produced by a single plant, is therefore seen in the
range of the number per head; in table 2 the range is from 12 to
23. But the daily data show that the range of values and the
average value shifts from day to day. The variations from day to
day are not therefore solely chance variations, since certain
elements of differentiation appear.

The totals of all heads with the same number of flowers give
what appears as a chance distribution. The chance collection of
data for any period of timxC, as the first ten days, the second ten
days, etc., would also give general summaries that would appear
as chance variations; that such data do not adequately represent
the individual seasonal performance is, however, very obvious.
Random collections would hardly reveal the presence of intra -
seasonal variation; it is only the tabulation and computation of
data for individual days from individual plants that clearly
brings out such facts.

For the sake of completeness, there are given with the tables
values whose significance will become clear later on. It is suffi-
cient to point out here that a is the computed flower number for

M ro ro

ro . ro

M N M ro M

(N M

O

00

<N
1

ro

il

II

II

2

M ro eg

364 Semi-centennial of Torrey Botanical Club

I iz£

Octobei

M U-) N M

M • U-) M

• \0 N

0'6i

N

M

M • U-)

September

0

ro 01 •

9-ii

00

M . -rf )-l IT)

8-81

M M M

2-81

N

• M M •

9-81

N
N

M M M M M

0-81

'mm •

6-81

- !

ro H r^t M

■ 1 6-li

1

(N • <N

8-81

ro M ro (M

I-8I

00

• U-> CO M

9-61

e-6i

August

M M M rO M LO

t7-8i

■ N Tj- ro

i-6i

cq M M

i-Ai

-1

M M -sT CNl M M

t"8i

ON

M H (N Oq

&61

&61

ro oq

1 S-6i

N

&61

0

ro fO ^

I'OZ

NO

V'oz

1 6-6i

CM -rf M

(N j 9*02

July

0 1

MOM

&61

ro <N

8'6i

g-oz

o 1

L'oz

00 1

M LO M

O'OZ

1

$-oz

! ! ; !

o-iz

0 1

! ! ^ ! M

V-iz

3q=

•AV

Stout & Boas: Statistical studies in Cichorium

8

<; o

hi

H

<

Pi
W

0^ <;

November

O'll

October

Csj M (N

L-Li

?!

M ; M n

o-gi

o-

M ; CS) M

Lgi

M to M M

e-Ai

rO

M 00 M

9-91

ON

; fO 01 M

rZ,i

M ro ro M ; M

6-91

September

OS

ro fO <N ^

£-Ai

<»

to ^ M

9-ii

9*Z,i

in

8-91

M fO <N

L-Li

M (N tJ- ro

1 6-li

o-Li

O

ro csj CO

Q-Li

fO

o-li

M (N sO M

L-gi

m

■ N O I-I H

i-li

M (N fO •

6-gi

• CN N M

i7-Ai

CNl ^ •

2-91

O

0-91

OS

M M CSJ M

g-Li

00

p-li

M M M

6-gi

so

ro

6-gi

0-81

Number
of flowers
per head

H M M H-l M M

1

Averages . .

II II

On

d

+

0)

II

II

[0

366 Semi-centennial of Torrey Botanical Club

the first day of blooming, b the calculated rate of decrease per day,
and / the number of days the plant was in flower, [o] represents
the average flower number of the plant, calculated as are all the
other values, from data collected throughout the flowering period.

Exceptions to the general rule of seasonal decrease occur.
One of these is indicated in the performance record given in table
3. The marked feature here is the decided uniformity in both
range and average number of flowers per head throughout the
entire period of bloom.

The plant here considered (table 3) bloomed for a period of
two months. The total number of heads counted was 248.
The partial variability seen in the range of 15 to 20 was quite
indiscriminate and coexistent as to time. As the season advanced
there was no appreciable change and the computed value for the
rate of decrease is +0.004, ^ slight increase.

Data of another plant showing no intraseasonal decrease are
presented in table 4. Here there is, as in table 3, no noticeable
change in the general average or range from day to day, except
for a few irregularities at the beginning and end of bloom in which
the values usually obtained are reversed; the very lowest numbers
per head appearing on the first two days of bloom.

In a few cases there is a decided though small increase from
the beginning to the end of the season. The greatest increase
observed was +0.025. In all these cases the plants had a low
flower number throughout the season. Table 31 gives data for
one of these plants.

A further variation in seasonal performance is shown by data
of table 15. Here there is a marked decrease both in the range
and in the average of flowers per head, but the lower numbers of
the range remain quite the same throughout; the shifting involves
chiefly the higher numbers.

B. Inter annual partial variability.

The collection of data from the same plant in successive years
makes possible a comparison of the performance of an individual
in different years in which the seasonal and growth conditions
may be different, and in which the different ages of the plant may
also give differences in vigor involving differences in time of
blooming, all influencing the performance of the season.

00 N

N

. M nO . M

N
N

NO

lO

ON

*^

CM

tn

1 ; ; M CM (N ; ; !

NO

NO

ON

NO

00

NO

! ! ; M cNj IT) ; ;

NO

On

CN) CN) ro 1— *

nO

— —

•

0
CO

• • • . ^ ^ ^ . .

nO

On

(N

. . . NO ro M . . .

(»

. ^ .

nO

O

^.

. «^ NO (N

o

• ■ *^ . w . . .

oo

. w M . >^ .

nO

'"I

ro

. <^ .

nO

On

C-i
N

. I-* IN •rl- ro .

nO

00

0

.

H?

On

00

. M M nO CN<

nO

On

o

— -
o

. <^ .

NO

; ; ; lo fo M ; ; ;

CNj

m

H

NO

o

; ; M M CN) ; M

^ '. i-i '.

CM

On

M ; M ; N CM ; ; ;

00

368 Semi-centennial of Torrey Botanical Club

Tables 5, 6, and 7 present data for the same wild white-
flowered plant (A) for three successive years (1913, 1914, and
1915). The age of this plant when transplanted in 1911 from
Wisconsin to the experimental garden was unknown. There has
been no very marked difference in its general vigor and habit of
growth in the five years that it has been under observation, except
that in 1914 part of the roots which were inadvertently somewhat
exposed were killed by winter freezing and there were fewer main
branches produced from the cluster of roots.

The performance of this plant (^4) in each of the three succes-
sive years shows a seasonal decrease that is quite characteristic of
the species. There is also rather close agreement in the ranges of
partial variabiHty, these being 21-13, 22-15, 22-12. The average
number per head and the standard deviation are also quite uniform
as follows: 1913, 17.2dzi.79; 1914, 17.8zb1.41; 1915, 17.3zb1.72.
There is also rather close agreement in the values for the first date
of blooming (a), but there was a considerable increase in the length
of the flowering period, (t) in 191 5 over that of the previous years.
The rate of decrease (b) was lowest in 191 5. The significance of
these facts and the means of proper comparison of such data will
be discussed presently.

The principal interannual partial variability is seen in the
length of the blooming period. When this is considerably shorter,
as in the year 1914, and the total amont of decrease remains much
the same, the rate of decrease (b) is necessarily more marked.

The wild white-flowered plant considered above was grown
from roots obtained in the field and its growth and vigor were
much more uniform in successive years than is the growth of
plants grown from seed. The latter, as a rule, exhibit in the second
year of growth a marked increase in general vigor as measured by
the number and size of the main stems,which gives a corresponding
extension of the flowering period with the production of more
flowers.

These aspects of interannual partial variability may be illus-
trated by TABLES 8, 9, and 10, which present the data collected
from a plant in the first, second, and third year of growth.

The records of this plant in the first, second, and third
years of growth agree in giving lower numbers per head as the

r

fs) ro to fo to to

w rj- rj- Tt M

to

mber

! ! ; ; fo tH ; ; ;

Septf

i i lO Tj- M ] 1 1 1

so

to

m

1 ; ; ; ; 1

o
to

CO

; M lo M ro ! ; ; ;

sQ

to

VO

»^ ! '-I ! '. '.

to
4

ro

! ro fo ; ; ;

° j ! ! o CO N ; : : :

sO

to

A.ugusi

00

M lo ro M ; M

lO

; w N ; ;

0\
to

„

N i-i ro M M M ;

■a

IN
SO

M M M Tt M ; ;

ST
sd

'. '. ! M ^ ro H-i ;

ro
t-^

; ! ! 0) uo M ; ;

CS)

<»

19.0

SO

! ; H M M M ;

(N

; M (N (N (N "xi- ;

to

^

; ; ; (M M IT) M M ;

00

>>

IT)

'. '. ! M ro ;

00

<—>

C4

; ; ; ; m lo m

to
06

0

'. '. ! M M ro ; ;

00

'. '. '. ; M o to

6\

! '. M M to ; ;

00

; ; ; M lo <N ;

00

'. '. ! M M !>. ;

ro

ro

d

o
c

!!!!!! M ;

ro

d

3

M o :

d

Number
of flowers
per head

too t^OO o o

M

o\
10
q

d

d
1

sO

11

II

N

0

(N

!>•

06

<N
1

CO

M

00

11

2

Semi-centennial of Torrey Botanical Club

M n N H

j> to ro

II II II II

S ^ S 2l

M M O M

Stout & Boas: Statistical studies in Cichorium

Fiox I ■ . ^^g^^^^ " I i6l

NOY.

! M ! M

October

CO

M M M N

0-Si

cf 1

rn ■

?!

(N M

e-si

00

ro

M

0*8 1.

o-Qi

M M

S-gi

LSI

September

" 1

M (N

L-gi

^ i

o-gi

S-gi

00
00

o-gi

M ro M M

£-Li

M

o-Si

" 1

0'17I

August

ro
t-^

<N IN

£-gi

P)

£-li

0

OS

o-gi

ro • M

i-ti

ro

M roo •

S-li

O

• M rt ro M

17-81

M • CM M ro

^•81

; M M f-i M ;

l-Li

July

; ro ; o M

^•21

: w ^

] M

g-6i

N

; M (M M

i-6i

0

MM ; 00 ;

S-8I

; ro CM n N M

9*81

M c>) ; fo N CM

r6i

N

; M M fO H

o-6i

00

. . .

O'OZ

fO

o

!>

00

ro
1

to

ro

II

II

2

372 Semi-centennial of Torrey Botanical Club

TABLE 8

Seasonal distribution of number of flowers per head for first year of
BLOOM OF an Fi generation PLANT (EsXA) no. 7 derived by crossing a

PLANT of BaRBE DE CaPUCIN (£3) AND THE WILD PLANT {A). DaTA FOR I913

Number of flowers
per head

August

Total

6

8

II

19

21

26

28

30

16
17
18
19
20
21

I

2
3
I

I

5
29
38
8
2

2
3
5

I

6
3

2
6
2

3
6
I

2
4
3
I

3

6
2

5
5

2
2

I

Averages

19.0

18.8

19-3

18.9

18.3

18.7

18.5

18.2

17.6

83

[o] = 18.6 a = 19. 1

\oi\ = - 3-17 b = - 0.045

[/] =12 t = 24

[tn = 69.89

season advanced. The range of intraseasonal partial variability
was slightly increased in successive years, but there are very
marked differences in the length of the blooming period, in the
total number of flowers produced, and in the rate of decrease.
The length of blooming period in the first year of bloom for this
plant was rather below that of the average one-year-old plant.
The average number of flowers per head for 1913 is 18.6, for 1914
it is 19.0, and for 1915 it is 18.7. The values of the first date of
bloom are 19. i, 20.9, and 19.9. Aside from the rate of decrease
the various values are not widely different. In this respect the
plant in question is one of the most uniform that we have studied,
exhibiting, perhaps, the least interannual variability with respect
to values [0] and a.

The interannual partial variability was usually more pro-
nounced than in the plant noted above. Frequently the annual
performance was quite divergent in nearly all respects. One of
the most marked of such cases is seen in the plant for which data
are given in tables ii, 12, and 13.

In the first year of growth (table ii), this plant exhibited a
rather short period of bloom, the number of flower heads produced
was low; the range was only 20-17. the second year (table
12), the period of bloom was much extended, the total production

M M CM ro M c>q

M M M <N

O 00

2 w

O h!

< 5

M (N ro M

ITi M CS) (M

On

O

d

d

o

1

II

II

o

o

M

ro

1

11

II

o

rc ro ro M

M M M M

6 o

374 Semi-centennial of Torrey Botanical Club

0.

2 i

0 K
Pi ^

Q W

01 >

i ^'

November

" 1 :

1

O'OZ

00

M CM

1

z-6i

^ 1 ■

N M .

1

8-91

M M

October

°0 1

01

M CS M M

8-91

M M

1

S-6i

1

<N M

1 9-81

1

e-8i

CM

CM CM

'. CM

1 ^-81

M fO M t-l ; M

1 s-ii

September

?? 1

: M ro O

1 s-si

c? 1

ly-> M M CM M

1 £-Li

m 1

M M ro ro

1 2-^1

; (N M M U-5 M

1

M M CM ] <N

1 9-91

M M ro [mm

g-Li

00 1 H

M M (N fO M <N

£-U

CM rO fO HH M

9-Li

M re CM CM CM

I-8I

August

M -tt CM CO ;

A-Ai

n"

M CM ;

6I1

; M M ro

o-6i

0

; CM ; M M

zsi

0\

o-gi

M PI ro

o-6i

cn

CN ro "mm

8-81

o

M (N M ro ro

S-6i

\o

MM ; to M M

6-81

o

; M U-)

t^-6i

July

<o

; M ro CM H

6-6i

CO

; CM H

f-6i

M M C^ M (N

o-oz

t'-Si

o\

; M ro CM ;

6-6i

IT)

; M ; M M M M f

9.02

: : fo fo ; : ;

0-03

0\

; M ; CM ; o M !

£-iz

; ; M ro CM M

o£z

061

00

M

q

6 M
I M

CM vO

Stout & Boas: Statistical studies in Cichorium 375

TABLE II

Seasonal distribution of number of flowers per head for the first year
of bloom of an fi generation plant (c x en) no. i derived by crossing

A WILD PLANT (C) AND A PLANT OF BaRBE DE CaPUCIN (£22). DaTA FOR I913

Number of flowers

August

Septemb

er

5

per head

7

8

13

14

16

19

25

27

30

13

23

29

0
H

17
18

2

I

2

I

6

I

2

I

I

2

3

I

I

12

19

I

4

3

2

4

5

8

4

5

7

7

50

20

3

2

I

4

3

I

2

2

4

22

Averages

19.4

19-3

18.8

19.4

19-3

18.5

19.2

18.7

19.4

18.5

18.7

18.0

90

[o] = 18.9 a = 19.2

[ot] = — 5.00 b = — 0.016

=19 ^ = 53
[t'] = 285.75

of heads greatly increased, the range of flowers per head was 22-13,
and there was a rather uniform and gradual rate of decrease. For
the third year of bloom (table 13), the period of bloom and range
of partial variability were quite as in the previous year, but there
were irregularities in the decrease in that the lowest numbers per
head were reached in the latter part of September followed by a
rather decided increase in number per head. For the successive
years the numbers for average flowers per head [0] are 18.9, 18.3,
and 17.9; the values for the first day of bloom (a) are 19.2, 21.8,
and 18.5, and the values for rate of decrease are —0.016, —0.068,
and —0.028. Both the average number per head [0] and the value
for the first day of bloom (a) are decidedly lower in the third year.

The interannual variation in regard to the intraseasonal
variabilities was marked in the case of a plant for which data are
presented in tables 14 and 15, and which have already been
referred to as unusual types of interseasonal performance. Here
the behavior in the first and second years of growth involve
marked differences in the decrease and range. In the first year
the daily range and average remained quite the same, in the second
year there was a decided decrease in average, but the decrease in
range was seen only in the higher number. The values of the
number for first day of blooming, 15.3 and 1 7.1, are widely differ-
ent, as are the rates of decrease +0.001 and —0.054. However,
the averages for flower number per head are quite identical, 15.3
and 15.8.

Semi-centennial of Torrey Botanical Club

0*l7I

1 0-171

o

1 S-Si

en

17-Sl

Oct(

On

0-91

e-9i

(N

1-

h-

1 V-Si

ro

H

ro t-

o-Z,i

1

o

N

1-

cq cv

VO

1 8-91

Li LI

?!

fo fo •

L-Li

mbet

^ 1-

a

V

c/)

e-8i

H

1-

0

ro

VO H

£■81

CO

A-8I

6-Li

■ (N CM M

i-6i

On

r^- ro

o-os

to N H

0-61

\o

Tt lo ;

i-6i

N

00 N ;

z-6i

V)

3

b£

00 1-

e-6i

3
<

m

lo (N ro

8"6i

00

8-02

VO

: Ti-vo

9-02

S'oz

g'oz

0

00

S-oz

VO
PI

<N CO to

£-oz

'3
•—1

rn
N

; VO ^

z-iz

g

Tt lO

P'lZ

'. N ;

o-iz

C O

Stout & Boas: Statistical studies in Cichorium

1^

October

N

W M

8-91

ro M •

z-gi

00

P) M M

z-Li

S-Si

M M M

t-Si

l-gi

M lO M

o-Si

September

M (N U-) M M

&Si

M fO

o-Si

g-fi

00

M M N

z-\'i

M ro -O • (N M

z'Si

MM ■ CO '^t M

i-gi

00

M

C"Si

M ■ ro IT) M

S-Qi

; ro M

z'Li

„

M •Tt ro

&gi

August

CO

M M M M

O'll

N M P) M

g-gi

; ro M

8"9i

M M ro M

z-gi

(M r:f M M

M M

o-gi

ro

M ro M ro

M

i-gi

0

; o fo M

S-Li

1/1

M rq

i-Li

: 00 M M

£-Li

July

ro

M 00 M M

z'Li

M'Tt Tt M

S'Li

M ro ^ M

L-Li

ro

N

ro IT, (St ;

6-gi

01 ro

£'Li

: : o

t-gi

tr,

M <N O M

L-Li

ro

L-Li

On

s-si

M o ro

M

o-6i

June

'. r^

21.0

ro

ro O O
ro in <N

II II

378

Semi-centennial of Torrey Botanical Club

< te

Oh <

H

2 Q

H

<! .

M 2

« O
W erf
M u

M Os M 00 C^J 1

0\

CO
CM

November

00

M M M

0'17T

i

0-gi

-1

CO •

0-Si

October |

M CM

l-Si

^ :

9"t'i

H

M ^ H

P-Si

o\ M :

rSi

M IT) fO M

^•Si

M ^ N

rSi

in ro

M

l-Si

00

o-Si

M <N U-) CM

8-171

\o

M ID M

o-Si

s-si

M 00 l-i

September

• o

l»

M vO !N • M

t->

N

• M M O • H

■n
w

W M

c?

(N N M

• C^

M ro fO

s-si

o

CM CO <N W

00

CM

• CM

o-Si

CO CO CO

m

• • 00 CM

• M so CM

rSi

Number of
flowers per
head

o 00

Averages .

CO O
CM CO
CM

II II II

Stout & Boas: Statistical studies in Cichorium 379

TABLE 15

Seasonal distribution of number of flowers per head for the second year

OF bloom (1916) FOR THE PLANT (£22 X A)-IO-nO. 17. DaTA FOR FIRST YEAR
given in TABLE I4

Number

July

August

■«

0

of flowers
per head

5

M 7

10

13

15

18

21

24

27

29

I

4

7

II

15

18

25

28

14
15
16

I

I

I

I

I

I

I

7

2
I

5
2

4
I

I

I
2

4
3
3

5
3
2

I
5
3
I

6
2

4
4
2

6
3

5
2

6
I

3
3

4
I

6

I

58
37
36
8

17
18

3
I

4
I

6

4
2

2

2

2

I

3

19
20

I

2

3

I

I

2

21

I

I

Averages

IT)

M

0

Cs
10

0

00
10

N
10

10

10
10

4

0
4

10
4

152

[o] = 15.8 a = 17. 1

[o/] = - 13.90 b = - 0.054

['1 =24 t = S4
M = 261. II

2. Individual variability

The various types of intraseasonal variability noted above
are in reality records of the performance of individual plants
during a seasonal period of growth. The differences noted between
them as individuals are, therefore, evidences of individual varia-
bility. Individuals may differ decidedly in respect to one or
more values, total production of heads, average and range of
number of flowers per head, or amount and rate of decrease.
These differences may appear among plants of the same age.

The interannual variability and especially that seen in the
first two years of growth indicates clearly that much individual
variability may exist in plants of different ages.

The inter-relations of individual variability and partial vari-
ability are here very evident. Any estimate of the former involves
the latter in one or in all its aspects. It seems, therefore, clear
that the question of an adequate judgment of the individual must
be based on full performance at least during one season of growth,
and that the comparison of individuals should be made for plants
of the same age.

380 Semi-centennial of Torrey Botanical Club

II. STATISTICAL TREATMENT OF DATA

The various sources of partial variability considered above
make clear some essentials in regard to the proper collection and
utilization of data in order that they may reveal the nature of the
character of flower number and its hereditary behavior. If the
flower number per head fluctuated with much the same range from
day to day, as it does for a few plants of chicory, the mean, the
standard deviation, and the coefficient of variability would be
quite sufficient to give an estimate of the individual and could be
determined from rather random readings with the magnitude of
error depending chiefly on the numbei of heads that were counted.
But this is not the case with the majority of plants. The flower
heads mature at diffcent dates and the number per head, as a
rule, decreases as the season advances. It will be shown later
this is to some degree related to the position of the head on a plant.

The partial variations from day to day are not solely fluctua-
ting. For most plants, as shown in tables i, 2, and 41 the daily
range of fluctuation changes in a somewhat uniform progression to
lower values. It seems to the writers that this element of change
should be recognized in any statistical treatment which attempts
adequately to determine values for a plant as a whole.

In this investigation, instead of calculating the mean and the
variability, as expressed either by the standard deviation or the
coefficient of variability, and using these as expressions to char-
acterize the flower number of an individual plant, the flower
number of the first day of bloom and the rate of change have been
calculated and used.

The latter expression, as will be seen more clearly presently,
is calculated as the mean and the variability would be from all
the data at hand. A starting point, a rate of change, and the
length of time through which this change takes place, gives an
index to the variability. Furthermore, one can find the average
flower number for any one particular day {On for if the flower
number is given for the first day of bloom and the rate of change
following (either positive or negative). We may assume, for the
present, that the rate of change is uniform and call it b, and we
shall indicate the flower number for the first day of bloom (which
is to be calculated) a.

Stout & Boas: Statistical studies in Cichorium 381

It is obvious from such typical data as have been presented in
the foregoing tables that the average flower number for any
particular day {on for tn, for example) involves the actual flower
number on the first day of bloom plus the amount of change,
either positive or negative, following that date. In respect to
flower number in chicory the change is usually a decrease, and
if this decrease be computed it can be expressed as a rate of
change for the season and designated as the value h. The
amount of change from the first day of bloom to any one particular
day (after the plant has been in bloom 4 days for example) can be
expressed by the number of days the plant has been in flower
multiplied by the rate of change, as Un. According to this con-
ception then On = a + and the following series of linear equa-
tions will express the series of averages obtained from the actual
observations day by day: oq = a -\- Uq . . . On = a -\- btn.

From the data the average of the flower number obtained
from day to day can readily be determined. This value for the
data collected for the plant £3 (see table i) arranged in the first
column of table 16 is 18.6.

The values of t are also known and can be expressed as devia-
tions from the average time of blooming cornputed from the known
dates of the collection of data. This average as given for plant
£3 in the second column of table 16 is 30.7. Since this average is
often a fraction the calculation may be simplified by measuring
the different times of the collection of data from the integer
nearest to the average so that /„ = ^ + when t is the integer
nearest the average and t„ is the deviation of tn from this integer.
If the difference between the true average and the integer used in
the calculation be considered as d then the full value for the devia-
tion from the true average is -f d.

The average of all the (r + dYs must, of course, equal zero,
because they express the plus and minus deviations from the true
average. The average of all the r's equals —d. The equations
then assume the form :

00 = a b {t -\- d -\r tq)

01 = a -\- b {t -\- d + Ti)

02 = a b (t -{- d + T2)

382 Semi-centennial of Torrey Botanical Club

On = a -j- b (t + d + Tn)

The average of this series is i [o] = a -\- b [t] ; because as shown
above [t] = — d and -\- d — d = o.

The value of a thus becomes equal to [o] — b [t] and of these
quantities [o] and [/] are determined from the data and are known.

For the purpose of calculating the value of b another equation
is needed and this may be obtained by multiplying each of the
above series of equations by t + d. Since, however, d is the same
value throughout we may ignore it and we have :

OoTo = {a + b (t -{- d) ]ro -\- br.^

OnTn = {a + b (t d) }r„ + &r„2

Averaging and substituting [r] = —d,

Solving this equation the value of b is obtained, and sub-
stituting the value of in i gives the value of a.

In table i6 are calculated the values of b and a for the plant
Es in 191 3 (see table I for values of 0) in order to illustrate the
method employed.

It will be more apparent in the presentation of data that fol-
lows that differences in the rate of decrease throughout the season
already noted in the general survey of intraseasonal partial
variability are undoubtedly the source of greatest individual
variability. The expression [0] = a -\- b [t] as developed above
can be used to express the general behavior of an individual in
respect to flower number, and from it are determined the values
of a and b to be used in comparisons.

The values of a and b therefore may be discussed further. In
the plant involved in table 16 the value of b is —0.065. This
is the value of the amount of decrease per day estimated from all
the observations. While the actual variation in the rate is more
or less variable in the data as collected, there is much in the be-
havior to suggest that until more of the factors contributing to
this are known and analyzed, it may be treated theoretically as a
uniform rate in each plant, an assumption which will admit of the
foregoing mathematical treatment."

2 [or] = - d{a+ bt) - bd'^ + b [r^]

3 [0] d --= dia + bt)

4 [or] +d[o] = b [r2] - bd''

Multiplying ihyd gives 3
Adding 2 and 3 gives 4

Stout & Boas: Statistical studies in Cichorium 383

TABLE i6

Calculation OF flower number of first day of bloom (a) and seasonal rate

OF decrease (b) FROM THE DATA FOR A PLANT OF BaRBE DE
CaPUCIN (£3) PRESENTED IN TABLE I

0

t

o(r + ./)

+ 3.4

0

- 31

- 1054

961

2.5

I

30

750

900

2.5

2

29

725

841

2.7

5

26

702

676

1.9

7

24

456

576

1.5

II

20

300

400

1.9

^3

18

342

324

0.7

IS

16

112

256

1-5

18

13

195

169

1.4

20

II

154

121

1.6

23

8

128

64

I.I

25

6

66

36

3-0

26

5

150

25

0.8

28

3

24

9

0.3

30

I

3

I

- 0.7

33

+

2

14

4

0.8

35

4

+ 32

16

0.4

37

6

24

36

1.6

40

9

144

81

I.O

44

13

130

169

I.O

47

16

160

256

1.0

50

19

190

361

1.4

55

24

336

576

0.8

60

29

232

849

1.4

64

33

462

1089

1-5

67

36

540

1296

1.4

74

AZ

602

1849

[0] 18.0 + 0.6

30.7

[r+rf] = 31 -0.3

\oi\ = - 29.49

W\ = 441-96

Substituting in formula derived above
(7) +0.6 = a + 316

(2) — 29.49 = + o.3(a + 31&) + 441-96& — (0.09&2) (last value so small that it

may be neglected)

(5) + 0.18 = + o.3(a + 316)
(4) - 29.67 + 441.966
.'. 6 = — 0.065

a = 18.0 + 2.6 = 20.6

The question will immediately arise whether the assumption
of a uniform rate of decrease is justified. If there were only
meager data at hand for each plant it would be necessary to use the
mean and the standard deviation as values. In such cases it would
be recognized that such values are not the most satisfactory ex-
pression of flower number, but that they are the best available
from the material at one's disposal. In chicory, however, there

384 Semi-centennial of Torrey Botanical Club

are sufficient data to enable the determination of a more adequate
expression for the behavior with respect to flower number; namely,
the theoretical value for the first day of bloom (a) and the rate of
decrease per day {b).

It is, however, evident that the rate of decrease is influenced
by a large number of factors such as length of blooming period,
position of heads on branches, kinds of heads, whether terminal
or lateral, etc., points which will be brought out later in the paper,
and that it is in consequence not uniform throughout the season.
The data, while sufficient to give us an approximate value for the
rate of decrease, are not sufficient to determine it absolutely, and
it must be borne in mind that the expression for flower number
used in this paper is not considered to be absolute, but merely the
most adequate to be obtained from the material at hand. This
value can be determined from all the data in the manner developed
above.

. The value of a can also readily be determined on the basis of
the observations. In the case of the data in table i6, a = 18.6 —
[( — 0.065) 31] = 20.6. This is a computed value for the flower
number of the first date of bloom.

It will be seen that two plants which start with the same flower
number per head and show the same rate of decrease must have a
different average number of flowers per head for the whole season
if their blooming periods are of different lengths. The average
will be lower the longer the blooming period. Just so the standard
deviation w'll be the greater the longer the blooming period. On
the other hand, if the two plants show differences in the rate of
decrease the average number of flowers per head for the whole
season may be the same in the two plants. The following com-
parison should make the point quite clear. The plant E3 (tables
I and 2) bloomed 74 days in 191 3 with an average flower number
for the season of 18.6 and a standard deviation dz 1.9. In 1914
it bloomed 91 days with an average flower number of 19.2 and
standard deviation of dr. 1.5. Does this mean that in 1914 £3
started with a higher number of flovvers per head than in 1913,
and maintained this characteristic throughout the season, or was
the starting point in the two years the same vvith a difference in the
manner of decrease? In using the method described above, for

Stout & Boas: Statistical studies in Cichorium 385

the plant £3 for the year 191 3, the number of flowers per head for
the first day of bloom was 20.6 with a decrease of —0.064 P^i day;
while for the year 1914 the average number of flowers for the
first day of bloom was 20.7 with a rate of decrease of —0.032.
The higher average in 19 14 was due to a slower rate of decrease.
To illustrate this point further we may take the plant (£3 X A)
no. 7 (tables 8 and 9) in 1913 and 1914. In 1913 the average
was 18.6 with a standard deviation of =t 0.9; in 1914, 19.0 with a
standard deviation of ± 1.7. In 1913 the blooming period was
24 days; in 1914, 70 days. The averages are the same as for the
plant £3, but the standard deviation is smaller for (£3 X A)
no. 7 in 1913. The values for the first day of bloom and the rate
of decrease show that the plants are different in respect to flower
number. In the plant (Es X A) no. y the difference between the
rate of change in 1913 and 1914 is not very great. In 1913 it was
— 0.045; in 1914, —0.053. The average flower number for the
first day of bloom is, however, quite different for the two years.
In 1913 it was 19. 1 ; in 1914, 20.9. This together with the different
length of blooming periods accounts for the differences in the
values of a obtained for the plants £3 and (£3 X A) no. 7.

III. DETAILED PRESENTATION OF DATA BEARING ON

I. Relation of length of blooming period to rate

OF CHANGE

One of the most obvious facts brought out by the data is that
in plants showing a seasonal decrease the rate of decrease is less
the longer the blooming period. The following correlation table
(table 17) brings out this fact very clearly. The plants used are
three-year-old Fi plants, of which there are data for no plants.

For a further study of the relation of rate of decrease to length
of flowering period these plants may be grouped as indicated.
As there were rather few plants with short blooming periods, it
was necessary to include in the first group all the plants that
bloomed less than 60 to 100 days. This group consisted of 16
plants. There were only three plants that bloomed less than 60
days, so these were left out entirely. In order to have a sufficient
number of plants in the fifth group those blooming from 130 to
150 days were here included.

386 Semi-centennial of Torrey Botanical Club

The table shows that there is a considerable negative corre-
lation between length of blooming period and rate of decrease.
The high values for the rate of decrease all fall in the first group,
that is the group containing the plants with the shortest blooming
periods. The fifth group, which is the group of plants with the
longest blooming period, has the lowest values for the rate of
decrease.

table 17

Correlation between length of blooming period and rate of decrease.
Data for iio Fi plants, 3 years old

Days of

Rate of decrease in

units (Xioo)

Total

bloom

O-I

1-2

2-3

3-4

4-5

5-6

6-7

7-8

8-9

9-10

10-11

40-50
50-60
60-70
70-80
80-90
90-100
lOO-IIO

I

I

I

I

2

I

I

I

I

I

3

I

I

2

I

I

6

}■

I

I

I

I

I

I

6

2

I

5
8

I

5
I

I

15
26

} 2
} 3
} 4

II0-I20

I

5
7
2

7

4

120-130
130-140
140-150

I

8

9
2

4

3

32
10
8

5
5

I

2

I

} ^

Total

3

19

27

27

13

12

3

2

I

2

I

IIO

r = — 0.67

The average flower number was calculated in successive twenty-
day periods for each group. Table 18 gives the average flower
number for these twenty-day periods. The five groups are very
much alike, excepting that there is, perhaps, a somewhat greater
decrease between the first and second twenty days in the plants

Diagram accompanying TABLE 18 showing curve for average decrease for

SUCCESSIVE twenty-day INTERVALS OF BLOOM,

Stout & Boas: Statistical studies in Cichorium 387

Z o

I I I

ro O O
O ro

odd

+ I

00 O sO O On O

On t-- w H mm
6 6 ^^ 6 6 6

I I I ++ I

PO t> >J0 O M

d d\ 00 t-^ oc

■o 00 o U-) j>
ro i> O ro 00 \0

ID O 00 lO 0\ 00
^ m' M

00 ro 00 ro O

6 1^ 6 6 6

I I I + I

:> Os O 00 M
0\ 00 «>• sO

00 O On On
O <N O ro
O M O 00 ro

On 00 N !

d hI

6 6 6

Q «

1 1

1 + 1 :

On O On O 00 sO
On On t-^ i>- r-^

Q ^

6 «

3 -°

lo ro 00 On
On M On ^

ro O On
M M d d
111 +

ro q q M 00

d On 00

r- 00 o
lo O

q ro

T T

d d\ 00

o -

w ?! .fa 3 5 ^
•fa ^ O -! .>< ^

c/5 H c/: c/2

is

5 V

O 00 U-5

r-o Tt M M W W

d d d d d d

I ++ I I I

00 O vO 0^ On O
N ro M On ro N O
Tl" O O 00 M

H M <N lo On lo

\0 iJO ro M M M

d d d d d d
I ++ I I.I

M M N 0\ O lO

O Tj- 0( M On 00 NO

4l 5 5 5

vO 00 vo IT)
ro i> O ro 00 NO
NO 00 On 00

M ■r}- 0^ 00 M

Tj- ro w O CN

6 6 6 6 6

I ++ I I

5 5 5 4I

00 O J> On On

O •rj- cs O Tf- ro

O >H vO (N 00 ro

ro oj m' M

000
I + +

0 o

1 +

ro ro !>• O \0 O

j>. sO On M o M

-< M M c>5 oi

4^ 4i 4^ -H 4^ -H

to ro 00 On
On 10 M On

M O NO N

0 00 O ro
N ro M ro
6 6 6 6

1 ++ I

M M (N t-H

-H -fl -H 4^ -H

00 o
10 t-- \0

M M vO

O M

d d
I +

On up Ov

5

>.

03 -

o

N

•fa. ^5

to c/} H c/: c/$

388 Semi-centennial of Torrey Botanical Club

with short blooming periods. In the plants with the longest
blooming periods, the greatest decrease comes between 40 and 60
days, while in the others it comes between 20 and 40 days. In
the last column are given the average differences for the successive
twenty-day periods. It will be seen that the greatest difference
is between the first and second period, so that here the rate of
decrease must be the greatest. Between the third and fourth
period there is an increase in flower number rather than a decrease,
and after that there is only a slight decrease, if any. The diagram
of table 18 illustrates this point. The differences between the
averages of the different periods have been plotted and it will be
seen that there is a marked change in the rate of decrease after
the first eighty days. Plants that bloom eighty days and less
will show a greater rate of decrease than those that bloom longer
for the reason that there is a slight increase or no further decrease
in flower number after the first eighty days. This accounts for
the large negative correlation between rate of decrease and length
of blooming period. This discussion also illustrates the fact that
the range of variability in respect to high and low flower number
per head is not very closely related to length of bloom.

The variabilities have also been calculated for the successive
twenty-day periods (table 18). The as show a decrease, then
an increase, and following this they again decrease. Since all the
factors influencing flower number are not known, one cannot define
the different variabilities exhibited by the different periods. We
know, however, that there are sources of variability during the
first period which do not exist later on, as, for instance, the coming
into bloom of the different branches. The decrease in variability
towards the end of the blooming season follows necessarily from
the fact that there is a limit to the lowest number of flowers per
head produced by a plant. The high numbers which appeared
early in the season do not as a rule appear late in the season; the
low numbers, however, appear early and do not decrease farther
as the season advances.

The range of variability for three-year-old Fi plants and one-
year-old F3 plants for successive twenty-day periods is given in
table 19. These plants are reported in detail in tables 18 and
20.

Stout & Boas: Statistical studies in Cichorium 389

TABLE 19

Range of variability for successive 20-day periods of bloom. Data for Fi

3-YEAR-OLD PLANTS (REPORTED IN TABLE 1 8). DATA FOR Fs I-YEAR-OLD PLANTS
(reported in TABLE 20)

Range of variability

Difference

Number of plants

|- First 20 days

30-14

16

107

Second "

27-10

17

107

Third " "

25-12

13

107

Fr

Fourth "

25-10

15

91

Fifth " "

23-10

13

76

Sixth " "

24-11

13

50

Seventh "

22-12

10

18

. First 20 days

33-10

23

110

Second "

24- 9

15

97

Third " "

23-11

12

78

Fourth "

23-10

13

39

^ Fifth " "

20-12

8

12

This table shows that the lowest flower number on a plant
may appear in the first or second twenty days of the blooming
period while the highest number appears only in the beginning of
the season and decreases as the season advances. This must, of
course, result in a smaller variability for the end of the season and
also explains why there is such a marked negative correlation
between rate of decrease and length of the blooming period.

The period of bloom for plants in the first year of growth is
shorter than in the succeeding years. There is, however, a
similar negative correlation between rate of decrease and length
of blooming period, as in the older plants discussed above. For
the behavior of plants in the first year of bloom \here are data on
no of the F3 plants, all blooming in the same year (1915). The
plants, for convenience, have been grouped into five groups, each
group blooming twenty days longer than the preceding. As
shown in table 20 the same sort of differences appear in successive
twenty-day periods as in the older plants, that is, the differences
are less or there may even be an increase in the later periods.

The variabilities for the F3 plants are gWen in the same way
as for the Fi plants, and here too the as are the greatest for the
first twenty days.

From the foregoing discussion, it is clear that there is less dif-
ference between the averages and that the variabilities are lower
for twenty-day periods after the first eighty days for the three-

390

Semi-centennial of Torrey Botanical Club

z o

Q ^

is c

O J> U-5 o

On Oi <N <N

6 6 6 6

111 +

O 0\ o o

O vO 00 O

00 o NO fc

M H d 6

00 vd lO

(N O ro to

M (N M O

O O

d M d
I I

4)T3

•2 <^

00 M M M

!>• vd vd

o o

\0 tM

\0 00

d d

ro r~ q\

to ^

•a

o . -

T3 x:
to O) H to

Average
Differ-
ence

— o.i6

— 0.12
+0.10

-o.i8

Number
of heads

O 0\ O O M

<3 vO 00 vO 00

oo_ 0_ \0 ro

Diflfer-
ence

0\ Os O 00

W M M M

6 6 6 6
1 1 1 1

b

M CO O

0 to o

1 ^1 4l 45

Number
of heads

Ov M ro O .
CN O to .
M M vo .

+ (N M ;

Differ-
ence

C\ 00 o ■

0 M M ;

6 6 6 ■

1 1 + ;

b

ro ^ M M

00 o ;

flflflfl :

Number
of heads

vo i> Os
(N "sf

Differ-
ence

ro U-)

M O

d d
+ 1

b

00 O

h1 5 5

Number
of heads

to oq
't 1.

Differ-
ence

oo
d
1

b

00 M

q i>-

-H -W

— j

Number
of heads

Os
ro

Differ-
ence

b

M

-H

First 20 days

Second " "

Third " "

Fifth " " 1

Stout & Boas: Statistical studies in Cichorium 391

year-old Fi's and after the first sixty days for the one-year-old
Fs's. In other words, most of the decrease has taken place during
eighty days in the first case and during sixty in the second. It is
quite obvious then, when purely mathematically considered, that
there should be a large negative correlation between rate of de-
crease and length of blooming period.

2. Significance of the range of variability in flower
number per head
A question which arises in this connection, and which appears
to be of considerable biological significance, is whether the total
amount of decrease or the actual range in variation in flower
number throughout the season bears any relation to the highest
flower number in the plant. In other words, do plants with high
flower number show a larger total decrease than those with low,
or is the amount of decrease about the same in each? The ques-
tion was suggested by the comparison given above of the data
for different periods of bloom, when it became evident that the
variability during the later periods was lower than that of the
earlier.

The following tables, one for three-year-old Fi plants and one
for one-year-old F3 plants (the same used above) will perhaps best

TABLE 21

Range of variability of flower number per head of io6 three- year-old

PLANTS of the Fi GENERATION

Number of

Maximum flower

Minimum flower number

Diff"erence between

plants

number

Range

Average

maximum and

average minimum

30

15-0

15.0

2

29

13-12

12.5

16.5

3

28

17-14

15.3

12.7

5

27

15-12

14.2

12.8

14

26

16-12

14.0

12.0

8

25

17-13

14.2

II. 8

17

24

1 5-1 1

13-8

10.2

19

23

16-10

12.6

10.4

21

22

17-11

12.9

9.1

13

21

16-11

13-2

7.8

3

20

13-11

12.0

8.0

answer the question. Here the total range from highest to lowest
flower number has been recorded, the plants with the same highest

392 Semi-centennial of Torrey Botanical Club

number are put in one group. The highest numbers for the io6
plants of TABLE 21 range from 30 to 20 and the lowest numbers
from 16 to 10; for the plants of table 22 the highest numbers
range from 33 to 17 and the lowest from 17 to 9. For the plants
as a whole there is therefore a greater range for the highest number
per head than for the lowest number.

table 22

Range of variability of flower number per head of 189 one-year-old

PLANTS of the F3 GENERATION

Number of

Maximum flower

Minimum flower number

Difference between

plants

number

Range

Average

maximum and
average minimum

I

33

13-0

20.0

I

27

13.0

14.0

5

26

15-13

14.4

II.6

3

25

15-14

14.7

10.3

9

24

15-13

13-8

10.2

25

23

16-10

13-6

9.4

41

22

17-10

13-9

8.1

48

21

1 7-1 1

13-5

7.5

36

20

15- 9

12.9

7-1

II

19

13-11

12.0

7.0

8

18

14-10

11.9

6.1

I

17

II. 0

6.0

The plants with highest numbers, however, show the greatest
range of variability as their lowest numbers are as low as the
majority of the plants with lowest values of the highest number.
A glance at the tables 21 and 22 shows this point. Of three-
year-old plants those with such high numbers as 30-26 have
lowest numbers of 17-12, while those with high values at 22-20
range from 17-11. Of the one-year-old plants the ones with high
values of 33-23 have low values of 16-10, while those of 22-17
have lowest values, ranging from 17 to 9. The extremes and the
averages of the lowest numbers are quite the same irrespective of
the higher values.

Different plants exhibit greater variation in respect to the
highest number of flowers borne in any head than in regard to the
lowest, and therefore an individual plant exhibits the greatest
range if the highest number is high for that plant. The variabili-
ties of individuals and of groups increase as the upper limits of
flower number increase.

Stout & Boas: Statistical studies in Cichorium 393

These facts have a special significance in suggesting that any
evolutionary change that may have occurred or that may be now
in progress in respect to flower number is affecting the high num-
bers more than the low. The numbers in the various classes, as
revealed in tables 21, 22, and 23, indicate that there are few
plants with extremes of highest flower value and that the greater
number of plants of the general population as grown have inter-
mediate values for the highest flower number. In other words,
highest flower number exhibits fluctuating variation of greater
extent than lowest flower number and is to a large measure
independent of the latter. Data will be given later regarding
the heredity of high values as expressed for a plant as a whole by
values of a, and also as to the effect of selection on high or low
values.

table 23

Range of variability of flower number per head for 219 one-year-old
plants of the f4 generation

Number of

Maximum flower

Minimum flower number

Difference between

plants

number

Range

Average

maximum and
average minimum

I

26

12.0

14.0

4

25

16-II

14.0

II. 0

2

24

14-13

13-5

10.5

7

23

1 5-1 1

13-4

9.6

19

22

16-12

14.1

7.9

45

21

17- 9

13-4

7.6

61

20

• 15- 7

13-3

6.7

55

19

15-11

12.9

6.1

20

18

12.3

5-7

5

17

14-12

12.8

4.2

3. Relation of flower number per head to position of

HEADS

A. Descriptive studies regarding position.

Early in the collection of data, it was observed that the first
heads to bloom are, as a rule, situated on the uppermost branches,
and also that the first head which opened on a branchlet or in a
cluster of flower heads is the terminal one. This at once suggested
that the seasonal decrease so uniformly observed may be related
to a succession of bloom involving a periodicity between develop-
ment of different main laterals and also between different second-
ary branches of main laterals.

394 Semi-centennial of Torrey Botanical Club

As a rule chicory plants are much branched. Grown from seed
there is in the first year a single main stem. In the variety "red-
leaved Treviso" there is a rather uniform duplication of the single
main stem, giving two stem elements usually quite pronounced
but cohering strongly.

From the main axis numerous laterals arise which are further
branched, producing bushy plants as shown in plates io and
.II. All branches end in flower heads, but considerable variation
exists in the development of the ultimate branches, not only for
different individual plants but among the different branches of a
single plant. In some plants many of the ultimate branches are
elongated, giving a divaricate habit with many heads that appear
solitary and decidedly terminal (plate 12, marked i). Branches
which are lateral to these are usually less elongated (plate
12, 2 on plate), so that the flower heads appear sessile, but
these [in turn may have further lateral but much reduced
branches. Very often several branches constituting a system in
the axil of a leaf are all much reduced so that several heads appear
much branched and closely compacted, as shown at points indi-
cated as 3 on PLATE 12. In some plants the ultimate branches
are well developed and the clusters contain few heads involv-
ing only the last few ranks of ultimate branches. Such rather
simple grouping is shown in plate 13, ^, in the series of laterals,
all from the same plant, which show a graded transition from a
lateral with a terminal and three sessile laterals (i); to two
laterals (3); to one lateral (4, 5, 6, and 7); the series illustrating
successive stages of shortening of branches. B of the same plate
has a somewhat more marked development of secondary laterals.
In other cases there is little development of penultimate branches,
so that clusters of numerous heads are frequent and the general
branching is more sparse. The branches shown in the plate are
near the apex of the main or large basal laterals ; the larger main
laterals near the base of a plant have larger laterals near their
bases, which in turn duplicate the branching system of the more
terminal parts of the main branches.

As all flower heads are in reality terminal for the particular
branches, the distinction of terminal and lateral heads is purely
a relative one. A terminal head blooms before a head that is

Stout & Boas: Statistical studies in Cichorium 395

immediately lateral to it. Thus in plate 13 the heads marked a
in each case bloom before the one (b) lateral to it with branches

A, 1-3; however, a blooms first, but the next to bloom is b, the ter-
minal of the most basal cluster rather than c, which is immediately
below the head a. The same general behavior holds for such cases
of reduced branching as are shown in C and D of plate 13.

B. Statistical studies regarding position.

A complete study of the flower number per head according to
position of terminals and laterals for all parts of the plant would
involve a series of numbering from all apices to base in succession
for all branches. For plants of the simplest branching even, this
would be a very involved study. However, some clue to the
relationship between position, time of blooming, and flower num-
ber can be gained by a comparison of terminals with heads that
are immediately lateral to them. Such data may be obtained
without discrimination between various branches, or they can be
obtained separately for each of the main lateral branches.

Data from the intensive study last mentioned above have been
obtained from five plants. The flowers of the heads were counted
and the data recorded vvith respect to position on the various
branches. All heads were counted with the exception of those
that opened on Sundays or on days of heavy rainfall so that the
data are nearly complete for all flower heads.

The behavior of one of these plants is recorded in table 24.
The data were collected in 1915 from an Fi plant of wild white X
Barbe de Capucin, { X £3) no. 4}, whidi was then three years
old. The data are given for the unbranched portion of the termi-
nal axis and for the various successive lateral branches. The
numbers in Roman are averages for terminal heads, either solitary
or in a cluster, and the numbers in italics are for laterals, the
averages being computed for each day. The daily averages for
terminals and laterals are given for all branches. In quite the
same manner the performance of another plant, an F3 of the cross
wild white-flowered X Barbe de Capucin {{A X £22) —9-4 — no.
14} in the first year of bloom is presented in table 25.

We may first consider the comparison of relative numbers and
values for terminals and laterals as such as to time of development.
For the plant reported in table 24, no laterals opened during the

. o

i z

o

- 1

Q

a .

<(, 7)

erf J

> erf

o o

IT) O O

o o o
06 d\ K

o o

o o

10 o

o 10

Oo 06

u-2 O

Q

'Z erf
<

O O

2

m5

August

1 O 1 ••roO

^ ::::::::::::::: ^ ::::::::: : : :

1 o ■

1 ^ '. ^ .

■ ro 0

: ! ^

o • o

'. '.

• • ro

■ o ■ o

.H.N

• w >o

. . N

u->

_ • •

::::::::::::: ^ ::::::::: .o : :

■ ro t>«

: ^

July

ro

o O • - t^-i^-Q - o

• ro 0

0\

!tN.; '.00 '. ;oo !^ '. '. '. '. '. '. '.

ro N tv

•o - o - o - o t^ - •. f^-oo ■ ^

;oo ''o '<i ! ; ! ; !vd ; ■ ; ; ; ; ;ts.
N.N.i-i.N N.; i~,_N_K,

. ■ 0 <M

• • ^ t;

• o O - O - "^OO - o - o - o - o ■

'. ' ' ' ' !^ ;o\'>n^f^!iN.'vi 'vi

.N N .N .HHK, _N .N .H .N .N

cs 0 0
iri vd

N

• ■ • - ts. - • - OOt^O (NO - t^-O ^T)

'. '. '. '. '. t^- '. '. ;<>i>-K.o ioK-iK-iis.; ; ! ;

00 10 «^

0

•O ■ • -Oo • - oo - O • •lOOO'^OOO'^- - mO

00 sO ro

•00»^- • ■ - O ■ m • ■ ■CNOt^OiOOro'^- - o

it^oooo '. '. '. '. '. '. ;vdyDr--oor^<io6iN. ! !fo!

w 0\ ro

\o

• w-)-0 • ■ -O - O -OOto-CM - O • ■ - uorr^

!°o !t^! '. '. '.CO . tr^ '. ; ioK-t^io !

On Os t\i r^i

^ vd ^

• oq'^q>o-oio-vo ■ • -Qoqro-o ■'^"^'^•'^'0

^ o\ ts. fn

M N

iT

i6.o
i8.o
17.0

20.0

16.7

16.6

18.2

16.8
17.0
17.0

17.6

17.2
16.3

0 ro ts

ro

to-rooio-O • - Ooc - O •t--.-u-)-ro-ro-i>--ro-

j>-;t-^ooj>-!t^; ;tv-d '. c6 '.>>■'.[-•'. c-' '. . i-d !

00 r^ Oo ts

N

ooi>»^(NOio-oo ■ ^ ■ 0 Q> m ■ 00 -co - ro-fo-ioq
r^VDJ>-K.r^6\i>- ] \6 , . c^K^ t>- '. t>- ;o '.CO '. '. <i ^

M rg rg
^ od

M N

o

•0 - lo-t^-o -NO - o'-t^-o - o -lo-r^-
! it^ir^ir^idviod '.06 \ '. 06 ;^d !h^! '. '.

.1-1 . >-< .1-1 .1-1 .1-1 .^-1 . >-< . . >-> .M

Ov

q - qqq - ro-w - Tt-vo-q - r^-'q q •

i>.;rCi/)t^;oo |o6 '.06 ;t^|od ! '. '. '. 0\ '.

00*

00

uo-t^-O - (N - ro - • - O -O - O - lO-o

d\'i>-'od '.06 '. 0\ '. '. . '. '. '. d '.CO '. 0\ '. CO '. '.

M.M.I-I.M.M...M...(N.M.M.M...

0 ^ <n

^06 t;

^ od Ov
M K,

00 - to-q - c^ .up.io - • - q -q ■ 0 -q • 0 - ro-
od ;d\!d\;od '. 6\ '. 6\ '. ; ;d\;r^!d !d\;d\!od ;

r- 0\ N 0
^od oc^

M N

ro

i>...ro iJO

6 '.'. '. 6 '. 6\ '. '.'. '.'. '. '.'.'.'.'.'. '.'. .'.'.'. '.

C^...O).M

^ dv : :

ro . q • q • U-)

6 '. 6 '. 6\ '. 6\ '.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'. '.

^ d : :

June

0

ro 0 • •

d : :
w . .

0\
M

" ° : :
- : :

Branches
starting at
top

(U

o'o s: ^

X V >

398 Semi-centennial of Torrey Botanical Club

TABLE 25

Seasonal distribution of flower number per head for first year of bloom of

PLANT (A) AND A PLANT OF BaRBE DE CaPUCIN (£22),^DAILY AVERAGES ARRANGED

(Terminals, in ordinary type; laterals, in italics)

Branches

June

July

stfirting"
at top

16

18

22

25

26

27

29

I

2

3

17.0

17.0

17-5

Main

16.3

77.0
16.0

16.0

16.0

13.0

7(5.0

17.0

19.0

I

17.0

16.0

17-5

19.0
17.0

2

17.0

16.0
17.0

18.0

77.0

3

16.0

17-5
19.0

17.0

16.0

IQ.O

7(5.0

17.0

16.0

15.0

4

17.0
15.0

7(5.5

17.0

16.0

18.7

18.0

16.0

5

17.0
17.0
17.0
18.0

7(5.0

17.0

19.0

20.0

16.0

6

i7-5_
16.0

16.5
15.0

17.0

i<5.5

16.0

7(5.5

19.0

19.0

19.0

7

17-5

16.0

15.0
16.0

17.0

16.5

15.0

8

16.0

16.3
15-0

15-7

15.5

7(5.0

19.0

18.0

19.0

17.0

16.0

16.5

15.0

16.0

9

16.0

16.5
16.0

19.0

18.0

16.5

16.5

10

i-5.0

7(5.7*
17.0
7(5.5
16.0

17.0

16.0

17-5
77.0

18.0

II

7(5.0

16.0

16.7

12

7(5.0

15.5

No. of 1
heads 1
with 1

4

17.2

6

17.4

3

17,3

I

17.0

4

6

16.8

2

3

19.0
2

5

18.0
2

10
18.5
I

7

17.4

6

7

16.7

6

6

16.0
2

II
16.3

8

5

15-4

7(5

7

16.3

70

6

15-8
22

averages I

16.5

16.5

17-5

16.5

17.0

16.7

16.8

16.0

16.4

75.<?

15-9

first five days of bloom; no data were obtained on the sixth and
seventh days of bloom, but on each of the eighth and ninth days
one head opened. From then on the number ot such heads grad-
ually but steadily increased and the number of terminal heads
decreased. From July 29 to the end of blooming period (August
16) only lateral heads bloomed.

It is also plain that the average number of flowers per head for
both terminals and laterals steadily decreases as the season of
bloom advances, and that both the maximum and minimum
numbers are lower for laterals. In respect to the total number of
flower heads, only indirectly shown by averages in tables, there is

Stout & Boas: Statistical studies in Cichorium 399

Table 25 — Continued

AN F3 GENERATION PLANT (A X Eii)-9-4- nO. I4, DERIVED FROM CROSSING THE WILD
ACCORDING TO POSITION OF BRANCH AND OF HEAD ON BRANCH. DaTA FOR I915.

July

Augxist

14.0

16.0

Ij-O

6
14.8

I5-0
15.0

15.8

19.0
I

15-0

3
16.3

15-0

15.0 1/(5.0

15-0

2\ 3
15-0 75.0

17.0

15-5

16.0

15-0

15-

17.0

I

17.0

17.0

I

17.0

14.0

I

14.0

15-5

16.0

3

15-7

16.0

3

16.0

15.0

15-5

3

15-7

14-5

2

14-5

17.0

15-5

14-3

7

14.7

15.0
14.7

I

15.0
5
14.0

15-0

I

15-0

also quite similar performance in that the maximum number for
both terminal and laterals is about midway in the period of their
appearance ; the maximum for laterals therefore is on a date slightly
later than the maximum for terminals.

In comparing the terminals and laterals as summarized in
respect to rate of decrease (h) and value of first day of bloom (a)
some differences appear. The terminals bloomed for a period of
28 days and gave a value of 19.6 for a and for b of —0.138. The
laterals bloomed 41 days and for these the value of a is 18.0 and
for 6 is —0.045. The rate of decrease in number per head for the
laterals is less than that of the terminals.

The other four plants, which were studied intensively, agreed

400 Semi-centennial of Torrey Botanical Club

with the behavior reported above for general performance of
terminal and laterals. However, minor differences were present, so
the performances of no two plants were identical in detail. This
may be shown by data for one other plant as given in table 25.
Here the laterals came into bloom much sooner and a few terminals
appeared very late in the season.

We may now consider the relative performance of different
branches of a plant.

For the plant reported in table 24, the first head to bloom was
situated on the lateral branch that was sixth in rank from the top.
On the following day a head opened on each of branches no. 4, 5,
and 7; on the next day heads opened on i, 2, 3, 8, and 9; on the
following day heads bloomed on the main terminal branch and
on 10-16. No data were collected on the 4th and 5th, but on the
6th, or eight days after first bloom ^ all the main and all the lateral
branches of the plant had produced at least one flower head.

In general the performance of each branch is quite like that
of the plant as a whole. Terminals come into bloom first, laterals
continue to bloom later than terminals, the average flower number
per head decreases as the season advances. A comparison of the
performance of different branches, however, shows that the averages
of the first flower heads on most branches are not decidedly dif-
ferent, which means that the rate of decrease in average number
per head for a plant as a whole is usually less during the first part
of the season while the different branches are coming into bloom
than it is later when all branches have come into flower and most
of the first terminals have bloomed.

Individual differences for the various branches in the relative
time of blooming are in evidence and constitute a factor contri-
buting to the rate of decrease as determined. For the plant re-
ported in table 25 eleven days elapsed before all the main branches
were in bloom. Some irregularity was also seen in that the seventh
and ninth branches started bloom with higher values per head
(19) than were realized in the first heads to bloom on branches
I, 2, 4, 5, and 6. Such a performance influences the actual curve
of average flower number and rate of change for the plant as a
whole, giving for the first few days an actual increase in the average.

The relative average performance of each branch and the

Stout & Boas: Statistical studies in Cichorium 401

average performance of the plant as a whole are shown for each
of the two plants discussed above in the diagrams of tables 26

table 26

Diagram showing decrease in flower number per head for plant {A X £3)
no. 4 AS A whole and for its different branches during a single season
OF BLOOM. Drawn to a scale with ordinates giving flower number

AND abscissas REPRESENTING PERIODS OF BLOOM

A. Entire plant.

B. Branches 4, 5, and 7.

C. Branches 17 to 25.

D. Unbranched portion of main axis and branches 10 to 16.

E. Branches i, 2, 3, 8, and 9.

F. Branch 6.

TABLE 27

Diagram showing decrease in flower number per head for plant as a whole

AND FOR different BRANCHES. FrOM DATA OF PLANT (A X E22)-Q-4-nO. 1 4.

Ordinates are flower number per head; abscissas represent periods of

BLOOM

A. Entire plant.

B. Branches 7 and 9.

C. Branches 8 and 10.

D. Branches 4, 5, and 6.

E. Branch 11.

F. Branch 12.

G. Branch 3.

H. LTnbranched part of main axis and branches i and 2.

and 27. To avoid the complication of many lines, branches that
started to bloom on the same day were averaged as indicated.

402 Semi-centennial of Torrey Botanical Club

The performances of the different branches of the plant shown
in TABLES 24 and 26 were more similar than those of the different
branches of the plant reported in tables 25 and 27. The graphs
in the latter case ovi^rlap and cross and the graph for the plant as a
whole is less in agreement with that of a single branch.

From such studies of the individual behavior of the different
branches we gain some clue to the immediate causes of the irregu-
larities which appear in the data collected from day to day, and
of the irregularities that appear in the rate of decrease in the num-
ber of flowers per head. Differences in the relative development
of the various branches, and the number of branches produced,
have a marked influence on the average flower number observed
from day to day and in the corresponding rate of decrease.

It is apparent that the irregularities are due chiefly to varia-
tions in behavior of the different branches during the first twenty
or twenty-five days of blooming and ate chiefly due to the number
of branches and to the rate with which these come into bloom.
The averages for the whole plant may show no decrease, a very
slight decrease, or even a slight increase (see table 27) during the
period of first blooming, due to the fact that day after day for a
longer or shorter time new branches begin to bloom all with a
high flower number per head. If, however, a considerable number
of branches have been in bloom for some time and are showing a
marked decrease in flower number per head when late branches
come into bloom, the higher flower numbers of the latter will for
a time retard the rate of decrease. The irregularities occur chiefly
at the beginning and the end of the period of blooming. In all
cases the rate of decrease is less until all branches are in bloom.

It would appear that the most typical development of a plant
is such that the branches from the oldest to the youngest show
characteristic differences in the number of flowers per head and
in the rate of decrease, which may be somewhat analogous to the
behavior of a plant at different ages. The uppermost branches
in comparison with the lower branches of the same plant are, as a
rule, smaller, have fewer heads, bloom for a shorter period, and
show less decrease in flower number. While the variations in
and among the various branches of different plants considered as
wholes are so pronounced that a perfect or exact type of develop-

Stout & Boas: Statistical studies in Cichorium 403

ment of this sort is not realized, the strong tendency to such a type
is more or less evident from the data which show that there is a
decrease for branches according to the relative time of blooming,
which merges into and contributes to the general decrease of the
flower number for the plant as a whole.

The more absolute calculation of the decrease of average num-
ber per head, therefore, involves the factor of position of the va-
rious branches involving their age, length of life, and number.
This decrease, due to position, may be considered as an unknown
c and its value calculated in the same manner as that of h. If n
be used to represent the number of the branch, we may introduce
these two new factors into the equation which now becomes
0 — a -\- ht -\- cn. Three simultaneous equations can be developed
by the same methods already used, giving

(1) \o\= a-\-h\t\^ c M

(2) \ot\ = a\t\^h \f \ + c \tn\

(3) \on\ = a [w] -f h {tn\ + c \v}\

In order that the rate of decrease for each branch be given its
full weight, uninfluenced by the high flower number of the branches
that begin to bloom later, the data were so arranged that the
same starting point was taken for all branches; that is, all the
observations for first day of bloom, second day of bloom, etc.
were grouped together, regardless of dates. From table 28 the
values of a, 6, and c for the plant {A X £3) no. 4 were calculated.
In order to simplify the calculations the observations are put
down as + and — deviations from 18.0.

Comparisons show that both series of calculated values cor-
respond fairly well with the observed values. It does not, of
course, appear from the data on this one plant that the values
computed when both time and position are taken into account are
in better agreement with the observed values than if time alone
be considered. In order that the reader may judge of the amount
of agreement, there is presented for comparison in TABLE 29 (i) the
flower numbers actually observed on various dates after the
first day of bloom without taking position into account, (2) the
theoretical values for the same dates with regard to the factor of
time, and (3) the values when both time and position are taken
into account.

y

404 Semi-centennial of Torrey Botanical Club

TABLE 28

Table for calculation of value of first day of bloom (a), where position of head upo^

WHICH DATA OF SEASONAL DISTRIBI.

Days

I.O
I.O
1.0
I.O
2.0
2.0
I.O
2.0
I.O

1-5

2.7

I.O

1-3
2.3
2.0

I.O

1-5
1-5

2.0

I.O
I.O

0.3

33.6

I.O
2.0
I.O
2.0
1-5
2.7

2.3

1-5
0.5

I.O

0.5
0.5

iS-7

■1.0
2.0

I.O

1-3
0.4
■0.5

1.5

0.0

6.7

0.2
0.6
0.0
0.0
•0.3
0.0
■1-5

I.O

I.O

2.0

0.8
1-5

I.O

0.2

I.O

— 0.1

-0.3
0.0
-3-0

-0.3
-0.8

I.O

0.0
0.8

0.5
0.0

0.7

•2.0
0.3

4.01 2.1

0.5

1-3

0.5

I.O

-0.3
0.0
0.2
-1.2
-0.6
0.0
■0.7
■0,2
-1.2
-0.7
-0.7
-1.6

•2.7

2.0
0.0

I.O
I.O
I.O

0.3

I.O
1.2
1.3
0.3
0.5
0.7
I.O

1-3
1-7

II. 7

■0.5
■0.3
■0.5
2.0

-1.4
0.2
-1.2

-0.4

-I.O

0.0
■2.4

-3.i|

•1-5
■0.7
■1.4
•2.8
■0.6
■1-5

■9.6

1-5

■i-S
■0.4
-0.4
■0.5
-1.0

•I.O

1-3
■0.5
■1.8
■1.0

1-7

■8.6

I.O

2.5
2.0

0.5
0.5
0.5

I.O

2.7
0.7

1.9
0.2

I.O

0.5

5-0

0.0
0,0

0.0

I.O

■I.O
■2.0
0.0
■I.O

16.0 —I.O

I.O
0.0
0.0

I.O

■0.7

0.0
I.O

3-3

0.0
0.0

0.5

— I.O

— 2.0
0,0

■I.O

■0.7

■4.9

-o.<

[o] =

[ot] =

[on] =

[t] =

- 0.2

- 8.3

- 7.1

10.8

= 187.79

[n] = 14.9

[W2] = 268.49

[tn] = 160.37

The collection of data and the calculations which take the
rate of decrease of the various branches into account, while simple,
are extremely laborious and consume much time. The collection
of such extensive data for a very large number of plants is scarcely
possible. Such data are desirable in the analysis of the sources of
the variability and may indicate the proper methods of obtain-
ing from a larger number of plants the data that will admit of
more general analysis and comparison.

The question naturally arises whether in a plant with a long
blooming-period such as chicory the random collection of data at
intervals of every third day throughout the flowering period will

Stout & Boas: Statistical studies in Cichorium 405

TABLE 2%— Continued

PLANT IS TAKEN INTO CONSIDERATION. DATA FOR THE Fl GENERATION PLANT {A X Ei) MO. 4, FOR
TION ARE PRESENTED IN TABLE 24

Days

16

18

^9

23

24

* 25

26

27

28

29

3*^

3^

32

1.0

- 0.5

0.0

1 .0

— 3.0

—3.0

- 3-0

— 2.0

— 1.0

0.0
1.0

0.0

0.0

0.0

0

"-'•0

0.3

0.0
— 2.0

0.5
0.7

— 2.0

— 2.0

— 1.0

— 1-3

-3-0

-2.5
-1.7
— 1.0

— 2.0

0.8

-0.3

— 2.0

— 2.0
-1.7

0.0

— 2.0

1.0
1.0

1.0

-1-3

— 2.0

— 1.0

— 2.0

— 2.0

— 2.0
-i-S

— 2.0

-2.3

0.0

- 1.3

— 2.0

-1.3
-0.3
— 1.0

-1-7

— 2.0

— 2.0

— 1.0

- 3-3

- 0.5

— 2.0

— 2.0

-1.7
— 1.2
-0.7

-5-3

-0.2

-2.3

-12.4

-2.6

— 2.0

-4-7

-6.7

-19.1

0.0

-5-9

2.0

-5-0

-5-0

— 2.0

-3-0

Totals

3-7
0.0
2.0
2.0

2.5

3.0

1.0

- 0.7
10.3

6.8
8.0
3.8

- 1.6

- 0.3
-11.6

- 1.3

- 0.4

- 7.9

- 3-2

- 7-9

- 16.4

- 4-7
-17. 1

- 9-1

- 7.5
-18.7

1 — 0.2 = a + 10.86 + 14.9c

2 — 8.3 = 10. 8a + 187. 79& + 160.37c

3 — 7.1 = 14.9a + 160.376 + 268.49c

Solving, h — — 0.086
c = - 0.095
a = + 2.2 (20. 2^

adequately represent the variations in flower number per head
and give data for a rate of decrease for the plant as a whole which
adequately considers the variation due to position. Some evi-
dence on this point is to be had from the relative dates upon which
branches with the same relative position come into bloom on dif-

table 29

Observed and theoretical values of flower number per head for an Fi
generation plant {a x £3) uo. 4, from data given in tables 24 and 28

0

4

14

18

29

Observed values

19.0

19.6

17.8

17.2

16.8

Values computed for time (a — 0.066/)

18.9

18.6

18.0

17.7

17.0

Values computed for time and position (a —0.086/ — o.095n)

19.0

18.4

17.S

17.6

16.2

406 Semi-centennial of Torrey Botanical Club

ferent plants. Such data for the main and for the successive
fifteen branches of seven plants are presented in table 30. For
each plant the date of first blooming of any branch is considered
as zero and all later blooming calculated from this. Observations
are not complete for all branches, as indicated by the dashes.

table 30

Relative day of first bloom for main stem and 15 branches numbered from
TOP down, for seven plants

Plants

2

3

4

5

7

Average

Branches, starting from
above

Day of first bloom

Main

0

I

4

0

6

5

2.7

I

0

0

3

0

3

II

2.8

2

0

I

3

0

2

6

2.0

3

I

I

2

5

2

2

5

2.6

4

I

I

2

I

5

I

5

2.3

5

I

0

2

I

5

I

4

2.0

6

I

0

0

5

0

4

1.7

7

2

I

6

I

3

2.6

8

I

4

4

3

8

I

2

3-3

9

I

4

4

3

6

6

I

3-7

10

I

5

4

4

8

6

0

4.0

II

3

4

2

4

9

8

0

4-3

12

3

5

4

10

10

0

5-3

13

3

5

4

4

3

3.8

14

3

5

4

4

5

4.2

15

3

4

4

8

4.8

None lower observed

While the number of plants observed is not large, it is suf-
ficient to show variations in the type of development and illus-
trates the fact that there is considerable variation in the position
of the branch that first comes into bloom and likewise in the rela-
tive time in which branches similarly placed come into bloom.
While it is generally one of the uppermost branches that blooms
first, the general development of a plant with respect to size,
number of branches, etc., may be such that, as seen in plant no. 7
of Table 30, the loth, nth and 12th branches from the top may
bloom first. The largest variability for branches of any one posi-
tion is eleven days, seen in the first from the top, and the least
is two days, exhibited by branches 13 and 14.

In spite of these irregularities, it appears that for the average
relative date of bloom for the different branches of individual
plants there is little individual variability and that one may secure

Stout & Boas: Statistical studies in Cichorium 407

quite adequate data by making collections from the plant as a
whole, especially when the data are well distributed over the
season and form a rather large total.

C. The phenomenon of intermittent annual growth.

Opportunity for further study of the influence of position and
time of development on the number of flowers per head was given
by a very marked variation of vegetative habit, giving discon-
tinuous development or two periods of growth in a single year.

Text-figure i. Two plants of the race (race 3) exhibiting intermittent growth-
Photo taken late in autumn.

To left a plant with first growth cut away, leaving only second growth.

To right a plant showing both first and second growth. The first growth was
about twice as tall as the second and when photographed was dead and dry while the
second growth was green and flowering profusely.

As a rule the seasonal development of the various branches is so
continuous and overlapping that there is a sequence of bloom and
no interruption in the blooming of a plant as a whole. The last
to develop of the main laterals are the most basal of the laterals
and are from the uppermost of the rosette leaves.

408 Semi-centennial of Torre y Botanical Club

TABLE 31

Seasonal distribution of flower number per head for a one-year-old plant of an

PLANT (A) AND A PLANT OF BARBE DE CaPUCIN (£22).

Number
of flowers
per head

August

September

2

5

8

10

12

21

24

26

29

I

5

7

II

13

I

13

I

14

I

I

I

T cr
■■■ J

I

I

I

7

7

3

3

2

4

4

4

2

I

16

2

I

2

6

/

3
I

J

5

7

2

7

2

4

2

3

4

2

17

I

I

I

2

2

4

J

J

3

7

4

I

I

I

I

18

I

7

7

3

3

I

I

I

19

I

I

7

I

Averages

17.6

17.2

16.6

16.3

16.6
16.5

16.7
16.0

16.3
16.5

16.4

16.7

17.0

17.2

7(5.<?

16.0

16.8

17-3

17-3

16.7

16.0

Old growth [o] = 16.7 a = 16.9

[ot] — + 2.79 & = + 0.009

[t] =26 t = 54
m = 271.55

In one line of descent there was a marked deviation from this
habit of seasonal growth which was very striking in that an entire
series of sister plants in 1916 exhibited this deviation. After the
branches that usually develop were through blooming, young and
new branches appeared from the axils of many of the rosette
leaves (which had died) ; these continued to grow, making a bushy
compact second growth quite distinct from the earlier growth both
as to time of development and to position on the plant. This habit
of growth is shown in text-figure i . To the right is a plant
with the growth of both periods shown; the older dead branches
of the first period extend above the newer growth. To the left
the old growth has been cut away, leaving only the branches of
the second period. Between the close of the blooming of the
first growth and the beginning of bloom on the second growth
there was usually an interval of several days.

Complete data collected for the flower heads of two periods of
growth for a single plant {{A X £22) —9-5-12 —wo. 6} are pre-

Stout & Boas: Statistical studies in Cichorium 409

TABLE zi— Continued

¥i GENERATION {A X E22)--9-5~1 2- HO. 6, DERIVED FROM A CROSS BETWEEN THE WILD

Data for 1916. Terminals, in ordinary type; laterals, in italics)

September

October

November

16

22

25

5-7

9-10

10-13

14-16

17-20

21-23

24-26

27-30

-2

4-6

8-9

11-14

16

I

I

Death

2

3

2

I

5

2

4

4

I

I

3

2

2

7

9

2
2

7

I

2
/

2
2
I
J

3

2

I

I

I

I

I

I

2

I

I

z
I

I

I

6

I

I

I
I

I

I

I

J

I

14.8

15-3

14.8

14.9

16.1

14.1
15-0

15.0

14.0

14.7

16.0

15.0
15.6

15.6
16.0

15-0
77.0

15-0

17.0

16.4

17.5

17.0

New growth [o] ^ 14.9 a = 15.1

[oi] = + 1.64 h = -\- o.oii

[i] =19 i = 42
= 142.83

sented in table 31. The growth of the first period ended bloom
on the 25th of September. The data collected for this period
showed that the averages from day to day throughout indicate an
increase in flower number per head {h = +0.009).

The new growth began bloom on the 5th of October and had
only partly completed bloom when it was killed by frost. The
data for the new growth are necessarily incomplete and mostly
from terminal heads. Data were collected from the new growth
every day, but to'avoid extending the tables the data are summarized
for every consecutive three days of collection ; the totals for each
column are therefore larger than when a column gives only data
obtained in a single day. What was collected showed that there
was here also an increase of number per head {b = +0.011).

The performances of the two periods were similar in yielding
a slight increase in number per head. The average number per
head and the value for the first date of bloom for the two were
much higher for the first period. The lower branches which formed

410 Semi-centennial of Torrey Botanical Club

TABLE 32

Seasonal distribution of number of flowers per head for a one-year-old plant of an

PLANT (A) and a plant OF Barbe de Capucin (£22). Data for

July

August

September

31

3

7

9

12

18 i 21

26

I

6

8 1 .X

13

16

7

J

7

7

7

I

7

3

I

7

I

I

I
I
I
2
I
3

/

5

2
2

2

J-

X

2

I

3

4

I

I

6

5

2

2

J

6

4

2

3

2

I

3
3

2
I

2
2

3

3

4

5

2

3

3

2

2

J

I

I

3

J

/

2

2

I

• I

7

2

I

I

I

15-5

17.9

16.0
77.0

16.4
17.0

16.0

J7.2

16.6 16. 2

/(5.7

17.2

17.0

16.2

16. 1

15-4

15-9

75-9

14-3

Number
of flowers
per head

13
14

15
16
17
18
19

Averages

Old growth [o] = 16.1

[ot] = - 7.90
[t] = 29

[t-] = 246.40

a = 17.0
b = — 0.032
t = 57

the growth of the second period started in with lower average
numbers per head.

Most plants of this race exhibited a minus value for rate of
change (an actual decrease) : the typical performance being as
shown in table 32. For this plant the growths of the two periods
both show a rate of decrease. The values of a and [0] for the new
growth are decidedly lower than those of the old growth.

4. Individual variations from the usual performances as
to seasonal decrease

Marked deviations from the usual behavior of seasonal de-
crease have been found. Data for two such cases have already
been given in tables 3 and 4.

Stout & Boas: Statistical studies in Cichorium 411

Table 32 — Continued

Fi GENERATION (A X E22)-Q~3-I2- flO. JI, DERIVED FROM A CROSS BETWEEN THE WILD

1916. (Terminals, in ordinary type; laterals, in italics)

September

October

November

19

22 j 25

27

5-7

9-1 1

13-14

16-17

18-20

21-23

24-26

27-28

1

30-1

2-4

6-8

9-ri

14

2

I

2

I

2
J
I
I

6

6
2
I

/

4
I
7
I
I

2

3

3

2

4
I
7

8
I

2

5
z
4

2

2
8
7

2

I
J

3
2
2
I
I

4

I

I

/

8

4

9

3

4
J
3

X

I

I

6
J
4

2
J

4

J

J

2

3

3

4

■3
J

I

I

I

*5

5

5

2

I

I

I

I

/

I

I

15-3

14.7
14.0

15.2

15.0

16.0
16.3

15-3
j(5.o

14.8
15-0

15-4

/5-0

15-7

14.8

15.2
15-0

15.2
74-7

15-0

14.7
15-0

14.4

14.6

15-2

16.0

New growth [o] - 14.9 a - 15.2

{ot\ - - 0.34 6 = - 0.003

[/] =19 / = 40
= 123.00

These two plants were grown from seed in the summer of 1915;
both were about 31/2 feet tall, rather sparsely branched, but with
numerous flower heads, which were typically terminal and lateral.
These plants were quite identical to all appearances in respect to
habit of growth with numerous plants which gave a marked de-
crease in flower nu;riber as discussed above. Although data were
not taken for terminal and lateral heads separately, it is evident
that there were only slight differences between such heads. Time
of blooming and position, it seems, were here not factors giving
partial variability. The performance of such plants is indicated
by the low values of h. If all plants exhibited such behavior, it
would not be necessary to consider the seasonal performance in
determining the values which were to be used for the comparison
of the different individuals.

412 Semi-centennial of Torrey Botanical Club

In the course of the studies a few plants were found which in
the performance for a season exhibited an actual increase of average
flower number per head. This is quite the reverse of the usual
performance. Data for one of the most marked cases of such
increase are given in the following table (table 33). This plant
was one of the F4 generation grown in 191 6 and the data as col-
lected distinguished between terminals and laterals for the dif-
ferent clusters. The average or mean number for all flower heads
is 17 and the value for the first date of bloom (a) is 16.4. Neither
of these is an especially low value. It should be noted that this
plant was one of a series of sister plants which were quite uni-
formly sparsely branched and had few solitary terminal heads:
the heads were in clusters, quite as shown in C and D of plate 13.
Such clusters evidently represent a shortened and compacted
system involving terminals of different relative ranks with their
respective laterals. The group is so compact that only the first
head to bloom can be definitely regarded as terminal. If the
group had not been compacted (but expanded as in A and B of
PLATE 13), many of these heads would have been terminal for
their respective clusters. The prevailing high numbers in the
so-called lateral heads of these groups as seen in table 33 may indi-
cate that in these compacted groups of heads the laterals of lower
ranks are crowded out and fail to develop as they may when the
branching is more profuse. However, it must be recognized that
many plants with the grouped-head habit, quite identical with the
one under consideration, showed the seasonal decrease most char-
acteristic of the species.

5. Variation in partial variability with the age of a plant
The data collected from plants in successive years of growth
may now be presented with respect to the very important question
of variation in seasonal performance of a plant from year to year.
In the case of all perennials and especially of herbaceous peren-
nials this point needs careful analysis before adequate comparisons
between individuals can be made. All of the plants studied, with
the exception of the variety red-leaved Treviso, grew as perennials,
and although plants of each series winter-killed, the roots and
basal portions of such plants were apparently fully alive at the end

Stout & Boas: Statistical studies in Cichorium 413

0 >.

1 i

Q o

^ o

Z to

2 <

< ^
Z

M M C\)

O

; i ; M ; ^ : i M

: o
•

ober

0

:
•

Oct.

; .■ ; ; ; ; ; : ;

: o
• 06

: 0
• d

: 0
• 00

N
P)

■

On

M M " M ; " i "

;

■ t-L

vO

: : : : : . : : h

• ^.

• \d

ptemb

M

. '. '. '.

:

• t-i

; ! 1^ ; ; : r\) ; :

;

• <i

C\J [ (\J [ C\j " f\) ] (\) [ \

; 0

'. '. '. '. ^ ! .' H '. '.

; <^

'. '. ; ^ ; ; H ; ;

■ ^

: : : : : ^ : : :

• °°

; ! '. '. ! ^ '. '.

; : ; ; tr^ m cm : -i- :

q ►-I

; ; ; ; N : <>i ; n

•

August

00

: H : ; : : N : :

q oq

IT) VO

IT)

'. '. '. ; M (\) M c\i ; ;

10 Oo

P)

'. '. ; ro ! (N M ; ;

OS

; N ; fo 1^ M CM M ;

M

0

; iH : so ; ro : ; : : :

w ;

ro

* ; N ; ;

q :

06 •

July 1

ro

q :

Number
of fiowers
per head

lo o 00 0\

M M M M M M

Averages

414 Semi-centennial of Torrey Botanical Club

of the previous season of growth. The plants of the red-leaved
Treviso variety died at the end of the first season of bloom.

A perennial plant of chicory grown from seed exhibits consid-
erable interannual variation in habit of growth, especially in the
first and second years of growth and bloom. In the first year of
bloom a single main erect stalk usually develops from the rosette;
in the second year, usually, several main stalks are produced
directly from the extreme basal and lower parts of the crown, or
even from clusters of roots that may or may not have become separ-
ated by the death of the basal portions of the stem of the previous
year. In a few years the cluster of roots is more or less increased
in size and number and the group of main branches is correspond-
ingly increased and more or less crowded together. The degree
of such development varies considerably. The wild white-
flowered plant A showed rather feeble increase or spread of this
sort; the wild white-flowered plant C was somewhat more vigorous
in its vegetative development; and plants of the Barbe de Capucin
(E series) were most vigorous in this respect, so that from E3 and
£22 in the third year of growth a large number (15 or 20) of stems
arose from the roots, and as some of these developed from small
detached roots they were weak and late in developing. The
crowding of such a number of stems also caused the poor develop-
ment of many of them. For the perennial plants upon which
data were collected, it has been the policy to allow no more than
three or four main branches to develop. These were selected
from the first shoots and were as nearly uniform in development
as possible; all other weaker or later branches from the root cluster
were cut away.

It has already been reported that in the second year of growth
from seed, plants are taller and more branched and have a much
longer flowering period with the production of many more flowers.
These differences in general habit of growth are represented by
such differences in data as appear between tables 8 and 9 for
example. These differences raise the question whether such values
as those calculated for the first day of bloom (a) and for the rate
of decrease (b) are consistent for a plant in the successive years
of its growth from seed.

Some judgment of this question may be gained by the correla-

Stout & Boas: Statistical studies in Cichorium 415

tion of data for those plants from which data were collected in
different years. The following tables (34, 35, and 36) present
such data, in values of a, which are grouped solely according to
the age of the plants.

TABLE 34

Correlation between flower number for first day of bloom (a) of one-
year-old AND THAT OF TWO-YEAR-OLD PLANTS

Flower number of 2-year-old-plants

Averages

o ^
-9 a

o ^

16 17 18

19

20

21

22

23 24

16

I

I

17

I 7

3

3

18

3

3

6

19

4

2

I

20

I

I

2

I

21

I

I

I

22-

23

I

24

18.5
18.6

19-3
20.6
19.6
20.3

24.0

0.68

TABLE 35

Correlation between flower number for first day of bloom (a) of three-
year-old PLANTS AND THAT OF ONE-YEAR-OLD PLANTS
Flower number of 3-
year-old plants

o w

<V 03
O ^

16

17

18

19 20

21 22

16

I

I

17

I

5

18

4

2

19

2 6

20

I

I

21

I

I

22

r

+ 0.68

Averages

17-0
18.8
19.7
19.8
20.5
21.0

TABLE 36

Correlation between flower number for first day of bloom of two-year-

OLD and that of THREE-YEAR-OLD PLANTS

o s

u C

Flower number of

3-year-old plants

18 19 20 21 22

Averages

18

2

18.0

19

I 2621

19.0

20

10 2 3

19-5

21

3931

20.1

22

I 2 I I

20.4

r = -\- 0.60

416 Semi-centennial of Torrey Botanical Club

Table 34 presents the correlation of data for first and second
year of blooming for 44 plants; the coefficient of correlation is
+0.68. Table 35, involving 26 plants in the first and third
years of growth, shows a coefficient of correlation of +0.68. The
values of a from 50 two- and three-year-old plants show a coeffi-
cient of correlation of +0.60 (table 36).

The values for the coefficient of correlation are remarkably
uniform, which may be considered as more significant than the
particular value. The correlation is also high, which, with the
distribution in the tables, indicates very clearly that plants which
exhibit a high value for a one year will also give high values in the
following years. The value for a is strongly individual and tends
to remain quite constant from year to year in spite of such dif-
ferences as exist in the growth of plants in successive years. At
least such is the case when the data used in the calculations are
collected as described from plants grown as the chicory plants
have been.

It should be pointed out that the plants utilized in this com-
putation belong to a small number of lines of descent and that
they are for the most part Fi, F2. and F3 progeny of crosses in-
volving the wild white-flowered plants {A and C), and those of
the cultivated variety (£3 and £22)- The parents were widely
different in general habit and the Fi and F2 generations likewise
exhibited wide variations in such characteristics as height, vigor,
general habit of branching, and period of blooming. The popu-
lation as grouped in tables 34-36 is composed of five lines of
descent. If these lines of descent showed great differences in
flower number, the correlation from year to year would appear
greater than it really is. From table 37, however, it is evident
that for plants of the same age the distribution of flower number
in the different lines is essentially the same. In this table the
frequencies of occurrence of flower number are recorded for each
line separately. The interval used in grouping the flower number
is 0.3; that is, the group 16.0 — .2 includes 16.0, 16. i, and 16.2.

Table 37 shows data for different lines of plants in the first,
second, and third years of growth. It will be seen that all the
lines overlap. The series (£3 X^)— 4 — , an F2 generation, is
the only one that shows in the first and second year one plant with

/

Stout & Boas: Statistical studies in Cichorium 417

a very high value for a. The other plants of this series, however,
show a range of variability similar to that of the other series.
In the third year one plant of the series (£3 X C) had a very low
number of flowers per head, but the rest of the series agrees with
the general distribution of the other series.

TABLE 37 '
Distribution of flower number per head (a) for one-year, two-year, and three-year-old

PLANTS OF different LINES

Flower
number

One year old

Two years old

Three years old

X

it?

X

X

X

X

X

3

X

X

X

X

^

<o

X

X

X

X

s

'I

X

X

5

X

X

§3

A

"o

X

S3

A

Co

X

X

X

S3

X

X

S3

X

Co
—

Co
X

X

Co
—

16.0- .2

•3-- 5
.6-.8
.9-. I
17.2-.4

•5-- 7
.8-.0

18.1- .3
.4-.6

.7--9

19.0- .2

•3--S
.6-.8
.9-. I

20. 2- 4

.S--7
.8-.0

21.1- .3
.4-.6
•7--9

22.0- .2

•3-5
.6-.8
.9-. I

23.2- .4
.5-7
.8-.0

24.1- -3

I

I

I

2
I

I

2

3
2

T

I

2
I

I

1 . .

I

I
I

I
I
I
I
I

I

2

3

I

3

I

I

I

I

I

I
I

I

I

I

••

I

I
3
I
3

2
I

2

2

I
I
I

I
I
I

2
I
I

I

3
I

I

I

I

I

I

I
I

I

I

I

3

I

I

I

I

I

I
I

2

I
2

3

I

I

I

I

I

I

2

I

2
2
I

I

I
I

I

I

I
3
I

2
2

I

2

2

I

••

I

I

I

I

3
I

2

I

I

I

I

2
I

I

I

I

I

I

I

I

I

I

It is quite evident then from this table that the differences in
the distribution of flower number between the different series are
not great enough to influence the coefficient of correlation for
plants of different ages.

The plants A and C were at least in the third year of growth in
1 91 3 when they were first used for data here reported. The

418 Semi-centennial of Torrey Botanical Club

general habit of growth of each was quite uniform in 1913, 1914,
and 1 91 5. The parents £3 and £22 were in their second year of
growth in 1913. During the winter of 1914-1915, the plant £22
was destroyed by pine mice. In the early spring of 1915 the roots
of each plant {A, C, and E3) were divided in two and grown as two
plants, from which data were taken separately. The data for
these various plants and for the hybrid generations are tabulated
in TABLES 38 and 39.

It is clear from these comparisons that there is a rather close
agreement in the behavior of a plant in successive years with re-
spect to the production of flower number as determined by the
calculated values for a. The nature of the development of a plant
and the many factors contributing to fluctuations in development
and production of flower number are such that exact agreement is
not to be expected. The variations in the value of a for any one
plant from year to year follow a rather uniform course. The
value is, as a rule, lower for the first year of bloom than for the
second and there is a further slight increase in the third year.
The behavior of a plant is more uniform in the second and
third years of growth. The data for plants A, C, E3, and £22
indicate that such is also the case in later years.

TABLE 38

Flower number per head (o) in three successive years of plants three or

MORE years old. DaTA FOR I913, I914, AND I915

Plant

Value of a

1913

1914

'915

A

19.2

19.2

f 18.6

I 19-3

B

19-3

18.0

I 18.5

, 17.6

C

21.5

21.8

1 23.0

1 21.8

£3. . .

20.6

20.8

' 20.8

^ 20.6

£22. .

23.0

23.6

dead

Table 39 gives the values of a for plants of different ages, for
the entire Fi and F2 population and for the various lines of descent.
For the Fi plants, as a group, the average the first year is 18.8;
for the second year is 20.6 and for the third is 21.0; the average

Stout & Boas: Statistical studies in Cichorium 419

TABLE 39

Flower number per head (a) for one-year, two-year, and three-year-old

PLANTS

Plants

One year old

Two years old

Three years old

Increase in
a from i

2 years

Increase in
a from 2
years to
3 years

Average
a

Num
ber of

cases

Average

a

Num
ber of

cases

Average

a

Num
ber of
cases

Fi's

18.8

36

20.6

55

21.0

105

-f 0.8

+ 0.4

(A X £22)

19. 1

9

20.7

10

20.9

35

-f 1.6

+ 0.2

(£22 X ^ )

19.7

5

20.4

6

20.0

II

+ 0.7

- 0.4

(E3 X A)

18.6

6

21. 1

8

19.8

12

+ 0.5

- 0.3

(A X £3)

18.4

3

20.3

6

I9.I

9

+ 1.9

— 0.8

(C X -E22)

20.1

4

21.5

5

21.4

16

+ 1.4

— 0.1

(£3 X C)

18.4

4

20.4

6

19.4

7

+ 2.0

— I.O

19-3

2

21.0

8

20.5

8

+ 1.7

- 0.5

(A X C)

18.1

3

19.8

6

20.0

7

+ 1.7

+ 0.2

F2's

18.5

63

19.4

23

+ 0.9

(A X £22)-4-. . . .

20.5

10

20.0

3

- 0.5

(£22 X A)-io-

Ser. I

17.6

II

18.2

10

+ 0.6

(£22 X A)-io-

Ser. II

17.1

15

(£3 X A)-4-Ser.

I

19.7

10

19-5

10

— 0.2

(£3 X A)-4-Ser.

II

19.2

17

increase of the second year over that of the first is just twice that
of the third year over the second year. The F2 generation has
been studied only for the first and second year of growth, but has
shown practically the same increase as did the Ft.

In regard to the behavior of the rate of decrease in successive
years, it has been pointed out earlier in the paper that the rate of
decrease becomes less as the blooming season becomes longer.
Older plants are more vigorous than one-year-old plants and
bloom longer. Because of thi's fact the rate of decrease becomes
less as the plant grows older. The question whether various lines
of descent are exhibiting diverse types of behavior, which are,
however, characteristic of particular lines, will be more fully
considered later.

6. Variation in range and distribution of modal

NUMBERS

A survey of the facts regarding the maxima or modal numbers
in chicory are of interest in relation to the question of specificity
of flower number.

420 Semi-centennial of Torrey Botanical Club

From data taken on an average of every third day throughout
the period of bloom, it often appears, as may be illustrated in
TABLE 10, that a well-pronounced maximum or mode is in evidence.
In this particular instance there are 86 heads with 19 flowers per
head, and about this mode there is quite a regular chance distri-
bution. In other cases, the distribution is less regular and a mode
is less pronounced. In no case, however, has there appeared evi-
dence of a strongly bimodal distribution. Of course, data on all
heads blooming on a plant would perhaps bring out the modal
number more prominently. The collection of data during only a
part of the blooming period would, of course, give quite different
modes for any plant which exhibits a rate of change.

The range and distribution of the modal values for various
groups of plants are given in table 40. Several points are clearly
indicated. The range of individual variability of modal numbers
is rather wide and extends from 14 to 22. Interannual partial
variability exists, as shown in the first section of the table, which
presents data of a single generation in the first, second, and third
years of growth. The distribution is somewhat irregular and the
number of plants observed is not as large as one might wish, but the
evidence is quite clear that the mode for a plant actually shifts to
higher values at least during the first three years of growth.

table 40

Distribution of modes for flower number per head for individuals of

VARIOUS generations AND LINES OF DESCENT

Plants

Fi First year . . .
Second year. .
Third 3^ear. . .

F3 and F4

(A X £22)-4-.
(A X E22)-9~.
{En X A)-io-
Summary. . . .

i?'s 1916.

Distribution of modes

14 15 16

9
2
I

41
10
51

13

31
31

62

5
41
84
130

17

10
3
7

20

14

19 20 21 22

2
26
30

Number
of plants

38
52
63

37
118

143
298

79

Average
mode

18.4
I9.I
19-5

18.8
16.0
16.8
16.8

17.4

The second section of the table presents data for the F3 and
F4 generations of three main lines of descent in the first year of

Stout & Boas: Statistical studies in Cichorium 421

bloom. Here family differences are to be noted : highest values
are seen in family (A X £22)— 4— and lowest values are seen in
family (A X E22) — P — • It will be seen by comparison with table
48 that these modes are high or low quite in agreement with values
of a.

The modes for the 1916 crop of the variety red-leaved Treviso
exhibit a rather wide range of variability, and the most frequent
class is not strongly indicated but appears to be at 17. It will
be shown later that the values of a for these plants (table 42)
are relatively high, although the range of variability of flower
number is wide. The distribution for the whole season is some-
what skew and the low modal values are not an indication of the
seasonal performance.

No attempt has been made to mass indiscriminately the total
counts made into a general table. It is obvious that such a
treatment would cover up individual and line differences, and the
final value thus obtained would be modified by the proportion-
ate number of individuals in the different races represented.

IV. CHARACTERISTICS OF FLOWER NUMBER IN THE VARIETY
RED-LEAVED TREVISO

This variety is one of remarkable vigor of growth. Grown
from seed sown in pans in a greenhouse during January there is
vigorous and rapid growth, giving large rosette leaves of extremely
robust habit and a large much-branched stem. The mature
plants have been from 4 1/2 to 6 1/2 feet tall (see Fig. i, Stout '17).
When thus grown from seed the plants are annuals. The exceed-
ingly vigorous growth usually reaches full maturity in late summer
and early autumn and the plants are usually through blooming
before the first late-autumn freeze.

The race grown has exhibited a type of teratological develop-
ment which consists in the production of two stems sometimes
completely separated above the crown, but usually united and
somewhat twisted for a distance, above which the two merge into
a single apparently normal stem. The fasciation usually does
not extend to the flower branches and does not directly affect the
individual flower heads. The extent of fasciation, the leaf shape,
and the degree of pigmentation in this strain all show considerable

422 Semi-centennial of Torrey Botanical Club

variation, but the plants grown in all three years were quite similar
in general vigor and habit of growth, although there was more or
less variation in the number of branches and the length of the
flowering period.

Data on flower number in this variety have been obtained from
two parent plants of 1913, and from three generations (1914,
191 5, and 191 6 crops) derived by further intra varietal breeding.
Until the summer of 191 5 all plants (with the exception of one
feebly self-fertile plant in 191 5) were found to be self-sterile.
The variety was kept in culture by crossing plants (usually im-
mediate sister plants) that were cross-compatible.

As a result of the vigorous and robust growth in the lines
grown of this variety as many as 2,000 to 3,500 flower heads per
plant were produced. At the climax of development 100 or even
150 heads opened on a single plant in a single day. The period
of bloom was also somewhat extended, but data were not used
unless a plant completed its bloom before the first autumn freeze.

Table 41 presents data for one of these plants. For this
plant the period of bloom extended from July 14 to October 21.
There is a decided decrease in both daily range and average.
The partial variability seen in the entire range from 31 to 7 was
the greatest in both directions seen in any plant of this variety.

Of the crop of 191 3, grown from commercial seed, there are
data for the two plants which were crossed in obtaining seed from
which all cultures were derived. For one of these the values ob-
tained were a = 19.5, b = — 0.084, and the range was 23-12;
for the other plant a = 17.4, b = — 0.081, and the range was
19-13. With respect to values of a, these two parent plants were
somewhat diverse.

For the 1914 crop, data were obtained from nine plants: these
gave values of a ranging from 18.6 to 20.8, with average of 19.8 dr
0.67. As shown in table 42, the values of these plants are quite
well distributed about the value of the parent having highest value
of a.

In 1915, offspring weie grown from crosses involving four
sister plants (of the 191 4 crop) for which the values of a were
almost identical and were quite of the average. The average
values for each line of descent according to immediate parentage

Stout & Boas: Statistical studies in Cichorium 423

October

1 :

mm]'

1 L-£i

CM •

• M

if c i

M • Tj- M

O'C T

H M

^J CM -MM

M M

3 Li.

M M M M

M M •

M CM M M

September

CM •

M M CM M

• CM • M (N M ro

C'C T

t a I

M CM M (>1 • M

L FI

MMMMMMCMM

00

• H (M M M

• M

/S- T T

Oil

M M

CM M M

CM

OS

CM M . CM <N M

M CM fC ■ ^ •

y b I

« • <N

F a I

August j

o

ro

M M ro

I'SI

VO

M M M

Z,-9i

M M M ro CM

• M

N j

M M • M

00

CM M M M M rO

9"Z,i

IH fO fO M M

• M

M

ro <N M CO

I 81

H

M M re CM M

e-gi

M fO CM M

o'6i

M • CM •

S*6i

M CM lO M

M

I'OZ

July

o\

N

• CM M ro

8"6i

VO

o

; M ; ro lo M

; M ro o ; ;

S-6i

0

; fO N CM M CM

; ; CM CO <N <M

8^2

] M

[ M

t^oo OlO M M rOTjuoo r^oo 0\0 M CM ro'^iovo r^oo 0\0 m

MMMMMMMMMMCNCMCMCMCMMCMCMtNOJfOrO

0\
00
m o

M CD
" I
II II

0\ o

CO
CO

424 Semi-centennial of Torrey Botanical Club

were all above 20 (see table 42) and the range had shifted to
decidedly higher values than had previously appeared. One
plant gave a rather low value of a, 16.9. The complete sterility
of all but one of these plants prohibited the study of selection for
high or low values in uniparental offspring. The very prevalent
cross-incompatibility also limited the possibility of testing such
selection in biparental heredity. However, crosses were made
from which six different progenies were grown in 191 6.

table 42

Distribution of values of flower number of first day of bloom (a) for the variety red-
leaved Treviso. Data for 1913, 1914, 1915. and 1916

Series

Distribution c

f V

alues of a

3er of
duals

Average

a of

a of

0

IT)

0

0

0
00

CO

0

ON

ON

0
0

VO

0

21.0

0

N

tr,

N

0

0

1 Numl
indivi

a of
series

parents

1913

I

I

2

18.5
19.8

20.8

I

I

4

2

I

9

±0.67

±0.99
±2.15
±1.56

±1.52
±1.79

±1.37
±1.49
±1.64
±0.77

\

' 19-5

1915

I

I

I

I

2

2

I

9

. 17-4
' 19-5

R4. . .

I

I

I

I

I

I

I

I

8

20.8

19.72
19. 81

i?5

I

I

2

2

5
3

I

3

I

I

I

I

I

19
12

20.6

. 19-9
19. 81

I916

Ri

3

2

2

I

19-5
18.4

18.6

. 19-7=^
no data

i?8

I

2

3

I

2

I

I

I

4
3
2

I

17
23
9

' 19.0

i?9

I

3

5

2

3
2

2

3

I

19.8
21.73

Rvo

I

2

2

19.0
19.7
20.8

19.2
19.0

20.8"

21.72

no data

^ 21.82

Rn.

I

I

X

I

I

2

I

3
6

5
3

2

I

I

20

Rn. . ,

2

I

4

I

17

\ 22.8

A glance at the results recoreded in table 42 shows that the
ranges of values per plant and the averages for the different series
are somewhat different. The lowest average of 18.4 ± 1.79 seen
in series 8 is lower than the average of the generations grown in
the previous two years. The range is also lower. The immediate
parents of this series were of lower than average rank, but the
offspring average still lower. The highest average (20.8) seen in
series 12 is the highest realized in any of the 1916 series and is

Stout & Boas: Statistical studies in Cichorium 425

quite of the average of the previous year. The parents of this
series were plants of high values. While the series showed some
regression, the variability of d=o. 77 was low and the average
value of a was higher than that of other series from parents of
lower value. The results obtained in these two lines suggest that
in this variety lines of descent may differ and that such differences
may be high or low according to the performance of the ancestry.

table 43

Distribution of values for rate of decrease (6) for the variety red-leaved Treviso.
Data for 1913, 1914, 1915, and 1916

Serie

Distribution of values of d

'1 '

8, 8

O I O

d T3

d of parents

Average
^of

1913

1914.
I915

-Rl,2,3 .
i?4...

i?5.
I916

Rs.

R9.
Rio ■
Rn.
R12

3 I

I

3

I

I

■0.084
■0.081

I —0.072
\ -0.081I
I -0.0862
\ —0.007
j -0.0862
[ -0.081I

no data
-0.059
-0.083

— 0.0923

— 0.088
-0.083
-0.087"
-0.0932
no data
-0.087"

— 0.124

— 0.082
-0.075

— 0.087
-0.093

— O.IIO

-0.057
-0.051
-0.075
-0.058
—0.071
-0.086

As shown in table 43 the values of b were more variable than
the values of a. The characteristic performance gave minus
values for the rate of change and there was on the whole less
variability than was seen in the hybrid generations reported
above; here there were no cases of + values, and only one plant in
the 0.000 class.

In this variety the branching of laterals was very complete,
as shown in plate 13, A and B, and there were, as a rule, rather
few heads in clusters such as are shown in plate 13, C and D.

426 Semi-centennial of Torrey Botanical Club

This habit and the abundant branching which result in the pro-
duction of such large numbers of heads presumably give oppor-
tunity for the development of heads of such ultimate laterals
as might be suppressed in a less robust growth, as is seen in dwarf
races. There should be opportunity to observe, in such a race,
the lowest possible range of flower number for the species. The
chance for development of partial variability.is most favorable. The
observations indicate that this is the case. In table 44 are shown
the ranges in number per head for all plants studied in this variety.
The highest flower number per head extends from 31 to 19 and
there were 40 plants out of 144 that had produced some heads
with 25 or more flowers.

table 44

Range of flower number per head for all plants of the variety red-leaved

Treviso

Number of

Maximum flower

Minimum flower number

Dift'erence between

plants

number

Range

Average

maximum and

average minimum

I

31

7.0

24.0

I

30

13.0

17.0

3

28

I5-II

13.0

15.0

7

27

13-10

II.4

15-6

II

26

12-7

10. 1

i.S-9

17

25

16-9

12.0

13-0

26

24

14-5

10.7

13.3

27

23

17-6

II. 2

II. 8

26

22

14-7

10.9

II. I

17

21

14-9

II. I

9.9

6

20

13-10

II-3

8.7

2

19

13-11

12.0

7.0

Furthermore, the low numbers range as low as 5 and in every
class grouped together for a particular high number some plants
produced one or more heads of eleven or twelve. The plant with
the highest number of flowers per head (31) produced heads ranging
as low as 7. The complete record of data for this plant is given
in table 41.

This variety is characterized by rather wide extremes of
partial variability and by high values of a and b.

Stout & Boas: Statistical studies in Cichorium 427

V. CHARACTERISTICS OF FLOWER NUMBER IN A HYBRID GEN-
ERATION OF A CROSS BETWEEN PLANTS OF RED-LEAVED
TREVISO AND A WILD PLANT (A)

The plants of this hybrid generation were all of vigorous vege-
tative growth. They were scarcely as tall as typical plants of
the red-leaved Treviso, but they were more abundantly branched
from the base of the main stem. The total number of flower heads
averaged about the same as for the red-leaved Treviso.

Data were collected from 50 plants of this generation. For
these the range of observed numbers per head extended from
28 to 10 (table 45). This range is less both for highest and
lowest number than in the red-leaved Treviso. The plant
with highest number, 28, also gave numbers as low as ii,
thus exhibiting in its partial variability almost the entire range
observed. For highest numbers alone the individual variability
ranged from 28 to 19; for lowest numbers it ranged from 15 to 10.
Greater differences exist in the highest numbers than in the lowest,
quite as has been noted for other cultures of chicory.

table 45

Range of flower number per head of plants of the Fi generation of the
CROSS between red-leaved Treviso and a wild plant (A)

Number of

Maximum flower

Minimum flower number

Difference between

maximum and
average minimum

plants

number

Range

Average

I

28

II. 0

17.0

2

25

15-10

12.5

12.5

2

24

15-12

13-5

10.5

7

23

14-12

12.7

10.3

10

22

1 5-1 1

12.7

9-3

12

21

14-II

12.3

8.7

15

20

14-10

II-3

8.7

I

19

II. 0

80

The values a and h have been computed for 12 plants of this
hybrid generation and these are given in table 46. These are
sister plants having for seed parent a plant of red-leaved Treviso
{R ser I. no. 6) whose performance gave a value of 21.5 for a and
—0.091 for h. These values are quite the average for the plants
of this race grown in 191 6 (see table 42).

As shown in table 46, the values of a for these 12 plants ranged
from 16.6 to 21.4, with the average at 1 9.1. All 12 plants gave
minus values for b.

428 Semi-centennial of Torrey Botanical Club

TABLE 46

Values for twelve Fi hybrids of a cross between a plant of the red-
leaved Treviso and a wild plant (A)

Pedigree

[0]

[f]

[ot]

a

b ■

i

RA ser. 3 — no. 19

15.2

48

- 21.58

701.00

16.6

- 0.030

91

" 13

17.0

48

- 15-05

740.21

18.0

— 0.020

91

" 21

17.0

47

- 23.75

781.87

18.4

— 0.029

94

" 15

17-5

52

- 17-89

932.42

18.5

— 0.019

97

" 9

i6.e

50

- 34.10

690.03

18.8

— 0.048

90

" 24

17.7

42

- 22.87

611.89

19.2

- 0.036

83

" 10

18. 1

52

- 25-35

936.72

19.3

— 0.027

99

" 8

17.4

48

- 34-19

856.32

19-3

— 0.040

97

2

19.2

46

— 11.26

675-65

20.0

— 0.017

87

" 22

17.6

48

- 40.51

432.25

20.2

- 0.055

91

" 16

18.2

53

- 39.36

926.86

20.4

— 0.041

103

" 18

17.9

56

- 67.04

1072.15

21.4

— 0.062

no

VL SELECTION AND HEREDITY: CHARACTERISTICS OF DIF-
FERENT PROGENIES AND VARIOUS LINES OF DESCENT DE-
RIVED FROM A CROSS BETWEEN A WILD PLANT (A) AND
PLANTS OF THE VARIETY BARBE DE CAPUCIN

It would seem that the computed value for the first date of

bloom (a) which is derived from a considerable amount of data

obtained during the entire period ot growth is the most adequate

single value obtainable that is sufficiently characteristic to serve

as a basis for judgment of the effects of selection and of heredity.

The value of the rate of change (b) can also be used as a value for

comparisons.

I. General comparison of progeny with immediate parents

AS TO values of a AND b

Whether flower number is directly inherited as such or in-
directly inherited through its relation to types of vegetative habit
of growth, some clue to the type or degree of heredity may be
gained by the comparison of the performance of offspring with
the values of the immediate parents.

With the appearance of a few self-fertile plants in the Fi
generation, it was possible to grow self-fertilized lines of descent
of uniparental lineage and data from such lines admit of ^ com-
parison of performance. On account of the self-sterility of a large
number of the various progenies (about 50 per cent.), it was not
possible to make extreme + and — selections.

The Fi generation of the cross wild white-flowered X Barbe de

Stout & Boas: Statistical studies in Cichorium 429

Capucin was of biparental descent. The summaries for compari-
son with the immediate parents are given in table 47. The
values of a for the Fi are from data collected in the third year of
growth and are averaged for all sister plants of a particular cross.

table 47

Average flower number for first day of bloom (a) for parents and their

Fi OFFSPRING

Parents

Fi offspring

Difference between

a

a

Average a

Number
of plants

Average a

parents and off-
spring

C, 21.7
C, 21.7
A, 19.2
A, 19.2
A, 19.2

E22, 23.3
£3, 20.7
£22, 23.3

C, 21.7
£3, 20.7

22.5
21.2
21.2
20.4
19.9

19
15

46

7
21

21.4
20.0
20.1
20.0
I9.S

- I.I

- 1.2

- I.I

- 0.4

- 0.4

It is rather striking that while in every case the average of the
offspring is below that of the average for the two parents, yet the
higher the average of the parents the higher is the average of the
offspring. The fact that the average of the offspring is lower than
that of the two parents suggests that the immediate parents gave
values that are high for the races involved and that the offspring
show regression. It will be seen that the difference between the
average flower number per head of parent and offspring is the
greater the higher the value of the parents. Accordingly, if one
used parents with very low flower number, in all probability the
offspring would give a somewhat higher average flower number
than their parents.

Further and more exact comparison of progeny with immediate
parent can be made in the three main lines of uniparental descent
constituting the F2, F3, and F4 generations. Table 48 presents
such data, indicating the pedigree, the distribution for values of a,
the variability measured by the standard deviation, the value of a
for the immediate parent, and the standard deviation of the parent
series (the series of sister plants to which the particular parent
belonged).

A glance down the columns of the frequency distribution shows
that the three lines of descent present, on the whole, values that
are rather uniform for each line. A comparison of the values of

430

Semi-centennial of Torrey Botanical Club

a for a series with that of the immediate parent shows a marked
agreement. Frequently, inside of a Hne, the offspring deviate
somewhat from the values of the line, or regress toward the average

Parent

cr of
series

o q\q\o M q moooooo moooooo roosoo o\ioioioo\m
woowMrqMdddMddoMddddddddM

-H-H-H4^-+^4^^^-H4^-H4^-H4^-fl4^-H-H^^4^-}^4^-H^^-H

q\0-^'^i>.| 1 roc^lOloOoo^>■oo^-0^oo^>roMMr^^^
0\i-imo\o| Io^^ooo^d^^•o6^--o6^~^^-■o6cx5^-•l>•d^

Offspring

b

o o^a^MMC^Hooo^qloooooooc^oc^o^^l^lJ^OMoo
M oowmohomom 6 dddddd odd oomo
4^ -H-H-H-H-H+I-H4^4^-H-H-H-H4J-H-H-H4^-H-H-H-H-H-H

Average
a

0\ *O0'^q\l>(N'^(Nu-)r^vO0000\r0f0Ov00\f000'rfMTj-

0\ o o ooo dNt^oo li-jvooooo r^oo r--o6 t^-t^t^vdoooo

jaq
-lun^

0<NO'O0\r0Tl-a\0'^r0'^O\iO'ONOiOO0\r000J>l>

Frequency distribution for values of a

o'£z

S'zz

O'ZZ

S-iz

O'lZ

M H (N M M M

£-oz

O'OZ

M

M

cs ! ro ! M

M H M

M

S-6i

M ! ro M

eq

ro M

o-6i

M H M (N H M M

ro M N ro

S-gr

M IH

roM rororoioroc^vO m m ro

o-gi

<N

! N

■^^MN^rqioooio '. '.

S-^i

ro

M ^ ^

Tj- M rt O O 0\ M !

o-Zi

c^ro*-"MMO<Nio2<^

S-gx

; ; '. ] O ro M

o-gi

r^MroMroi-i(Nir5i-(

S-Si

ro ro M M M

N

o-Si

! ! <N ; ! '. !

ro \0 \0

ION I I f\l

1 (M I

I I I I I I
fy^ rrj rr^ Or^ ^

tq ^ i:^

I I

o o o o o o

H H H H N N N
I I I I I I I

and are hence not in close agreement with the actual value of the
parent. There are, however, numerous cases of almost exact
agreement.

It is also to be noted that there is a decided reduction in vari-

Stout & Boas: Statistical studies in Cichorium 431

ability of various series of F2, F3, and F4 generations over that
of parent series of the Fi. For the latter the standard deviation
is ±2.07; for the former the values are mostly less than ±1.00.

^= 1

0

1 °

0
M

10

00

0

0\
0

M

00

0

0 00
M (N
0000

ro 0\ 0\ 00 01
N M ro ro ro

0 0 0 0 0 q 0

rj- 0 0
oj ro M
0000

On
ro

q

1 0

1

6

1

6

1

d
1

6
1

1

1

d
1

6 6 6 6
MM

6 6 6 6 6 6 6
M M M 1

6 6 6 6
MM

d
1

be
c

^ 10 \0

^ 0\ 0\

000

M

m
0

N

00
0

0

0

0\0 0 M
0 CO

0000

(N 0\ 0 On 0 M
M M ro ro »0 ro
0000000

CO »0 M IT)

M M 01

0000

0
10
0

flfspri

0 0

1 1

0

1

1

0
1

0
1

0
1

1

0
1

0000
MM

0000000

M M M 1

0000
MM

0
1

0

jaq
-uin^

^ 0

(N

0

0\

0\

0

(N IH M M M

On r/) 00

H M

i>

o6i'o—

ogi-o-

.....>-(..,................

oZi-o—

. . ..................... .

091 'O —

' M

M

M

oSi-o—

O'l'I'O —

"0

o£i-o —

C/l

<U
3

OZI'O —

>

oii-o—

OOI'O —

c
0

o6o'o—

! ! ; ; ! <^^ !

01

a

OgO"0 —

M

ro

.1-1. .

i-i ! ! ! i-i ! ."

-5

oZo-o—

M

H

N '. ! (N !

>^

090*0—

M

0\ i M 01 M ro !

! (N ; 01

s

1)

oSo-o—

M l-l

10 M i M M (N ro

M 01 01 00

3
w

0^0"0 —

M

0\ M uo ro ; On

ro 10 01

o£o'o—

M IN 0 M

ro On 0 c^j ro

oq M

0S0"0 —

! 00 00 M ! M

vO ro :

oio-o—

^ : M :

ro ro ro !

OOO'O

; 0 (N ; ; ;

01 M 01

0I0"0+

; ^ M : : :

01 M M

0Z0'0+

I I tNi fn I

U-) IT) H N NO

I

Oo l-l N 1^ N N H

c 1 I'l'T T I I r r r rrrroooooooooo

^ I 1 I I I I 1 I I I I I I 1 I I I 1 1 1 I I I I

I T I

^ fc^
Xc|x

[ii [Ih

The variability of sister plants of a series decreases with line breed-
ing.

The performances of these lines of descent and pedigreed
series with respect to rate of change (b) are given in table 49.

432 Semi-centennial of Torrey Botanical Club

The frequency distribution is given for classes of o.oio extent.
Here, as for values of flower number for the first day of bloom (a)
the lines show some noticeable characteristic differences; for line
descending from the Fi plant {A X £22) no. 4 the — values were
most in evidence. In the other two lines a considerable number
of plants exhibited + values. The heredity of the type of seasonal
change is in general indicated by line performances. In respect
to immediate parentage, the agreement is less close than for values
of flower number. There has been no attempt rigidly to select
for various values of rate of change. That the performance is
more sporadic for rate of change than for flower number will be
further indicated in the study of various races.

2. Detailed analysis of six races as to inheritance of

FLOWER number PER HEAD

In the F3 and F4 generations of the family with the original
parentage wild white-flowered (A) X Barbe de Capucin (£22),
it became evident that the pedigreed breeding in lines had resulted
in the development of several types, races, or elementary species
which differed from each other most decidedly in habit of growth.
This gave opportunity for the study of the flower number in rather
widely diverse races. Comparisons can be made with reference
to the relation of various habits of growth to performance in
flower production. The detailed records of lines of parentage
also give opportunity for further analysis of the influence of selec-
tion and the degree to which heredity is in evidence.

A. The history and characteristics of a semi-dwarf, sparsely
branched race {race i, or line {A X -E22) — 4-J — )• In mature
development for the first year of growth this race was character-
ized by low stature (i 1/2 to 2 1/2 feet tall) and by a sparse and
coarse branching habit. The branches were very brittle, and
there was a very general death of the branches due to suscepti-
bility to a bacterial or fungous disease. The death of the tips of
the branches added considerably to the dwarf-like appearance and
decreased the total number of flower heads that came to bloom.
Twelve plants of the F3 and thirty-one plants of the F4 of this race
were quite uniform in the general vegetative habit of growth.
In the rosette stages, however, the plants which later made smallest

Stout & Boas: Statistical studies in Cichorium 433

growth were decidedly dwarfed. The general habit of this race
is well shown in plate ii, which also shows the effects of the tip-
rot.

table 50

Statistical constants (discussed pages 380-385) for two generations (F3 and
F4) OF a semi-dwarf, sparsely branched race (race i) with those for the

immediate ancestry

Pedigree

[0]

[ol]

[.2]

a

5

i

Fl. {A X £22)

no. 4

18.5

21

— 10.22

150-33

19.9

— 0.066

41

F2. (A X E22)-4-

no. 9

17.6

26

-13-18

296.63

18.7

— 0.044

58

" 6

18.3

10

- 5-56

50.00

19.4

— 0.109

22

(U

" 7

17.9

18

-17.72

145.92

20.1

— O.121

35

0
C

" 8

18.2

14

-10.52

71.40

20.3

-0.153

27

<

"4

17.9

19

— 19.82

155-54

20.4

— 0.127

41

I

18. 1

40

-36.47

538.60

20.9

-0.068

77

2

19.6

20

-13-81

181.87

21.0

-0.073

41

" 5

19.8

28

-12.39

289.55

21.2

-0.056

55

" 10

19.7

21

-15.89

196-53

21.4

-0.083

42

" 3

19-3

19

-18.45

148.85

21.6

— 0.120

37

F3. (AX En)-4-3-

no. 6

17.0

23

-10.32

205.13

18.2

-0.052

47

" 9

17.0

18

-19-54

141-36

18.6

— 0.142

38

" 10

17-5

33

-17-95

297-95

19-5

— 0.060

63

" 8

18.4

IS

— 10.40

105.69

19.8

— 0.096

33

" 7

17.9

14

— 12.90

78.78

20.2

-0.163

31

I

19.8

14

- 1.91

92.00

20.2

-0.031

35

" 2

19.1

15

- 7-75

98.82

20.3

— 0.077

37

" 12

18.3

33

-29-45

400.21

20.7

-0.074

68

" 3

18.9

14

-11.23

82.29

20.7

-0.132

34

" 4

19.1

24

-18.47

235-71

21.0

— 0.078

55

0

" 5

18.7

15

-18.45

116.38

21. 1

-0.158

40

03

" II

18.6

33

-34.12

403-09

21.4

-0.085

71

Fi. (A X E22)-4-3-ii-

no. 18

17.8

25

- 8.40

269.89

18.6

-0.031

60

" 15

16.8

25

— 26.26

337-00

18.8

— 0.079

58

" 3

18.0

34

— 24.22

498.08

19.7

— 0.049

75

" 20

18.8

31

-13-72

432.56

19.8

-0.032

69

" 25

18.2

25

-24-23

298.39

20.2

— 0.081

59

" 41

19.7

33

— 12.72

372.00

20.8

-0.034

65

" 21

19-3

29

-18.49

318.25

21.0

-0.057

56

" 9

20.2

28

- 9-29

288.10

21.2

-0.034

55

" 8

19.4

16

-15-55

98.62

21.9

-0.156

33

" 24

19.4

16

-15-38

97-75

21.9

-0.156

31

A most characteristic feature of flower number per head in this
race, as seen in F3 and F4, is its high value. The average number
(table 50) per head [0] is high, which indicates that heads with
high flower number were strongly in evidence. There was a
pronounced tendency to produce some heads with extra high
numbers. In the F3 series {A X £22)-4-J- of 12 plants there

434 Semi-centennial of Torrey Botanical Club

were six plants, having some heads of 24 or more flowers (see
TABLE 51). One plant (no. 12) produced a head of 33 flowers,
other plants had heads of 27, 26, and 25 flowers. For the F4 crop
of 31 plants data were collected from 10 plants that were least
afi'ected by the tip-rot. These, likewise, showed a tendency to
produce a few heads with unusually high numbers per head: seven
of the ten plants had some heads with 24 or more flowers each.
For no plant was the highest number less than 21. The minimum
number in any head for these plants, however, was quite as low
as that ordinarily observed. In this race irregular and sporadic
variations in partial intra-annual variability occurred, giving a
few heads with high numbers. The lowest values realized in a
plant during a season were, however, quite or almost as low as
were seen in series having no such high numbers.

TABLE 51

Range of variability of flower number per head of one-year-old plants
of the f3 and f4 generations of race i

Maximum

Minimum flower number

Diff"erence between

Number of plants

maximum and
average minimum

flower number

Range

Average

Fs. (A X E,2)-4-3-

I

33

12.0

22.0

I

27

13.0

14.0

I

26

12.0

14.0

I

25

15.0

lO.O

2

24

I3-II

12.0

12.0

3

23

15-10

II. 7

II-3

2

22

T4.0

8.0

I

21

13.0

8.0

Fi. {A X E22)-4-3-ii-

I

26

12.0

14.0

3

25

15-II

13-3

II. 7

3

24

16-13

14-3

9.7

2

23

15-II

13.0

10. 0

I

22

12.0

10. 0

The general tendency in this race, however, is toward high num-
bers per head. The values for [o] and for a computed from all the
data are consistently high. In respect to seasonal change, all plants
of this race exhibited a marked decrease, as may be seen in the
values of h given in table 50. The performance with respect to
terminals and laterals was quite as is most typical of the species.

The record of pedigree for the two series (F3 and F4) of this

Stout & Boas: Statistical studies in Cichorium 435

race shows that the descent has been consistently from the plants
having highest values of a. For the parent of the F4 series (which
was {A X £22) —4-3 — no. 11), the value of a was 21.4, which was
the highest of the series. The parent in the F2 was plant (A X
£22)— 4 — wo. j; the value of a for it was 21.6, also the highest of
the series. The value of a for the Fi plant used as a parent was
19.9. The values of a for the two original parents was as follows:
for the wild white-flowered plant A, 19.2 in 1913 and 19.3 in 1914;
for the plant of Barbe de Capucin, £22, 23.0 in 1913 and 23.6 in
1 91 4 (these values are of slightly higher relative value as the
plants are older than one year).

The race is, therefore, one of high average values, the average
of a being above 20.0. Selection for high values of a has main-
tained a high average. The race has shown a strong tendency to
irregular sporadic partial variability in that high numbers of
flowers are frequently developed in a few heads.

B. The history and characteristics of a dwarf, sparsely branched
race {race 2 or line {A E22) — 4-6 — ) . At the extreme right
foreground of plate ii are shown several characteristic plants
of this very decidedly dwarfed race. In general habit of growth
and susceptibility to the tip-rot this race is quite like the semi-
dwarf just described. It is, however, of much smaller stature and
is less branched. Nine plants of the F3 and twenty of the F4 have
been quite uniform in general appearance when mature. Sta-
tistical data on flower number were taken from six plants of the
F3 and from nine plants of the F4, these plants being the ones least
injured by the tip-rot.

In agreement with the extremely dwarf habit, the total number
of heads produced by these plants is low, the average for the nine
plants studied of the F4 being 95 for the entire season of growth.
This dwarf race is a sister race of race i, noted above, in that the
two descended from the same Fi plant {A X -E22) no. 4* The
number of heads open in any one day was so low that on only one
date were 10 heads open in a single day on any of the 1916 crop.
The period of bloom was somewhat shortened.

The highest flower number for any head was 22 and the lowest
was 12. There was no tendency to the production of extremely
high numbers per head for a few heads, as seen in the semi-dwarf

436 Semi-centennial of Torrey Botanical Club

race ; the values of [o] and a average slightly lower (tables 52 and
48).

All plants of the race showed a seasonal decrease { — b) with
the characteristic relation of laterals and terminals.

The ancestry of this race diverged from that of the semi-dwarf
race in that the F2 parent selected was a different plant (no. 6)
having somewhat lower values of a (19.4) and of [0] (18.3). The
selection for this race has not been consistently for lowest values
of a, as the plant of the F3 used as parent (A X £22) —4-6 — no. 3)
was the one having highest values for a. The F4 has shown an
increase of average of a over that of the preceding generation.

table 52

Statistical constants for two generations (F3 and F4) of a dwarf, sparsely

BRANCHED RACE (RACE 2). ANCESTRY GIVEN IN TABLE 50

Pedigree

[0]

[t]

a

d

t

Fz. (A X E22)-4-6-

no. 4

16.4

12

- 5.26

68.53

17.4

— 0.081

29

" 8

17.1

17

- 9.89

183.26

18. 1

-0.05s

51

" 5

16.1

26

-36.07

355.68

18.7

— 0.099

68

" 10

16.7

32

-36.12

471-30

19.2

— 0.077

70

I

16.6

14

-16.50

89.29

19.2

-0.183

39

" 3

18.0

22

-31.01

255-64

20.7

— 0.I2I

48

^"4. (A X E22)-4-6-3-

no. 17

17.6

14

- 3-70

92.50

18.I

-0.039

29

I

18.1

10

- 3-26

41.57

18.9

— 0.079

19

" 16

18.2

12

- 3-40

45-88

I9.I

— 0.076

21

" 3

17.9

17

- 9.43

131-46

I9.I

— 0.071

36

2

16.8

22

-26.88

205.77

19.7

-0.134

45

" II

17.7

16

— 12.69

90.50

20.0

— O.I4I

30

" 15

17.6

17

-17.94

113-25

20.3

-0.157

34

" 21

18.2

16

-13.63

97.17

20.5

— O.I4I

32

" 23

18.6

16

-17-03

100.55

21.3

— O.171

31

The performance of this race indicates that extreme dwarfing
in chicory affects decidedly the total number of heads produced,
but has no pronounced effect in reducing the average number of
flowers per head or in changing the general character of intra-
seasonal partial variability as exhibited by minus values of h.

C. Characteristics of a race exhibiting a second period of growth
during a single season {race j or line {A X E22) —QS — The
peculiar and unusual habit of growth seen in this race has already
been described and reference made to the illustration (text-
figure i). Detailed data for three plants of this race have

Stout & Boas: Statistical studies in Cichorium 437

also been presented (tables 31, 32, and 33) and discussed as
illustrating (i) increase in flower number per head during a period
of growth and (2) the fact that values for the second period of
growth are lower than those for the first.

For judgment of the general performance of this race, there
are data for 35 sister plants and for 15 of these there are also data
from the growth of the second period, all of which are given in
TABLE 53.

For the period of first growth, it is seen that values for a range
from 15.5 to 18.3 with an average of 16.9 ± 0.62. The values
are very uniform and are decidedly low when compared with those
of the semi-dwarf race (race no. i) considered above. The most
decided variability of this race is seen in values of rate of change
in flower number; these range from —0.048 to +0.018. These
differences in rate of change did not seem to involve any differ-
ences in vegetative habit of growth, in the grouping of flower
heads in clusters, or in the total number of heads produced.

Reference has already been made to the general performance of
the growth of the second period and to the evidence that the values
for such growth are as a whole lower than those of the earlier
growth.

In its ancestry, this race descended from an Fi plant {{A X
£22) no. q} from which no data were obtained.

For the parent selected for the F3, {A X E22) —Q-no. 5, the
value of a was 16.5 and that of b was —0.022. The parent of the
F4, (A X E22) —9-5 — no. 12, was one of 13 sister plants whose
values for a ranged from 15.2 to 18.0, averaging 16.6 with a stand-
ard deviation of ±0.86; its a value was 17.7 (high for the genera-
tion) and its h, —0.024,

As to value of a, the line of parentage has therefore exhibited
rather medium values.

Considering the value of a for the first period of growth, the
values are quite characteristic for the line of descent as a whole
(tables 27 and 53).

D. Characteristics and history of a semi-robust brittle-stemmed
race {race 4). The plants of this race are coarsely and somewhat
sparsely branched, the branches are thick and very brittle and
bear such small leaves as to appear almost leafless. The thirty

438 Semi-centennial of Torrey Botanical Club

TABLE 53

Statistical constants for one generation (F4) of a race exhibiting a second

PERIOD OF GROWTH DURING A SINGLE SEASON (RACE 3) WITH THOSE FOR THE
IMMEDIATE ANCESTRY. FOR FIFTEEN PLANTS OF RACE 3 CONSTANTS FOR THE
SECOND PERIOD OF BLOOM ARE INDICATED IN ITALICS AND ARE GIVEN DIRECTLV
BELOW THE VALUES FOR THE FIRST PERIOD OF BLOOM

Pedigree

[0]
—

[^]

[ot]

[^2]

a

b

t

F2. (A X £22)-p-

no. 5

iS-9

26

- 6.88

327-38

16.5

— 0.022

58

3

15-9

50

-32.47

992. C7

16.6

-0.033

113

I

17. 1

54

3i'6i

I 14^*28

i 0 .u

— 0.027

115

Fs. (A X E2i)-Q~3-

no. II

14.4

30

- 8.50

318.00

15.2

— 0.027

60

" 3

14.6

32

- 9-90

412.00

15-5

— 0.027

70

u

" 5

15-5

40

- 5.28

625.42

15-8

— 0.008

84

icest

" 9

15.1

49

-14.78

935-97

15-8

-0.015

107

" 7

iS-7

31

— 11.20

364-38

16.6

— 0.029

65

<<

I

15-9

32

- 8.78

378.24

16.6

-0.023

66

" 4

16.0

30

— 7.22

397-15

16.6

— 0.019

75

" 13

iS-9

50

-15-47

946.32

16.7

— 0.016

108

" 10

15.6

44

-20.35

799-37

16.8

— 0.027

97

" 6

15-8

30

— 11.02

340.85

16.8

-0.033

64

2

15-3

25

-24.29

244.74

17.7

-0.097

52

" 12

16.5

51

— 24.60

981.17

17.7

— 0.024

109

" 8

16.5

37

-26.59

674.92

18.0

— 0.041

93

Fi. (A X E22)-P-J-I2-

no. 35

14.6

34

— 10.00

392.83

15-5

— 0.026

71

" 18

15-9

32

+ 3-8i

411-35

15-6

+0.OC9

69

" 9

16.0

34

+ 4-90

381.09

15.6

+0.013

64

15.2

27

- 4-19

244-83

15-7

— 0.017

51

" 24

I4-Q

19

+ 0.03

119-32

14-9

0.000

37

15-7

35

- 2.36

429.14

15-9

— 0.C06

69

" IS

13-0

22

+ 2.77

107.37

14.6

+0.020

39

15.4

33

-10.75

412.33

16.0

— 0.027

68

" 38

13-2

19

- 1-44

118.69

13-4

— 0.013

40

15.3

30

— 10.29

356.32

16.2

— 0.029

63

" 30

14.9

16

+ 2.12

87.46

14-3

+ 0.025

33

" 32

15-9

37

- 5-91

480.28

16.3

— 0.O12

75

" 16

17.0

33

+ 7-94

435-79

16.4

+0.018

77

16.5

33

+ 0.56

434.00

16.5

+0.001

65

" 17

13-2

19

- 3-12

94-32

13-8

— 0.003

37

" 21

15-9

27

- 7-21

274.21

16.6

— 0.026

52

16.2

26

- 5-41

274.67

16.7

— C.019

61

" 10

13.6

18

— 0.30

129.69

13-6

— 0.002

40

2

16.7

39

- 0.44

448.50

16.7

— 0.001

75

" 22

16.6

28

- 2.98

335-83

16.8

— 0.008

63

16.8

27

+ 0.23

259-05

16.8

+0.001

53

" 33

I4-Q

19

+ 0.03

101.83

14.9

+ 0.007

35

" 12

16,7

26

-13-33

277.80

16.9

— 0.048

55

16.7

26

+ 2.79

271-55

16.9

+ C.OO9

54

" 6

14.9

19

+ 1.64

142.33

13-I

+ 0.007

42

" 37

17.3

35

+ 4-32

435-13

16.9

+ 0.010

72

" 19

15.9

38

-18.22

610.59

17.0

— 0.029

83

16.1

29

- 7-90

246.40

17.0

-0.032

57

" 31

r3-i

19

- 0.34

123.00

13-2

— 0.003

40

Stout & Boas: Statistical studies in Cichorium 439

TABLE sz— Continued

Pedigree

i^J

r/21

20

i6.8

— 9 7/1

A A ft m
44U. /y

T 7 0

— 0.006

72

16.9

'?7

~t~ 1.27

482.38

17 0

+ 0.003

TK
1 0

3

-'JO

18

+ /.i(5

1 24.00

1 5-1

-j-0,070

d2

16.8

— 7. 27

363.86

17 1

— 0.009

62

1

J ? 4
J- j'H-

19

— I -74

1 33-40

75.(5

— 0.013

40

16.6

1 '2
00

— 8 7<

ACtK. T 7

T 7 7

— 0.022

61

-'JO

TO

-4- 2 27

128.28

J -y

— O.OIQ

5

16.8

26

— ^7

277 80

17 1

— 0.019

j4

17 2

— I 00

Ac;o 27

17 ^

— 0.004

" T /I

I'i 8

I Q

— 4 (57

7 ? 0
J -y

— 0.006

33

Rac€

17 =^

4- 0. fJd

409.71

17.4

-|-0.002

7 1

4

T 7 T
i. / . i.

7 7

— "? t;7

COT /iS

JO •■- '40

17/1
J. / .4

— 0.007

76

16.9

34

— 6.12

410.44

17.4

— 0.015

68

I

-TP

+ 1.92

134-24

7(5.0

+ 0.074

41

16.7

27

- 6.95

261.05

17.4

— 0.026

55

" 34

15-4

19

+ 0.08

127-83

15-4

— 0.007

40

" 13

17.0

35

- 9-91

548.54

17.6

— 0.018

74

" 23

16.9

35

-10.58

437-52

17.7

— 0.024

69

" 26

17.2

37

- 4-98

471.08

17.8

-0.015

74

" 29

17.0

39

— 12.02

539-00

17.9

— 0.022

79

" 8

17.7

36

- 7-48

463.08

18.3

— 0.016

72

mature plants of the F4 series ranged from 24 to 36 inches in height.
This series of plants are shown in plates 10 and 11, field no. 53.
The habit of growth differs from that of the first growth of race 3
in being less abundantly branched ; although maturing early there
was no tendency to the development of a new and second period
of growth except in one plant.

The values for twenty-nine plants of this race are given in
TABLE 54. Values of a range from 15.2 to 18.2 and average
16.8 with a standard deviation of d=o.82.

The pedigree of this race is almost identical with that of race 3.
The immediate parents are two sister plants which differed only
slightly in value of a.

The rather uniform values obtained are in harmony with the
uniformity in habit of growth and general vigor seen in this series.
The low values which characterize the series are quite identical
with those of the immediate parent and constitute an excellent
illustration of the general results that the offspring of parents with
low flower values also tend to give low values.

E. Characteristics and history of a semi-dwarf bushy race
(race 5). This race is decidedly different from any other that
has been isolated thus far. The rather small stature and very

440 Semi-centennial of Torrey Botanical Club

bushy habit of growth are well shown in plate id, which is from a
photograph of a row of sister plants of the F3 generation. The
numerous branches are slender and tough. The mature height
of these plants in the first year of growth ranged from 26 to 32
inches and in the second year (12 plants living) none exceeded 36
inches in height.

TABLE 54

Statistical constants for one generation (F4) comprising two series of a

SEMI-ROBUST, BRITTLE-STEMMED RACE (RACE 4). CONSTANTS FOR IMME-

DIATE ANCESTRY GIVEN IN TABLE 53

Pedigree

L^J

\ot\

v- J

3

r 4.

(A V 77nn"\ n e T
\A A JLL2i)—Q—J—I—

no. I

I '^.2
J.

— 4.47

"^00. K2

iq.8

— O.OII

66

3

JO

371,17

— 0.014

64

2

I"? 6

70

oV

— T I I

488.31

16 c;

— 0.023

72

4

•^6

0^

— 13.02

4'?l 28

16.6

— 0 ClIA

70

17 .

r 4.

(^A X iL2i)—Q—3~0—

no. I

14.5

14

— 5-28

106.17

15.2

— 0.051

0 /

23

IK 1

2'i

— 5.66

198.82

I K.A

— 0.029

47

2

lA 0

0^

— 8.19

371.41

— 0,022

6^

3

I 0

20

— 6 4<

167.33

— 0 0'?0

41

14 0

20

— 8.48

16.0

— 0 QK<

40

" 6

I <.2
J- j.^

2A

— 8.21

230.94

16.0

— 0 O"?

40

" 29

14.9

32

-14.88

371.17

I6.I

—0.039

64

" 12

lS-2

29

-11.30

305.15

16.2

-0.033

58

" 31

15-2

24

-10.78

216.41

16.4

-0.052

49

" 22

15.6

32

-10.57

408.91

16.4

-0.025

67

" 16

15-4

26

-10.56

243.67

16.5

-0.043

SI

" 8

I5-I

28

-13.56

261.74

16.5

-0.051

55

" 19

15-7

27

-10.49

287.75

16.7

-0.037

57

" 21

15.8

24

- 9.72

236.94

16.8

— 0.041

49

" 9

iS-5

33

-16.61

411.96

16.8

-0.039

68

" 17

15-9

27

— 10.76

271.68

17.0

-0.039

55

" 24

I5-I

39

-24.19

481.92

I7.I

-0.050

69

" 27

15-9

35

-13.59

398.92

17. 1

-0.034

65

" 26

15-8

29

-17.74

408.77

17. 1

—0.044

69

" II

iS-5

33

— 21.12

403.22

17.2

-0.052

65

" 4

15.8

28

-23.06

439.67

17-3

-0.052

68

" 13

16.3

27

-11.87

289.10

17.4

— 0.040

57

" 5

15.6

26

-19.69

271-95

17.5

-0.073

55

" 28

16.3

39

-13.15

440.88

17.5

-0.030

70

" 20

16.0

37

-16.31

386.27

17.6

—0.042

61

" 10

15.9

36

-25.88

495.46

17.8

-0.052

72

" 7

16.4

26

-15-93

157.50

18. 1

-0.065

51

" 15

16.8

27

-15.23

287.75

18.2

-0.053

56

" 18

16.8

21

-II. 81

181.31

18.2

-0.065

43

Full data for the performance of the series (£22 X ^) — JO-Jj —
are given in table 55. The values of a range from 17.1 to 19.1
with an average of 17.9 with a standard deviation of d=o.59.
For all plants the rate of change was a minus value. The rather

Stout & Boas: Statistical studies in Cichorium 441

limited individual variability is quite in accord with the very-
marked uniformity of vegetative habit. It may be noted that
the immediate seed parent of the F3 series gave a value of a at
17.7, which was slightly above the average of the series to which
it belonged. In fact the F2 series (£22 X A) —10— showed a
tendency to low values, the range dropping to 15.3 (see table 48).
The values of the F3 therefore average somewhat higher and there
is less variability among individuals.

Three plants of this F3 series were selected as seed parents for
the F4 generation. The plants of all three series were quite uni-
form in the vegetative habit of the race, as will be seen in plate
10, to the left, to the foreground, and to the right of field no. 49.
The series to the right, however, was late in maturing and the
photograph does not show the mature branching.

The three F4 series of this race were derived from sister plants
differing imperceptibly in vegetative habit. In regard to the values
of a, however, one parent (no. 13) gave the lowest value (17.1)
of the Fa; the other two exhibited values slightly above the average.

The F4 offspring of the parent selected for lowest value among
the F3 of this line gave a range for values of a of 15. i to 17.7 with
an average of 16.4 and a standard deviation of ±0.66, The
mode has shifted to lower values than were seen in any generation
of ancestry. The range, however, was not extended to values
lower than some reahzed in the F2 (table 48).

The characteristics of the other two series were quite identical.
The range of one is from 16.6 to 19.3, the average 17.8, and stand-
ard deviation ± 0.56, while values for the other range from 16.5
to 18.4, with an average 17.3 and standard deviation ±0.51.
These values agree quite closely with the average values seen in
the F3 from which the series descended. The immediate parents
were only slightly higher.

The results here obtained demonstrate that pedigreed line
breeding from different parents of a race which itself appears very
uniform in the F3 may isolate strains that are slightly different in
performance, the selection for lowest value especially giving a
strain in which the range is decidedly shifted.

F. Characteristics and history of a tall-growing race (race 6).
This race was isolated or segregated in the F3 generation. In

442 Semi-centennial of Torre y Botanical Club

TABLE 55

Statistical constants for two generations (Fa and F4) comprising four

SERIES OF A semi-dwarf, BUSHY RACE (RACE 5) WITH THOSE FOR LINE OF
immediate PARENTAGE. VALUES OF a AND h FOR ANCESTRY GIVEN IN TABLES
48 AND 49

P d' r
e igree

L'J

r/21
J

I

TT. f Ti^c V A 0 Trro r\\r\

17.9

41

22 86

542.12

19.7

—0.043

77

J7 2" ^-^22 /\ ^ J J- U

no. 13

17.0

22

~ 5-34

100.76

17.7

— 0.034

44

no. 13

15-9

40

21-75

713-00

17. 1

0.030

97

4

0"

— 27.41

17-3

0.023

1 19

I

tA t
iO.i

54

""23.74

i ouu. 7 i

17-3

— 0.023

1 1 0

1 1

15-9

37

— 23.07

544.28

17.4

— 0.041

'78
7°

7

16.7

27

— 10.03

328.61

17-5

—0.031

Ao
03

" 8

16. 1

4i)

22. 2

759-41

17-5

0.030

92

10

16.2

47

— 27'00

933-57

17.0

— 0.029

n8
90

3

16.7

50

— 23.24

1 1 16.00

17.0

— 0.020

115

9

e T
0 J-

20.54

929.18

17.0

0.031

105

16.6

43

25.02

725-50

18. 1

— 0.034

8n

" 16

16.8

44

-JO Co

23-03

750-45

18.2

C.032

90

5

17-4

34

~i3-49

493-52

t8 1
i.0.3

— 0.027

/o

0

17. 1

45

— 23.72

693-63

T C r-
18.5

— 0.035

08
90

17

10. I

54

— 50.70

1 160.97

10.5

— 0.044

113

2

16.7

47

—37.10

795-68

10. 0 ■

— 0.046

97

15

17.4

40

—33-52

950.12

19. 1

— 0.035

lOI

X*4. (,£-22 A /± J— 10— I J— 12—

no. 20

17.0

25

"T 4-23

239-37

10. 0

+0.017

51

19

17.0

28

1 J- -4^

299 6 2

iu.9

_j_Q_QQ

t;6

25

16.7

30

— 3.0O

397-13

17.0

— 0 009

A8
00

1 0

16.9

30

— 0.24

300.70

17.0

— 0.002

61

15

16.9

27

— 3.12

282.00

17.2

O.OII

e A
b4

23

16.7

2 1

— 7.24

240.65

17-3

— 0.029

49

13

17-3

27

— 0.59

299.29

17.4

— 0.002

0

17.0

— 2.91

172.69

17.4

_ ■ '

43

1 2

17.2

22

— 5-77

477.78

17-5

'-'0

10

16.7

30

— 13.29

521.08

17.0

0 026

73

17

lo.O

10

+ 2.40

106.75

17.0

+0.023

35

" 2

10. cS

32

— 11.83

oAA
300.00

T T 8
17.0

— 0.032

A/i

04

0

±0.0

30

— 10.81

333-69

TT 8

17.0

— 0.033

"3

14

17. 1

31

— 9-51

355-05

17.9

—0.026

"3

5

17.4

30

— O.Oi

338-73

t8 r\
i o.u

0.019

62

3

17.4

27

— 5-93

279.30

10. 0

— 0.02 1

57

" I

16.9

26

— II. 89

28A <

18.0

— 0.042

56

" 16

17-3

24

- 7-43

217-67

I8.I

-0.034

49

" 7

16.8

24

-14-34

247.11

18.2

-0.058

51

" II

16.8

22

-12.65

187.75

18.3

— 0.067

44

" 22

17.6

23

- 9-04

215.06

18.6

— 0.042

48

" 4

17.8

21

- 8.55

162.27

19-0

-0.052

41

" 21

16.8

27

- 9-73

107.60

19-3

— 0.091

55

F4. (£22 X A)~io~i3-ij-

no. 12

15-2

34

-f- 1-31

438.25

15-I

"1-0.003

69

" 36

15-4

32

— 1.69

398-61

15.5

— 0.004

64

" 27

iS-4

35

- 2.39

433-63

15-6

-0.005

69

" 15

15-7

24

+ 0.97

264.00

15-6

+0.003

50

" 26

15-6

35

— 1.62

441-13

15-7

—0.004

70

" 33

ie.5

24

+ 5-72

234.06

15-9

+0.025

50

Stout & Boas: Statistical studies in Cichorium 443

TABLE 55 — Continued

Pedigree

\J\

\ot\

1* J

5

T C 7

0^

— 7 QQ

361.00

16.0

— 0.023

62

T 6

12

— 6.20

360.50

16. 1

— 0.014

61

9

15-5

29

16. 1

— 0.020

64

22

15-7

32

— c 87

/1 28 n c

16. 1

— 0.014

68

30

15.0

1 1

6b

— 16.60

16. 1

— 0.029

76

34

iU.U

34

4- c /16

"2 8n ^ n

16. 1

-|-o.oi4

66

1 1

T c 6

31

— 7 76

7^6 78

16 1

— 0.023

64

19

16.0

0^

— 6.04

-2/1^ ir-2

16. S

XW. J

—0.018

61

35

■loo

— 12.68

AA1 67
44 / .u /

16

— 0.028

67

17

15-9

30

— 70?

/ion 11

16

— 0.020

67

" 16

16. 1

27

— 6.29

278.00

16.7

—0.023

71

2 1

T C t

33

418 A"?
4x0. 4j

16 7

xw. /

— 0.0'?7

70

" 28

— 16.28

603.04

16.8

— 0.027

82

" 18

16. 1

06

— T 9 T C

C/18 20
o4""-'^

16.8

— 0.022

71

/

"2 T

— 12.92

406.85

16.8

— 0.032

65

" 8

T C
15.0

33

42 1 .09

16.9

— 0 01Q

'-'-'Joy

68

4

15-9

37

16.40

502 .00

T 7 T

— n cti 1

77

20

16.2

31

9o°

oo4o /

T T T

J. /. i .

— 0.029

62

24

iOO

23

— 6.62

z L 1 .zy

T 7 '7

— 0.029

48
40

" 6

r c R

— 20.22

/ITT C C

T 7
/ -O

— 0.049

71

T r c

39

C C C A A

T 7 "2
■■^ /-O

76

16.4

0 J

— t6 2 ^

420.00

T 7 7

— 0 Old

67

r 4. \tL22 /\ ^ j—iu—ij—j-

48
40

no. 41

10.9

23

1 3-97

211 2

T(S c:

~t~o.o 19

23

TiS c

1U.5

31

U.

347-09

16.6

— 0.002

60

" 16

16.6

^ 1

-4- n CO
r ^oy

310.86

16.6

-(-0.001

56

30

10.5

— 2 8c:

16.8

— 0.010

o4

5

10.5

24

— 4.22

22867

— 0.018

0*-'

16.6

^ 1

— '2 "J 2

2 1;8 2 T

T 7 n
X / .u

— 0.013

o"

24

17-0

n '7r>

u. /y

'?^S2 2 T

T T T
X /. X

— 0.005

00

39

16.4

26

8 n-2

■28 c 7n

1 /.I

— 0.028

OT'

17.4

27

\ ^o4

317-73

T T 2

-(-0.007

■^8
0"

27

10. J

25

— 9.66

306.74

17.2

=;6

6

17.0

28

— 4.24

31 1.48

17.4

— 0.014

55

" 13

17.0

30

- 5.95

321.76

17-5

— 0.017

62

" 12

17.2

25

- 3.81

326.67

17-5

— 0.012

58

" II

17.6

31

+ 5-07

348.04

17-5

+ 0.015

64

" 8

16.8

31

- 5-75

272.32

17-5

— 0.021

60

" 37

17-3

31

- 3-26

371-17

17.6

— 0.009

54

" 22

17.0

21

- 6.38

188.35

17-7

-0.034

43

" 33

17.2

28

— 11.20

336-28

18.2

-0.034

58

" 31

17-3

25

-11-54

266.21

18.4

-0.043

54

this race there is a vigorous development of the main axis, which
reaches a height of from 3 3/4 to 5 feet in the first year of growth
with rather weak development of lateral branches; laterals from
near the base of the main axis are wanting and there is feeble de-
velopment of secondary laterals from such laterals as develop.
The main axis is very leafy with leaves of robust growth and of
gradual transition to rosette leaves. When mature the plants
of this race appear as shown in plate 10 (field label 49) . Through-

444 Semi-centennial of Torrey Botanical Club

out their growth they are decidedly in contrast to the other races
already described.

table 56

Statistical constants for two generations (F3 and F4) comprising three

SERIES OF A TALL-GROWING RACE (rACE 6), WITH THE VALUES FOR THE F2 PARENT

Pedigree

L' J

h

F2. (£22 X A)-io--

no. 8

17. 1

40

- 6.56

344.28

17.7

— 0.019

67

F3. (£22 X A)-io~8-

no. 4

16.8

38

— 9.40

535.58

17.4

— 0.017

76

" 13

17.3

39

— 5.46

561.36

17.7

— O.OI I

78

" IS

16.4

44

— 23.31

758.37

17.8

— 0.032

91

" 14

16.6

35

-18.56

491.63

17.9

— 0.038

75

" 12

16.3

34

— 25.00

518.30

18.0

— 0.049

78

" 8

17. 1

18

— 7.06

129.31

18. 1

— 0.053

37

" 3

17.4

31

— 13.78

518.83

18. 1

— 0.027

74

" 6

17. 1

21

— 7.12

124.47

18.3

— 0.057

52

" 7

17.5

28

— 13.67

365.16

18.5

— 0.037

72

" II

17.8

25

- 9.66

226.33

18.9

— 0.043

50

" I

18.2

39

— 17.17

619.32

19.3

— 0.028

83

" 5

18.5

28

— 11.72

320.26

19.5

— 0.036

60

F4. (£22 X A)-io-8-i3-

no. II

15,8

30

— 16.47

317.63

17.4

— 0.052

62

17.0

27

— 6.70

301.75

17.6

— 0.022

61

" I

17.0

34

— 11.75

440.32

17.9

— 0.027

59

" 16

17. 1

34

— 10.54

391.55

18.0

— 0.026

61

" 5

17.6

21

— 4.31

199.53

18. 1

— 0.022

48

" 2

18. 1

16

— 9.28

989.33

18.2

— 0.009

32

" 17

17. 1

21

— 12.81

215.50

18.3

— 0.059

51

" 14

17.0

28

— 14.70

297.15

18.4

— 0.049

56

8

17. 1

24

— 15.67

238.72

18.7

— 0.066

49

" 6

17.4

29

— 16.02

333.09

18.8

— 0.047

.61

" i<
^ J

17. 0

2A

— 18.4=;

217.24

19.0

— 0.085

47

" Q

17. 1

— 24.90

416.33

19. 1

— 0.060

67

" 12

17. 1

— 16.15

177.72

19.3

— 0.091

49

" 7

17.7

33

— 24.10

414.04

19.6

— 0.059

52

F4. (£22 X A)-io-8-i4-

no. 9

15.7

29

— 2.35

310.05

15.9

— 0.008

58

I

15-7

31

- 4-38

372.48

16.I

— 0.012

63

" 18

15-7

34

— 12.66

394.87

16.8

-0.032

68

" 5

16.2

29

- 8.95

256.57

17.2

-0.034

57

" 6

15-4

36

-23.36

482.00

17.2

— 0.049

73

" 8

16.5

30

- 8.69

322.41

17-3

— 0.027

58

" 16

16.9

31

- 4-24

433.37

17-3

— O.OIO

70

2

17.1

27

- 5-24

278.50

17.6

— 0.020

54

" 7

15-6

33

-24-39

399.46

17.6

— 0.062

66

" 10

16.0

32

-22.83

397.21

17.8

-0.057

67

" 3

16.9

33

-10.53

382.29

17.8

— 0.026

65

" 12

15-9

32

-25.17

381.58

18.0

-0.065

64

" II

16.4

27

-25.97

391-91

18.2

— 0.067

63

" 15

17.4

29

-10.39

310.05

18.4

-0.033

58

" 4

17-3

29

-13.95

307.71

18.6

— 0.046

60

' 13

17.7

30

-26.93

349.57

19.9

-0.073

61

For the F3 series of twelve plants (table 56) values of a ranged
from 17.4 to 19.5 and averaged 18.3 with a standard deviation of

Stout & Boas: Statistical studies in Cichorium 445

d=o.62. The further performance of this race was observed in
two series of F4, the progeny of two plants, with values of a which
were somewhat lower than the average of the series. In regard
to the actual range for values of a and averages, one series was
decidedly higher, more uniform and less variable; its progeny
digressed toward the higher average of the generation to which
the parent belonged. For the other series ( — 10-8-14 — ), the
average a was somewhat lower than the value of the immediate
parent, and the range was extended to lower values than were
seen in the parent series. All plants of the F3 and F4 of this race
gave minus values for rate of change.

The two radically different races, semi-dwarf bush (race 5)
and the tall-growing race (Race 6), descended from the same Fi
parent (£22 X A) no. 10.

The six races noted above all descended from three sister Fi
plants which were quite alike in general vegetative characters.
The three main lines of descent split up into six rather widely dif-
fering races. In regard to values of a, it may be noted (table 48)
that races that segregated from the same line were of the same
general performance. Values for races i and 2 were relatively
high, those for races 3 and 4 were relatively low and those of
5 and 6 were more intermediate.

The differences in values for the lines and the noticeable
uniformity within races, especially in the F4, is proof that some-
what slight differences in number per head are hereditary. Al-
though self-sterility of many plants prevented a rigid testing of
the effects of selection, there is decided evidence, as noted above,
that selection for high or low values within a race is in some degree
effective.

A further point appears to be clear. The very marked vege-
tative differences may affect very strongly the total number of
heads produced, but such vegetative differences are only slightly
concerned with changes in the performance as measured by such
values as 0 and a.

446 Semi-centennial of Torrey Botanical Club

DISCUSSION AND CONCLUSION

The aim of the earher investigators who made statistical studies
of number of flowers per head was to obtain exact descriptions of
species as such. By determining the number that is characteristic
of various species, the degree of specific differentiation in respect
to the character of flower number would appear. The results
of the various statistical studies (of Ludwig especially) of flower
number, although chiefly directed to ray-flowers, have been con-
sidered as indicative of specific differentiation, as noted in the
review of literature. That the number of rays is not closely
or rigidly stable is very evident, however. A rather wide range
of variability is nearly always in evidence, and this has led investi-
gators to judge a species by some such value as the average or
modal number. When thus judged, it seems clear that species
differ and that a species may tend to maintain a particular maxi-
mum. The studies with chicory, viewed from this standpoint,
indicate that the total number of flowers per head may range in
this species from 5 to 34 and that the number per head most fre-
quently produced is somewhere from 17 to 19.

The view held by Ludwig maintains that the numbers for
ray-flowers characteristic for species as such conform to such a
series as that of Fibonacci, hence there is proof of discontinuous
variation or mutation in the evolution of species. However,
the data are not fully in accord with this view. It appears that
there are some maxima that fall on certain primary numbers of
the Fibonacci series and such cases do suggest that fundamental
rhythmic processes of development may have occurred giving spe-
cific series of numbers for such organs as rays in flower heads. That
all flower-number differences seen among and within species thus
arose is not indicated, for there are evidently numerous cases of
maxima that do not fall on either primary or well-recognized
secondary numbers of such series. It will readily be recognized
that the indiscriminate statistical study in populations of such a
variable character as number of flowers per head centers the atten-
tion on general or average performance and fails to recognize
the effects of individual and partial variability of a character.
Furthermore, a study of individual and partial variabilities

Stout & Boas: Statistical studies in Cichorium 447

furnishes valuable clues as to whether the origin or evolution of
different flower numbers per head is of continuous or discon-
tinuous nature.

When the data for an individual plant of chicory are massed
there is most often a rather pronounced mode or maximum. A
certain number of flowers per head on a plant occurs with greatest
frequency. The modal number, however, is not the same for the
various individuals of a mixed population or even of a race: for
some individuals it falls on as low a number as 14; for others it
falls as high as 22. In relatively few cases do the maxima fall
on one of the primary numbers of the Fibonacci series and the
grand maximum certainly does not thus fall.

A consideration of variation within the individual (partial
variability) shows that the number of flowers per head in chicory
may range on a single plant from 7 to 31. The primary numbers
8, 13, and 21 of the Fibonacci series are thus all represented on a
single plant and the range extends almost to the next in order, 34.
Furthermore, as a rule, all numbers between the extremes are
represented on an individual. The variations within the individ-
ual are certainly continuous rather than discontinuous, at least
to the extent that there is no rhythmic discontinuity with jumps
to one after another of a series of maxima.

The questions of chance variation, differentiation, and sym-
metry deserve special mention. In chicory, it is clear that the
differences in number per head are not purely chance variations
that are due to undiscoverable factors. As fully noted in the
literature review above, nearly all the statistical studies, both in
methods of collection and in treatment of data, have considered
the variations in flower number to be purely chance; but the re-
sults obtained in chicory suggest that there may be present
sources of variation that are due to differentiation in the sense
used by Pearson. The data here reported show that position is a
factor influencing flower number. As terminals bloom before
their immediate laterals, differences according to position appear
in the course of the succession of bloom, giving intraseasonal
partial variability. In this sense the more lateral and later-
blooming heads are differentiated from the more terminal and
earlier-blooming heads of the same plant.

448 Semi-centennial of Torrey Botanical Club

However, the distinction between chance variation and dis-
coverable differences according to position (differentiation) in
such similar organs as flower heads in chicory is by no means clear.
Partial, chance, or fluctuating variations are of such intergrading
ranges that differentiation according to position would not be dis-
covered by such indiscriminate and chance methods of study as
have usually been employed. Considering the first terminals and
the last laterals which bloom on a plant, there is, as a rule, a very
marked difference. There is also a difference between terminals
of various ranks. But for certain terminals the number of flowers
per head is identical with that of certain laterals ; one class grades
into the other. A strict interpretation of differentiation, however,
would perhaps lead one to the view that even the variation seen
in any one day among closely homologous terminals or laterals
produced on a plant is significant of still finer grades of differen-
tiation.

Such a view would evidently recognize a symmetry of develop-
ment giving primary, secondary, and tertiary branches. If such
be the case, it is clear that the transition from one to the other is
more continuous than discontinuous; at least the gradations are
slight. Although differentiation is in evidence, and is readily
discoverable, it is not obviously discontinuous.

In chicory the repeated branching and deliquescent habit of
growth with terminal and lateral branches of various ranks and
series give opportunity for a full expression and development of
marked differentiation of the sort discussed above. If the dif-
ferentiation which does occur in chicory in development of number
of flowers in such homologous organs as flower heads is any measure
of differentiation between individuals, races, or species as such,
it is indicative of continuity rather than discontinuity.

In judging an individual as a whole for flower number per head,
the collection of data at the beginning of the blooming season will
give a different estimate from what would be obtained at the close
of the blooming season. As a rule, higher values prevail in the
first part of the blooming period. The variability observed will
be less the shorter the period of time covered. There will be less
variability with the limitation of data to terminals only, to laterals
only, or to terminals or laterals opening during one day.

Stout & Boas: Statistical studies in Cichorium 449

The collection of data from a number of individuals to be
compared in any sort of statistical study may therefore involve
several decided elements of error. The collection of the same
amount of data is not alone adequate, for the two sets may repre-
sent different relative periods of bloom. Even if data were taken
for a period of bloom, as for the first ten days, they would most
often represent different proportions of the entire period of bloom.
Data from the last period of bloom only would give an unreal and
apparent similarity between plants, as all plants are more nearly
alike in the lower numbers per head produced during the late
season. Data for the first period of bloom would emphasize the
differences in the higher numbers per head. Indiscriminate col-
lection of data from individual plants therefore involves so many
sources of misrepresentation as to be of little use in formulating
any conclusions. It is quite possible that much the same condition
exists in respect to the hereditary studies of such characters as the
size of flowers (as Goodspeed and Clausen, '15, have suggested)
or as the size and weight of fruit produced or of any other character
which is subject to such degrees of partial variability.

The data for chicory indicate that there is much individual
variability in such a character as number of flowers per head. Such
differences as total number of flower heads and length of blooming
period are quite closely correlated with vegetative vigor and
variations of this sort are seen even in closely inbred and very
uniform races. Much more fundamental differences exist espe-
cially between plants of different races or between races as such.
The modal number and the computed value of the first day of
bloom may be quite different. The rate of change as determined
by values of h may be decidedly different, even to the point of
exhibiting no seasonal decrease or even of revealing an actual
increase in number as the season advances. While such cases are
not numerous they are sufficiently frequent (for this the data are
fully convincing) to indicate that the usual processes of noticeable
differentiation according to position which result in seasonal de-
crease may not be in evidence or may even be reversed in their
operation so that the differences between terminals and laterals
as to number per head do not appear and hence both terminals
and laterals show much the same range of variation. These

450 Semi-centennial of Torrey Botanical Club

individual differences arise among plants of the same age that
have been grown under as similar conditions as is possible under
greenhouse and garden culture.

The individual variations observed are to be considered as
fluctuating and continuous. They indicate that the character of
flower number is constantly varying, giving differences upon
which selection may operate in the isolation of races.

The performance of races with marked differences in vegetative
habit of growth gives some clue to the extent to which the char-
acter of flower number per head can be modified in correlation
with different habits of growth. The various races range in habit
from extremely robust and much branched races with annual pro-
duction of large numbers of flower heads to small dwarf races which
produce only a few flower heads in a seaison. The total number
of heads produced and the length of the blooming period are greatly
modified in such cases. The character of number per head, how-
ever, remains more constant. In the most vigorous types we have
found, as a rule, high values for the computed number of first day
of bloom. Also, it is in the most vigorous races that partial
variability is greatest in degree, for here lowest numbers per head
(as low as 5) are found.

The possibility of isolating races which will exhibit character-
istic differences (although often slight) in flower number is well
demonstrated, though rigid selection is here difficult because of
the limitations imposed by self -incompatible plants. The con-
tinued selection in self-fertilized lines of descent for extreme and
unusual characteristics appearing within a race is impossible in
chicory because of self-sterility and the limits of the effects of
selection cannot be so rigorously tested as is possible in a species
fully self-compatible.

Hereditary variations of slight differences and of continuous
gradation appear in every group of organisms with which ex-
tended studies of heredity have been made. The two current in-
terpretations of such phenomena are (i) those of the strict adherents
of the genotype theory who see only recombinations of Mendelian
units and (2) those of the selectionists who see evidence of con-
tinuous alterations in constitution upon which selection may
operate.

Stout & Boas: Statistical studies in Cichorium 451

The experiments of Castle and Phillips ('14) show that selec-
tion for variations in color patterns of rats led to a gradual increase
or decrease in the amount of pigmentation, ultimately giving quite
diverse patterns. These results have been continuously and con-
sistently interpreted by Castle to indicate that "genetic factors
are themselves variable" and that such factors may be a»ltered
gradually but permanently by repeated selection and ''that one
must reject the conception of modifying factors and conclude that
the character has a high degree of genetic stability yet is subject
to continuous genetic fluctuation" (Castle and Phillips, '14;
Castle, 'i6a, 'i6b, '17, and other papers).

Much the same sort of variability has been demonstrated,
however, in progenies propagated asexually by Stout ('15) in
Coleus and Jennings ('16) in Difflugia. Here the fundamental
hereditary characters are shown to be subject to slight and heredi-
tary alterations.

Jennings ('17) points out that modifying factors, postulated
by the opponents of the doctrine of fluctuating change in heredi-
tary constitution to account for a series of graded variations in
sexually reproduced and cross-bred organisms, are themselves
alterations.

Johannsen ('03) claims that in respect to such a variable
character as weight and size of seeds an ordinary population of
beans consists of a number of genotypes (races) which can be dis-
covered by growing and comparing the progeny of single seeds.
Each progeny constituting what he calls a pure line was con-
sidered as breeding true except for occasional mutations. Johann-
sen thus sought to apply the doctrines of discontinuous mutation
then so recently announced by de Vries and to establish the geno-
type theory to account for the isolation of races differing in heredi-
tary constitution (genotype). The variations appearing within a
line of progeny were considered as purely due to environment
(phenotypic).

There are at least two methods of attacking the problem as to
the inheritance of variations, such as those of weight of seeds
within a line of progeny:

I. By comparing the seed weights of progeny of small and large
beans irrespective of immediate parentage within the line and
/ rrespective of a consideration of individual or partial variability.

452 Semi-centennial of Torrey Botanical Club

2. By comparing the progeny of different plants which vary
in the average of their total product, with the estimate of individual
variabilities which takes into account the partial variabilities that
are present.

Either method is good if properly carried out. The first
method, however, is less satisfactory in that it does not consider
partial variability. The extremes of partial variations may include,
as they do in chicory, the very lowest and the highest values of
range. It is, however, the first-named method that Johannsen
chose to use. Furthermore, the far-reaching conclusions published
by Johannsen ('03) were based on the results of one crop for
which the offspring of small and large seeds produced by the same
plant were grown and compared as to size and weight of seeds
which they produced as classes. As far as any line was concerned,
the selection constituted a test for the heredity of partial vari-
ations only, as Belling ('12) has already noted. The negative
results obtained indicate that the weight of individual seeds
produced in the same pod or in different pods on the same plant,
as was the case in each of the 19 lines, is largely due to differences
in nutrition (phenotypic), resulting from position in the pod or
to relative position of the pod. It seems clear that adequate tests
for the heredity of variations in the character of weight of seeds
in beans should be based on the performance of plants as wholes
rather than on the weight of the individual seeds.

Johannsen (^13) reports progenies beyond his first generation from
a known parent plant (his 1902 crop) from only two of the original
nineteen plants. These two were plants of near the average
rather than the extreme performance. In these two lines of pro-
genies, however, the performance of individual plants was not
considered and as selection was on the basis of the weight of indi-
vidual seeds, those selected for high and low weights may have
repeatedly had the same parentage.

Even with this method, however, there is evidence that heredi-
tary variations of fluctuating nature arose within a line. In
line I, the weight range of beans produced by the parent plant
(1901) extended from 550 to 750 milligrams. The next crop of
seeds (produced by at least several sister plants) ranged from 350
to 900. For several years later the average weights of the heavier

Stout & Boas: Statistical studies in Cichorium 453

and the lighter seeds planted were beyond the limits of the entire
range of the seeds of the original parent plant. The actual ranges
of the successive crops, which might show multimodal irregularity,
are not given; only the average weights of all above or below the
mean are presented. Here, however, the data show that the
range of variability in offspring of a single plant may far exceed
that of an original parent. While this increased variability is
interpreted by Johannsen as within the range of variability of the
genotype such results are not at all inconsistent with the view that
actual changes in hereditary constitution are in evidence which
when subjected to such careful selection as that employed by
Castle and Phillips may result in the isolation of further genotypes
within the "pure line."

The necessity of considering the performance of individuals
as units in judging a progeny, or a line of progenies, and of repre-
senting on this basis the individual variations rather than the
partial variations occurring within the individual, when such
fluctuating characters as weight of bean seeds and number of
flowers per head in composites are involved, is well shown in the
studies of chicory reported above. The greatest diflference be-
tween individuals and between races as such (in respect to flower
number) is seen in the production of higher numbers per head rather
than of lower numbers. The tendency of the variabilities is to
extend the range of flowers per head to higher numbers keeping
the range in lower numbers much the same. There is not a de-
cided shifting of the entire range to higher and lower values for
the plant or the race as a whole.

This characteristic variability among races, to extend or to
limit the range in high numbers rather than to shift the entire
range, indicates that racial differentiation is here quite different
from that which appears to have prevailed in the races which de
Vries ('oi) isolated in Chrysanthemum segetum. However, de
Vries confined his attention chiefly to terminal flowers and
practiced a rigid selection for extremes only. In isolating the
race with 21 rays per terminal head, various intermediates ranging
as low as 13 for a maximum were discarded. At any rate, the
isolation of two extreme races by de Vries is no proof that other
and intermediate races could not also be isolated quite as they
have been in chicory.

454 Semi-centennial of Torrey Botanical Club

The differences between varieties or races of chicory are, con-
sidered as a whole, quite continuous. The average number of flowers
per head for the first date of bloom (a) (computed from the seasonal
record of the various members) for the different races isolated in F3
andF4 ranges from 204 to 16.7. These averages, however, are from
distributions that are overlapping (table 48). If the races having
highest and lowest values of a be massed in population, the curve
will be bimodal, the irregularity of which depends largely on the
relative numbers. If the race with intermediate values be also
massed, then the curve becomes monomodal.

It is clear that in chicory flower number per head is a character
that is subject to wide and continuous variability that is (i) par-
tial (existing among the parts of a single individual and here
involving also elements of differentiation according to position) ,
(2) individual (characteristics of plants as wholes based on their
entire record), and (3) racial (fluctuating about a mode that can be
somewhat closely maintained by selection). The differentiation
between races as such, however, is no more marked than the dif-
ferentiation involving position that occurs among the various
parts of the individual. Most especially is the character of num-
ber of flowers per head subject to modification during ontogenetic
development and epigenetic processes of growth. Flower number
per head is very different, therefore, in nature from the general char-
acter of the inflorescence as a whole and from the character of flowers
as individuals. All flowers are quite alike, all are in heads, but
the number of similar flowers that are thus grouped is variable.

The operation of heredity in such a character as flower number
is seen in the isolation of races which may be maintained by such
selection as was possible in chicory. Within each race, however,
there are further variations, continuous in gradation and of the
same nature as those appearing in a more mixed population, which
are unmistakable evidences of the instability of characters and
hereditary units.

BIBLIOGRAPHY

Bateson, W. 1901. Heredity, differentiation, and other conceptions

of biology. Proc. Roy. Soc. 1901. 69: 193-205.
1903. Variation and differentiation in parts and brethren.

1-23.

Belling, J. 191 2. Note in Am. Breeders Mag. 3: 311, 312.

Stout & Boas: Statistical studies in Cichorium 455

Boas, F., & Wissler, C. 1905. Statistics of growth. Rep. U. S.

Comm. Educ. 1904: 25-132.
Burkill, I. H. 1895. On some variations in the number of stamens and

carpels. Jour. Linn. Soc. Bot. 31: 216-245.
Castle, W. E. 1916a. Is selection or mutation the more important

agency in evolution? Scientific Monthly 88: 91-98.
1916b. Variability under inbreeding and cross breeding. Am.

Nat. 50: 178-183.

1917. Piebald rats and multiple factors. Am. Nat. 51 : 102-114.

Castle, W. E., & Phillips, J. C. 1914. Piebald rats and selection.

Carnegie Inst, of Washington Pub. 195: 1-56.
Clark, C. F. 19 10. Variation and correlation in timothy. Cornell

Univ. Agr. Exp. Sta. Bull. 279: 302-350.
Cook, O. F. 1907. Aspects of kinetic evolution. Proc. Washington

Acad. Sci. 8: 197-403.
Danforth, C. H. 1908. Notes on numerical variation in the daisy.

Bot. Gaz. 46: 349-356.
Goodspeed, T. H., & Clausen, R. E. 191 5. Factors influencing

flower size in Nicotiana with special reference to questions of inheri-
tance. Am. Jour. Bot. 2: 332-373.
Haacke, W. 1896. Entwicklungsmechanische Untersuchungen. Biol.

Centralbl. 16: 481-497, 529-547.
Harris, J. A. 191 5. The value of interannual correlations. Am.

Nat. 49: 707-712.

Helguero, F. de. 1906. Variazione ed omotiposi nelle infiorescenze

di Cichorium Intybus. Biom. 5: 184-189.
Jennings, H. S. 1916. Heredity, variation and the results of selection

in the uniparental reproduction of Difflugia corona. Genetics i:

407-534-

1917- Observed changes in hereditary characters in relation to

evolution. Jour. Washington Acad. Sci. 7: 281-301.

Johannsen, W. L. 1903. Ueber Erblichkeit in Populationen und in
reinen Linien. 1-68.

1913. Elemente der exakten Erblichkeitslehre.

Klebs, G. 1906. Ueber Variationen der Bluten. Jahrb. Wiss. Bot.
42: 155-320.

Koriba, K. 1908. Variation in the ray-flowers of some Compositae.

Bot. Mag. Tokyo 22: 86-90, 121-128.
Love, H. 1911. Studies of variation in plants. Cornell Univ. Agr-

Exp. Sta. Bull. 297: 593-677.
Ludwig, F. 1887. Die Anzahl der Strahlenbluten bei Chrysanthemum

Leucanthemum und anderen Kompositen. Deut. Bot. Montaschr.

5: 52-58.

456 Semi-centennial of Torrey Botanical Club

1888. Weitere Kapitel zur mathematischen Botanik — V. Die

Zelltheilung und der gesetzmassige Aufbau der Bacillarienbander —
VI. Das Vorkommen bestimmter Zahlen bei Organen hoherer
Gewachse und das Vermehrungsgesetz des Fibonacci. Zeitschr.
math. u. naturwiss. Unterr. 19: 321-338.

■ 1895. Ueber Variationskurven und Variationsflachen der Pflan-

zen. Bot. Centralbl. 64: 1-8, 33-41, 65-72, 97-105.

1896. Weiteres iiber Fibonaccicurven. Bot. Centralbl. 68:

1-8.

1897a. Nachtragliche Bemerkungen iiber die Multipla der

Fibonaccizahlen und die Coexistenz kleiner Bewegungen bei der
Variation der Pflanzen. Bot. Centralbl. 71: 289-291.

1897b. Beitrage zur Phytarithmetik. Bot. Centralbl. 71:

257-265.

1898. Ueber Variationscurven. Bot. Centralbl. 75: 97-107,

178-183.

• 1904. Zur Biometrie von Chrysanthemum segetum. Fests-

chrift zur Feier des siebzigsten Geburtstages des Herrn Prof. Dr.
Paul Ascherson. 296-301.

MacLeod, J. 1899. Over de veranderlijkheid van het aantal rand-
bloemen en het aantal schijfbloemen bij de Korenbloem ( Centaurea
Cyanus) en over Correlatieverschijnselen. Handel, vlaamsch nat.
en geneeskundig Congres 61-72.

Miiller, O. 1883. Das Gesetz der Zelltheilungsfolge von Melosira
{Orthosira) arenaria Moore. Ber. Deut. Bot. Ges. i : 35-44.

1884. Die Zellhaut und das Gesetz der Zelltheilungsfolge von

Melosira {Orthosira Thwaites) arenaria Moore. Jahrb. Wiss. Bot.
14: 232-290.

Pearson, K. 1901. Mathematical contributions to the theory of
evolution. I X. On the principle of homotyposis and its relation
to heredity, to the variability of the individual and to that of the
race. Part I. Homotyposis in the vegetable kingdom. Phil.
Trans. A. 197: 285-379.

1902a On the fundamental conceptions of biology. Biomet-

rika i: 320-344.

1902b. Cooperative investigations on plants — I. On inheri-
tance in the Shirley poppy. Biom. 2: 56-100.

1906. Cooperative investigations on plants — HI. On inheri-
tance in the Shirley poppy. 2nd Memoir. Biom. 4: 394-426.

Reinohl, F. 1903. Die Variation im Androceum der Stellaria media
Cyr. Bot. Zeit. 61: 159-200.

Stout & Boas: Statistical studies in Cichorium 457

Schiiepp, O. 1913. Variationsstatistische Untersuchungen an Aconi-
turn Napellus. Zeitschr. Induk. Abstain, u. Vererb. 10: 242-268.

Schwendener, S. 1878. Mechanische Theorie der Blattstellungen.

1885. Ueber Scheitelwachsthum und Blattstellung. Sitzungs-

ber. K. Akad. Wiss. Berlin. 921-937.

ShuU, G. H. 1902. A quantitative study of variation in the bracts,
rays, and disk florets of Aster Shortii Hook., A. Novae- AngUae L.,
A. puniceus L. and A. prenanthoides Muhl., from Yellow Springs,
Ohio. Am. Nat. 36: 111-152.

1904- Place-constants for Aster prenanthoides. Bot. Gaz. 38:

333-375-

Stout, A. B. 191 5. The establishment of varieties in Coleus by the
selection of somatic variations. Carnegie Inst. Washington Publ.
218: 1-79.

1 91 6. Self- and cross-pollinations in Cichorium Intyhus with

reference to sterility. Mem. N. Y. Bot. Gard. 6: 333-454.
1917- Fertility in Cichorium Intyhus: the sporadic occurrence

of self-fertile plants among the progeny of self-sterile plants. Am.

Jour. Bot. 2: 375-395-
Tammes, T. 1903. Die Periodicitat morphologischer Erscheinungen

bei den Pflanzen. Verh. K. Akad. Wetensch. Amsterdam. II. 9:

1-148.

1905- On the influence of nutrition on the fluctuating variability

of some plants. Verh. K. Akad. Wetensch. Amsterdam. 1-14.

Tower, W. L. 1902. Variation in the ray-flowers of Chrysanthemum
Leucanthemum. Biom. i: 309-315.

Vogler, P. 1908. Variationsstatistische Untersuchungen an den
Dolden von Astrantia major L. Beihefte Bot. Centralbl. 24M 1-19.

• 1909-1910. Variation der Anzahl der Strahlenbluten bei einigen

Kompositen. Beihefte Bot. Centralbl. 25^: 387-396.

1910. Neue variationsstatistische Untersuchungen. Jahrb. St.

Gal. Naturwiss. Ges. 1-32.

191 1. Probleme und Resultate variationsstatistischer Unter-
suchungen. Jahrb. St. Gal. Naturwiss. Ges. 33-71.

1912. Das "Ludwigsche Gipfelgesetz" und seine Tragweite.

Flora 104: 123-128.

Vries, H. de. 1899. Ueber die Periodicitat der partiellen Variationen.
Ber. Deuts. Bot. Ges. 17: 45-51.

1901. Die Mutationstheorie. Vol. I.

Weisse, A. 1897. Die Zahl der Randbliithen an Compositenkopfchen
in ihrer Beziehung zu Blattstellung und Ernahrung. Jahrb. Wiss.
Bot. 30: 453-483.

458 Semi-centennial of Torrey Botanical Club

Yule, G. U. 1902. Variation of the number of sepals in A^iemone
nemorosa. Biometrika i : 307-309.

explanation of plates 10-13
Plate 10

Field view of chicory crop of 1916. Field label 49 designates a series of the tall-
growing race [race 6, series (£22 X A)-io-8-i 5-]. Plants of the semi-dwarf
bushy race (race 5) are shown at the right [series (£22 X A)~io-i3-5-,] in front
[series (£22 X A)-jo-i3~i3-], and at left [series (£22 X A)-io-i3-i2-] of field
label 49. Field label 53 marks a series (A X £22)-p-5-(5-. of the brittle-stemmed
race (race 4).

Plate ii

View in field of chicory, crop of 1916. To left, field label 53 marks a series of the
brittle-stemmed race shown also in plate 10. Next to the right are plants of the
semi-dwarf race [race i, series {A X E22)-4-3-ii-]. In front of label 63 are plants
of the dwarf race [race 2, series {A X £22)-4-(5-j-]. Field label 63 designates plants
of series {A X E22)~9-5-6- and label 57 marks series (A X E22)-Q-4-io-, both of
which constitute distinct races thus far unnamed.

Plate 12

Branches of chicory from near middle part of plants. A is from the wild white-
flowered plant A used as a parent of various families; B is from plant £3 of the
variety Barbe de Capucin, and C is from an Fi hybrid of £3 X A.

Heads that are solitary and on terminal or elongated stems are indicated by i.
Sessile flower heads lateral to more terminal heads are indicated by 2, Clusters of
sessile heads are shown at 3.

As shown, especially in A, the terminal head blooms before a head that is imme-
diately lateral to it. In the flower clusters the first head to bloom is the main
terminal.

The three branches illustrate some of the variations that occur in regard to the
number of solitary and grouped heads.

Plate 13

Segments of a single large branch from each of four different plants showing the
grouping of heads, the relative development of successive laterals on a branch, and
the variation which is seen in different plants (individual variability) in the character
of the clusters of fliower heads.

A is from a plant of the variety red-leaved Treviso. (R. ser. 11, no. 14). At 8
is a terminal segment with a solitary head at apex and at first node. Segments 7
to 4 inclusive show a terminal and a single lateral at each node with various stages of
elongation of the stem bearing the terminal. Segments i, 2, and 3 show further
development; the succession of bloom indicated by letters.

B is from a plant of the Fi generation of the cross between a plant of red-leaved
Treviso and the wild white-flowered plant A (plant RA, ser. 2, no. 5). Shows rather
extreme development of ultimate branches giving few sessile flower heads, a char-
acteristic quite marked in plants of red-leaved Treviso and in wild plants.

C and D are from two plants of i_the brittle-stemmed race (race 4) showing
reduction of the ultimate branches with production of clusters of sessile flower heads.
In such clusters are terminals and laterals of different ranks.

Mem. Torrey Club

Volume 17, plate 12

STOUT AND BOAS: CICHORIUM INTYBUS. Branches showing various

ARRANGEMENTS OF FLOWER HEADS

THE TRIMORPHISM AND INSECT VISITORS
OF PONTEDERIA

By Tracy E. Hazen

Barnard College, Columbia University ,

(with plates 14 AND 1 5)

The family Pontederiaceae is notable as containing the only
known heterostyled species among monocotyledonous plants
(with one possible exception*) and is further remarkable among
heterostyled plants as furnishing the only recorded examples of
distinctly zygomorphic or irregular flowers in such plants.

Fritz Miiller,! writing in 1869 from Santa Catharina in southern
Brazil, described a Pontederia which had for several years been
introduced as an ornamental plant in the colony of Blumenau, and
which increased with incredible rapidity by asexual propagation;
the species he thought to be P. crassipes, and from the fact that
the flowers showed the same relative positions of long and short
stamens and style found in the mid-length-style form of the well-
known Lythrum Salicaria, he was convinced that he was dealing
with a trimorphic species of which only the mid-styled form had
been introduced. He found another species growing wild on the
banks of the Itajahy-mirim, which presented long- and short-
styled flowers, but no mid-styled form could be found there.
From the finely toothed petal segments shown in Miiller's figures
of this second species (loc. cit. f. i-j) it almost immediately oc-
curred to me that it was the plant which I have regularly seen
labelled Piaropus azureus (Sw.) Raf., growing with the more
famous water-hyacinth at the New York Botanical Garden, and
later I discovered that it is so identified by Miiller himself (as

* It is stated by Kerner (Pflanzenleben 2: 369. 1891; Eng. Ed. 2: 374) that
flowers of Colchicum autumnale present three style lengths, but his brief description
does not indicate any corresponding difference in stamen lengths such as always
accompanies truly trimorphic species.

t Miiller, F. Ueber den Trimorphismus der Pontederien. Jen. Zeitsch.
Naturwiss. 6: 74-78. 1871.

459

460 Semi-centennial of Torrey Botanical Club

Eichhornia azurea) in a paper published eleven years after his first
report.* In this paper the first-mentioned species is positively
identified as Eichhornia crassipes [ = Piaropus crassipes (Mart.)
Britton], the water-hyacinth, which has so conspicuously exhibited
the same habit of rapid vegetative propagation in the St. John
River in Florida, and Miiller reports that during 1 881-2 he found
long-styled plants of this species, hitherto known only in the mid-
styled form. These long-styled plants he thought could have
appeared in the Blumenau region only as the illegitimate offspring
of mid-styled parents.

In the English edition of Hermann Miiller's classic handbook
on flower pollination! published in the same year as this last paper,
Pontederia (Eichornia) crassipes [sic] is described as existing in the
colony of Blumenau "in long-, mid-, and short-styled individuals."
This statement is doubtless an error on the part of the translator
and editor, for it would seem improbable that Hermann Miiller
should have had such information from his brother Fritz at that
time. This error appears to be transferred to the other species
in the Engler-Prantl treatment of the Pontederiaceae, for therej
it is stated that Eichhornia azurea has trimorphic flowers, while of
E. crassipes only a long- and a mid-styled form is known. Our
view of the misapplied character of these last two reports finds
confirmation in the carefully edited handbook on flower pollination
by Knuth§, where no later original work on these species is indi-
cated than Fritz Miiller's second paper, and where any such almost
certainly would have been mentioned if it had been published.
Further examination of the water-hyacinths would be of con-
siderable interest.

* Miiller, F. Einige Eigenthiimlichkeiten der Eichhornia crassipes. Kosmos
13: 297-300. 1883.

t Muller, H. The fertilisation of flowers through insects, 561. 1883. (Trans-
lated by D'Arcy W. Thompson.)

t Schonland, S. Pontederiaceae. Engler & Prantl, Nat. Pflanzenfam. 2*: 73.
1888.

^ Knuth, Paul. Handbuch der Blutenbiologie. 3^: 113, 114. 1904- It is per-
haps worth while to call attention to the fact that the English translation of this
work, issued in three large volumes, covers only the first two volumes of the original;
the third German volume (posthumous) devoted to extra-European plants, and
therefore most useful for American students, contains several references to American
literature which nearly escaped the attention of the present writer, owing to the
failure of the translator and editor to mention the abridged character of the English
edition.

Hazen: Trimorphism and insect visitors of Pontederia 461

Darwin* in his book on heterostyled plants, published in
July, 1877, reports a ''third species of Pontederia'' recently dis-
covered by Fritz Miiller, which had all three flower forms growing
together in the interior of Brazil. I was at first inclined to suppose
that this was another species of Piaropus, for it seems strange that
so acute an observer as Darwin should have been so bound by the
conservative English conception of these plants current at that
time, as to fail to distinguish between the large-flowered, many-
seeded forms then beginning to be segregated on the continent as
Eichhornia,\ and the small-flowered, one-seeded Pontederia; but
that this third species was a true Pontederia is attested by Fritz
Miiller himself six years later {op. tit. 297) when he also states
that it was collected near Curitibanos in the highlands. This
species may have been Pontederia rotundifolia L., or one of the
imperfectly known allied forms; details of Darwin's description
indicate that it could not have been our P. cordata L., or P. mon-
tevideensis Hort. (see note, p. 466) although the former is reported
to grow as far south as Argentina, and the name of the latter
suggests for it a South American origin.

The first mention of heterostylism in our native pickerel-weed,
Pontederia cordata, is found in a brief note by Mr. William H.
Leggettt a prominent early member of the Torrey Club, and
probably first communicated in a meeting of the Club. In
November, 1875, he reported having noted during a previous
season an appearance of di- or trimorphism in this plant, but his
somewhat inaccurate description, based upon examination of dried
flowers, led Darwin to express doubt whether the species is really
heterostyled. In August, 1877, the month following the publica-

* Darwin, Charles. The different forms of flowers on plants of the same species,
183-187. 1887.

t The very natural genus Piaropus (thick feet) was established by Rafinesque in
1837 on Pontederia azurea and P. crassipes (Flora Telluriana 2: 81, 82); nevertheless
these two species continued to be treated under the generic name Pontederia in most
of the literature for about four decades thereafter. If the principle of priority is
ever to be maintained, there can be no justification for such provincialisrn in science
as that practised by the Brussels Congress in ordering Rafinesque's well-founded
name for a small strictly American genus published in Philadelphia to be rejected
in favor of the Teutonic Eichhornia Kunth (Enum. Plant. 4: 129. 1843) published
six years later, even though Kunth's name was latterly attaining some degree of
currency among continental writers.

J Leggett, W. H. Pontederia cordata. Bull. Torrey Club 6: 62, 63. 1875.

462 Semi-centennial of Torrey Botanical Club

tion of Darwin's book, Mr. Leggett reported* that he found on
examination of growing plants that Pontederia cordata "is as truly
trimorphic as Lythrum Salicaria, or even more so." His brief
account appears to have received little notice in this country, for
only one of our manuals mentions the trimorphism, and to answer
some of the questions raised by Mr. Leggett as to the insect
visitors, the relative fertility of the three forms, and the function
of the peculiar glands which beset the flowers, was the purpose of
the investigations now to be reported.

Early in July, 191 6, while searching for another plant seen
during a previous season at Areola, a trol ey station about midway
between Hackensack and Paterson in Bergen County, New Jersey,
I was attracted by the opportunity to secure a good photograph
in natural surroundings of our pickerel-weed — a plant which does
not seem to grow extensively in the immediate vicinity of New
York. Here at Areola it was abundant in two long ditches in a
pasture, and presently the attempt to obtain a photographic
record of its numerous insect visitors became a fascinating pursuit.
I made many visits to the station during the remainder of July,
August, and September, always laden with cameras and butter-
fly-net and killing-bottle. Flowering spikes were generally
brought back from the field for laboratory study, and though all
the flowers opening on any one morning begin to fade often by
mid-afternoon if the day is sunny, later if it is cloudy or humid,
nevertheless when brought to the greenhouse, the spikes would be
furnished with freshly opened flowers for several successive morn-
ings. The flowering proceeds in general from below upwards, but
as not more than one of the three or four flowers of a single spikelet
or sessile cluster of buds is open at one time, the main portion of
the spike may be well clothed with new flowers for some days.
In this way the biological advantages of conspicuousness of the
whole inflorescence and economy in condensation of the axis and
spacing of the open flowers are maintained at the maximum degree
of efficiency.

As Mr. Leggett discovered, the species consists of three kinds
of plants, each kind bearing a flower of somewhat different form.
At one time I thought the different plants might be recognized by

* Loc. cU. 6: 170,171. 1877.

Hazeis : Trimorphism and insect visitors of Pontederia 463

differing shades of color in the flowers, as described by Fritz Miiller
for Eichhornia azurea, in which he found the short-styled flowers
regularly of a deeper blue, and the long-styled flowers of a paler,
more violet tint; but on another day all of the spikes were indis-
tinguishable in this respect. By a little observation of the position
of the anthers, however, one may easily recognize the three forms
without having recourse to dissection of the flowers (plate 14).

Fig. I. Pontederia cordata L.; flower of the long-styled form with the perianth
tube cut lengthwise and laid open to show the typical position and relative height of
all the stamens ( X 5).

In all of the three flower forms the three narrower sepal-seg-
ments and the three ovate petal-segments are united into a tube
about seven or eight millimeters in length, which, however, often
has four narrow slits in its lower part between the segments on
the anterior or lower side of the flower, so that it would seem to be
less efficient as a nectar receptacle than many tubular flowers.
These clefts are so inconspicuous that they do not appear in any
of my sketches made from fresh flowers, but are evident in
material preserved 'fa formalin (fig. i) and in pressed specimens,
from which one might suspect that they are formed in part at

464 Semi-centennial of Torrey Botanical Club

least by shrinkage. Similar clefts are found in Piaropus. The
three spreading segments of the lower lip form a convenient landing
platform for smaller insects. The three segments of the upper
lip are rather erect and form a sort of standard with a conspicuous
mark on the posterior petal-segment which has been regarded as a
nectar-guide for insects, a large double blotch of bright yellow —
rarely two separate spots as often described, at least in plants from
several regions examined by the writer. The yellow pigment is
located in a layer of cells immediately underneath the inner epider-
mis, and appears to be diffused in the cytoplasm of these cells;
they extend outward in irregular scallops on the periphery of the
spot, producing a border slightly more greenish in tint, hardly
perceptible to the eye, but noticeable in photographs made with a
color-screen too light to bring out the correct value of the main
part of the spot (plate 14). Except in the region of this yellow
blotch, the cells of both the inner and the outer epidermis show a
rather violet-blue pigment dissolved in the cell-sap, and in addition
each cell contains a conspicuous globule of a clear indigo-blue
color, consisting either of solid amorphous anthocyanin or possibly
of a tannin or protein substance impregnated with pigment.*

In the lower part of the perianth tube the blue color is
lacking and some chlorophyl may be present, but examination
with a lens reveals pink spots due to single large subepidermal
cells containing a pigment dissolved in the cell-sap, of a slightly
purplish-red tint and having an acid reaction. Similar hypo-
dermal cells are found throughout the pistils, where they also
contain red pigment at least in the short-styled form (fig. 8).
In the middle and upper regions of the perianth, where the epider-
mal cells are blue, much larger scattered hypodermal cells occur
in abundance (fig. 2) but when dissected out, these are found to
be colorless; to assume, however, that anthocyanin is absent from

* When the cells are treated with hydrochloric acid and osmic acid (employed
as a tannin test) this deep blue globule turns to a wine-red color, and the dissolved
anthocyanin may be precipitated in small globules of the same color. If copper
acetate is used, the globule turns to an emerald green color and the precipitated antho-
cyanin shows the same tint. Such a combination of dissolved and solid anthocyanin
was first described and figured in Gilia by Hildebrand in 1863, but has been little
mentioned since until 1906, when Gertz in his important work Shidier dfver Antho-
cyan reported the wide occurrence of such a condition; summarized in Miss Whel>
dale's book. The Anthocyanin Pigments of Plants 32-35. 1916.

Hazen: Trimorphism and insect visitors of Pontederia 465

them is unsafe, since Willstatter finds that in Centaurea and some
other plants purple, red, or blue anthocyanin may readily change
to a colorless isomer. These large cells were first noticed in ma-

FiGS. 2-5. 2. Outer surface of sepal-segment: showing stomata in the epi-
dermal layer; large subepidermal idioblasts shaded with broken lines, and three
raphide-sacs (r.) in the same layer; the mid-vein (m.v.) and a lateral vein (/.v.)
lying underneath; on the surface, hairs of two forms ( X 105). 3. Typical simple
glandular hairs on the lower part of the perianth tube (X 105). 4. Hairs near tip
of filament of mid-length stamen ( X 113). 5. Cross section of filament from a bud
3 mm. long; showing a raphide-sac. and several of the elongated hypodermal idio-
blasts shaded (X 113).

466 Semi-centennial of Torrey Botanical Club

terial preserved in formalin and in alcohol, by both of which their
contents are turned to a reddish-amber color.*

Both the red-pigmented and colorless cells respond to several
of the usual tests for tannins, and their contents may be similar to
those of certain idioblasts which occur in the diaphragms of the
stem, both in Pontederia and in Piaropus. Olivef thought the
substance in these idioblasts was probably a fatty oil, but RothertJ
reports that although these cells are filled with a strongly refrac-
tive, ordinarily red-brown substance, yet in autumn the substance
may be colorless, and can then be determined by the customary
reactions to be tannin. I have also found these stem idioblasts
to be colorless in plants of Piaropus azureus and Pontederia mon-
tevideensis growing in the conservatory in mid-winter, when they
likewise seem to give the tannin reactions. Nevertheless, these
microchemical tests for tannins, unsatisfactory at best, are so
complicated by possible mixtures of other substances that we feel
they are merely suggestive here. The large cells of the perianth,
described above for the first time so far as I can find, appear to
resemble the subepidermal idioblasts in the petals of Fumaria
officinalis discovered by Zopf§ and at first described by him as

* In writing this report on the pigment-containing cells, and on other peculiar
cells which have been tested with reagents, I have depended somewhat, for checking
up observations on preserved material, on plants of Pontederia montevideensis Hort.
growing in the conservatory of the New York Botanical Garden, which furnished the
only available fresh flowers; comparative studies, however, indicate that all the
structures in question are so similar in the two species as to leave no room for doubt
of the applicability of these statements to P. cordata. The plants of P. montevideensis
came from Cambridge, England, in 1901, and have been propagated vegetatively;
the origin of the species appears to be unknown, and its botanical characters are un-
described, according to Bailey. In vegetative habit it closely resembles the narrow-
leaved forms often distinguished as P. cordata lancifolia (Muhl.) Morong, but I am
confident it is distinct from that. Rothert (Bot. Zeit. 58: 96. 1900) probably had
the same plant from the Berlin Botanical Garden, under the name Eichhornia monte-
vidensis, which he says shows such complete agreement in leaf structure with P-
cordata as to lead him to suspect that it is no Eichhornia, but a Pontederia. The
spelling montevideensis has been adopted from Index Kewensis Suppl. 4: 188. 1913.
and is in harmony with the French practise in forming the name of the inhabitants
of Montevideo, though montevidensis is said to be more in accord with Latin usage.

t Olive, E. W. Contributions to the histology of the Pontederiaceae. Bot.
Gaz. 19: 183. pi. 17- f- 5. 6. 1894.

X Rothert, W. Die Krystallzellen der Pontederiaceen. Bot. Zeit. 58: 78. 1900.

§ Zopf, W. Ueber die Gerbstoff- und Anthocyanbshalter der Fumariaceen.
Bibl. Bot. i^: 20. 1886. For more literature on this subject see Solereder's Syst.
Anat. of Dicotyledons (Eng. Ed.) i: 57. 1908.

Hazen: Trimorphism and insect visitors of Pontederia 467

containing a blood-red tannin; later it transpired that Zopf had
confounded true anthocyanin receptacles with ''sac-cells" occur-
ring throughout the vegetative structures of Fumariaceae, which
he finally considered to be alkaloid-receptacles, though Heinricher,
who gave the name "sac-cells," states that their contents are a
mixture of substances, including a fatty oil.

The role of these peculiar perianth cells in Pontederia (and in
Piaropus) can only be surmised at present, whatever the nature
of their contents. In fixed and stained sections they often behave
much like mucilaginous or gummy substances, and if of such a
nature might possibly function in protecting the perianth from
danger of desiccation until after anthesis, when the upper part
promptly rolls up and soon dries, though the tube persists as an
increasingly fleshy envelope around the ovary until the seed is
mature. Even in the open flower, some of these large hypodermal
cells are often found with their thin protoplasmic layer collapsed,
and the contents apparently discharged; half of such a cell is
shown at the bottom of fig. 2. It is interesting to note that the
stamen filaments in both Pontederia and Piaropus, which have
blue anthocyanin in their epidermal cells, are well supplied with
these long, mostly subepidermal idioblasts; but in the case of
Pontederia montevideensis , though they are conspicuous in fila-
ments of flowers grown out of doors in September, they appear
to be entirely absent in filaments of mid-winter, conservatory-
grown flowers, while still persisting in the perianth of the latter.
In April, after two or three weeks of sunshine, the flowers of
the same conservatory plant have the cells sparingly developed
in the filaments and showing a pink anthocyanin color. That
a temporary suppression of such structures should occur in con-
sequence of lack of need for them is rather incredible; the sugges-
'tion, rather, presents itself, that the contents of these idioblasts,
as perhaps also the numerous raphide-sacs which are early found
in a similar position in perianth, pistils, filaments, and most
abundantly in the anthers, are after all only in the nature of
by-products of metabolism.

The outer surface of the perianth is clothed with spreading
glandular hairs (fig. 2) ; their elongated terminal cell, rich in
protoplasmic contents and sometimes binucleate, is not infre-

468 Semi-centennial of Torrey Botanical Club

quently smeared .with a secretion which behaves under reagents
much Hke certain globules visible inside the cell; the other cells
show very scant cytoplasmic contents. Many of the hairs, how-
ever, particularly toward the tip of the perianth segments, have
one cell notably distinguished from the others, often by its swollen
ellipsoid form, but always by its strongly refractive, colorless con-
tents contained in the vacuole which practically fills the cell. To
anticipate a possible suspicion that these hairs are abnormal, it
may be remarked that they also occur on the perianth of Pon-
tederia montevideensis and of Piaropus azureus. On the upper
part of the stamen filaments are hairs of apparently similar char-
acter, though consisting of only three cells, a basal cell set in the
epidermis, a terminal globular secreting cell, and between them a
barrel-shaped cell with colorless refractive contents (figs. 1,4).
This cell is perhaps slightly more resistant to reagents than the
swollen cell of the perianth hairs, but in both the presence of
tannins is indicated, though probably not associated with exactly
the same other substances that may be found in the hypodermal
idioblasts. These peculiar stamen hairs are developed early
(fig. 5) and in buds only three millimeters long the barrel-shaped
cell occasions difficulty in sectioning, much more than the hypo-
dermal cells. More thickly sprinkled over the upper part of the
filaments, and also the upper part of the long- and mid-length
styles (figs. 9, 10) are simpler hairs, consisting only of the globular
secreting cell and a basal cell. LeMaout and Decaisne* figure
the short-styled pistil as fringed on one side with numerous spread-
ing hairs, but I have always found it almost entirely devoid of
such structures.

Growing plants of the form known as Pontederia cordata lanci-
Jolia, collected 23 March 1918 near Tampa, Florida, by Professor
and Mrs. R. A. Harper, arrived in New York in good condition a
week later, with only the flowers withered. These somewhat
dried flowers, when soaked out in water, showed a blue color in
the idioblasts of the hairs on both filaments and perianth; the
subepidermal idioblasts in the upper part of the perianth also
showed a fine deep prussian blue color, though all the anthocyanin
had disappeared from the epidermal cells, except for the solid

* LeMaout et Decaisne. Traite general de Botanique. 607. 1868.

Hazen: Trimorphism and insect visitors of Pontederia 469

globule. This same change of the colorless idioblasts to blue has
also been observed in fading flowers of P. montevideensis treated
in a similar fashion, while the pink idioblasts remain unchanged
in appearance.

It would be a matter of much interest to be able to determine
the function of these three sorts of hairs. Kerner and Stahl*
would doubtless regard the perianth hairs as a protection against
undesirable creeping insect visitors and snails. If this were their
function, here it might have been more easily secured if the hairs
had been developed on the large spathe-like bract just below the
flowers. Goebelert regards the glandular and tannin-bearing
hairs abundantly present on young fern shoots as serving in a
much higher degree as a protection against desiccation, by dimin-
ishing transpiration, and by absorbing and storing water or con-
ducting it back to young tissues. Pontederia shows a strong
tendency to dry up on the slightest provocation, and the hoary
glandular covering so conspicuous all over the young buds and
even over the stem down to the point of insertion of the bract,
may well furnish a protection against excessive transpiration.
Along this line, it is also suggestive, that the hairs on the stamens
and styles are chiefly found on the parts exposed when the flower
is opened, and that they are almost entirely absent from the short-
styled pistil, which is so completely enveloped by the perianth
tube as to need no other protection. The longer stamen hairs do
also, in some cases at least, catch the pollen from the anthers and
hold it in the most advantageous position in relation to insect
visitors, but it can hardly be supposed that so specialized a form
was evolved for such a purpose. KnuthJ found that the flowers
of Sicyos angulata L. acted upon a photographic plate much more
strongly than their inconspicuous greenish-white color would lead
one to expect, and suggested that this may be due to the numerous
glands covering the flowers which possibly "act as so many mirrors
or lenses receiving and reflecting light, so that their glitter strongly

* Stahl, E. Pflanzen und Schnecken. Jen. Zeitsch, Naturwiss. 22: 557-684.
1888.

t Goebeler, E. Die Schutzvorrichtungen am Stammscheitel der Farne. Flora
69: 483-497. 1886. See also Gardiner, E., & Ito, T. On the structure of the
mucilage-secreting cells of Blechnum and Osmunda. Ann. Bot. i: 30. 1887.

t Knuth, P. Handbook of Flower Pollination, i: 87. 1906.

470

Semi-centennial of Torrey Botanical Club

affects gelatine sensitized by silver bromide, and also the optic
nerves of insects." If such a theory has any basis in fact, it
might be applied to the hairs of the stamens, and possibly also of
the styles of Pontederia.

There can be no doubt that the numerous insect visitors of the
pickerel-weed seek it for the nectar it affords. It was at first
supposed that this was secreted by the basal region of the perianth,
but on examination of sections I found that the fleshy character
of the tube is due chiefly to the presence of numerous air chambers
separated by diaphragms much as in the stem structures, and that
nothing like nectar-secreting cells can be detected there. In
sections of the ovary, however, are found conspicuous epidermal
cells lining the three slit-like cavities left by the incomplete fusion
of the carpels (figs. 6, 8) ; these and one or two layers of cells
beneath them stain deeply because of their rich protoplasmic
contents and large nuclei, and there can be no doubt that they
secrete nectar which flows freely from the open lower end of the
narrow cavities to form an accumulation in the perianth tube.
These secreting cells were indicated in the figures of Pontederia
given by LeMaout and Decaisne* and by Wilson Smith, f but their
significance was not discussed. Similar septal nectaries were
discovered in 1854 by BrongniartJ in several genera; their histo-
logical development was more exactly studied by Saunders§; and
their variety of form and phylogenetic development in many
genera of the Liliales, Scitaminales, and Bromeliaceae have been
more elaborately set forth by Schniewind-Thies. ||

Ovary sections of Pontederia show also several groups of cells
whose large nuclei and abundant protoplasmic contents present
practically the same appearance as those of the septal nectaries;
these groups of cells are found imbedded in the tissue of the an-
terior carpel, and in the solid portion of the two anterior septa
lying above the septal nectaries, which do not reach higher than

* LeMaout et Decaisne. Loc. cit.

t Smith, R. Wilson. A contribution to the Hfe history of the Pontederiaceae.
Bot. Gaz. 25: 324-337- pl- 20. f. 54. 1898.

% Brongniart, A. Memoire sur les glandes nectariferes de I'ovaire dans diverses
families de plantes monocotyledones. Ann. Sci. Nat. Bot. IV. 2: 5-23. 1854.

§ Saunders, E. R. On the structure and function of the septal glands in Knip-
hojia. Ann. Bot. 5: 11-25. 1890.

II Schniewind-Thies, J. Beitrage zur Kentniss der Septalnectarien. 1897.

Hazen: Trimorphism and insect visitors of Pontederia 471

the middle of the ovary (figs. 7, 8). Often even in a young ovary
these cells abut on small schizogenous cavities, as at c' in FIG.7,
and in older sections of Pontederia montevideensis , where they are
more extensively developed, the cavities may form elongated

Figs. 6-10. 6. Section through lower half of ovary, short-styled form (X 57)-
7. Section through central part of same ovary (X 100). 8. Longitudinal diagram
of similar ovary, constructed from threee sections (X 35); from buds 3 mm. long.
9. Curved style-tip of mid-styled flower with small and mid-size pollen grains
germinating on the stigma (X35). 10. Six-parted stigma of long-styled flower,
three of its divisions lying behind those shown; epidermal cells indicated only at
the top; below the elongated subepidermal idioblasts, shaded (X37). a, antho-
cyanin idioblasts, shaded, mostly subepidermal; b, bundles of raphides; c, groups of
secretory cells, at c' abutting on small schizogenous cavities; d, embryosac; e, empty
loculi; n, septal nectaries.

canals lined with secretory cells. The fact that these cells appear
to be functional long after these of the septal nectaries cease to
show any trace of secretion indicates that they may belong to a
quite different category.

472 Semi-centennial of Torrey Botanical Club

The stamens are always in two sets ; a longer set of three on the
anterior side of the flower, consisting of a pair opposite the lateral
petal-segments, with a median one slightly longer, opposite the
lowest or anterior sepal-segment* the three short stamens
of the posterior set have the shortest one always opposite
the upper blotched petal-segment, flanked by a longer one on
either side opposite the lateral sepal-segments (fig. i) ; the stamens
of this pair show about equal length in the short-styled flower, but
in the long- and mid-styled flowers one is longer than the other,
so that the three anthers of the shortest sets stand one above the
other in the narrow tube in such a position that an insect's pro-
boscis would almost certainly graze all three when seeking the
nectar accumulated in the basal portion ot the tube. The longer
stamens w^ould generally be described as inserted on the perianth
tube about at the throat, but they may easily be traced as thickened
ridges down to the base of the tube.

The stigma of the long-styled form reaches a height of 1 2-1 3.5
mm,, averaging (in ten flowers) about 12.5 mm., and correspond-
ing fairly closely with the height of the longest stamen of the
mid-styled form, which is 13. 5-15 mm., and also with that of the
longest stamen in the short-styled form, which is 1 3-14.5 mm.
The stigma of the mid-length-styled form reaches a height of 7-8
mm., corresponding with the longest of the mid-length stamens
of the long-styled form, whose anther stands about 9-10 mm.
above the base of the ovary, and also with the mid-length stamens
of the short-styled form, whose median or shortest another stands
6.5-8 mm. above the base of the ovary. The stigma of the short-
styled form is only 2.7-3 iri^n- above the base of the ovary, and
the shortest stamen of the long-styled form has the tip of its
anther 3-3.5 mm. above the base of the ovary, while the anther
tip of the shortest stamen in the mid-styled form measures 2.6-3
mm. above the base of the ovary. Measurements of all three
stamens of the mid-length sets would show a closer correspond-
ence with the mid-length style, but it may be noted that the
stamens of the shortest sets are all generally taller than the shortest
styles.

The ratio of the average height of the long pistils to that of the
mid-length pistils is approximately as 100 to 60; and the average
height of the long pistils to that of the short ones is as 100 to 22.

Hazen: Trimorphism and insect visitors of Pontederia 473

Fig. II. Pontederia cor data; flowers of the three forms in approximately
natural position (X 4). The dotted lines with arrows indicate the six legitimate
pollinations. From each flower the left lateral petal-segment with its stamen, and
half of the left sepal-segment and its stamen have been cut away, in order to show
the position of the short stamens with their inverted anthers and the short pistil.
Drawn 28 August 1916.

474 Semi-centennial of Torrey Botanical Club

The six legitimate crosses which may take place between the
six sets of stamens and the three different pistils are indicated by
the dotted lines and arrows in figure i i .

The pollen grains from the different stamens show differences
in size similar to those reported by Darwin for the water-hyacinths
and for the Pontederia found by Fritz Miiller in the interior of
Brazil ; those of the two longest sets of stamens from mid-and short-
styled flowers are largest, indicating that they are adapted to
pollinating the long-styled pistil; those of the very short stamens
from long- and mid-styled flowers appear about a quarter as large

Long-styled Mid-styled Short -styled

raid-stamens longest stamens longest stamens

shortest stamens shortest stamens mid-stamens

FiG. 12. Pollen grains from the three flower forms, to show comparative size,
and similarity of those taken from stamens of similar length in the different forms

(X 200).

(their actual volume averages only one seventh as great) indi-
cating that they would pollinate only the short-styled pistil; while
those of the mid-length stamens of the long- and short-styled
flowers are intermediate in size, indicating adaptation to polli-
nating the mid-styled flower (fig. 12). Unfortunately for con-
venience of exact comparison, the pollen grains of our species
usually are not spherical, as intimated by Darwin for the species
studied by him, but they are here rather ellipsoidal or lemon-
shaped; in only ten out of three hundred measured have I found
isodiametric grains.

In order to obtain as definite a record as possible, five flowers
of each form were selected, each flower from a different spike.

Hazen: Trimorphism and insect visitors of Pontederia 475

except in the case of the short-styled form, of which only three
spikes had been preserved, and from each flower five pollen grains
from the longest stamen and five from the shortest were selected
at random and measured by means of the eyepiece micrometer.
The twenty-five measurements from each of the six stamen types
were then averaged. Some weeks later a duplicate set of twenty-
five measurements was made in the same manner, and while they
averaged uniformly slightly greater than those of the first set, the
difference was only such as might be expected because of the
personal equation, though in each of the three forms of the second
set it was noticeable that one or two of the flowers were especially
vigorous, as shown by a considerable increase in size of all grains,
both larger and smaller, above the average for similar anthers in
other flowers. Nevertheless, when the second set of measurements
was combined with the first, the ratios between the different sized
grains diverged from those obtained from the first set by only a
negligible amount, indicating the substantial reliability of the
work. The fifty measurements of each type of pollen give the
following result:

Long styled form, from median mid-length stamens 37-44 X 33-79 microns

from shortest stamens 23.94X21.69

Mid-styled form, from longest stamens 46.33 X 41.61

from shortest stamens 23.95 X 20.95

Short-styled form, from longest stamens 45.84X41.32

from median mid-length stamens 36.94 X 33.01 " ^

It will be seen that the pollen grains from the two sets of longest
and shortest stamens correspond most closely, the mean diameters
of the shortest sets differing by less than half a micron, but
even the divergence between the mean diameters of the two
mid-length sets amounted to only seven tenths of a micron.
There is a much closer correspondence here than in the
smaller number of measurements made by Francis Darwin
for his father on the Brazilian plant. It does not appear
whether Darwin selected pollen grains from more than one
anther of each type, and there is considerable variation in different
plants. The extremes found in Pontederia cordata may be of
interest. The grains from the median (i. e., longest) mid-length
stamen of the long-styled flowers showed such averages in dif-

476 Semi-centennial of Torrey Botanical Club

ferent flowers as 35.4 X 34-2, 39-3 X 36, and 36.6 X 33 microns;
the similar grains from the median (i. e., shortest) mid-length
stamen of the short-styled flower showed such averages as 34.8 X
32.7, and 38.1 X 33.6 microns. The average for a single flower of
grains from the longest stamen in the mid-styled form ranged
from 42.3 X 37.2 to 49.5 X 42.3 microns; while the small grains
from the shortest stamen of a single long-styled flower furnished
such averages as 22.8 X 23.1, 24 X 21, and 25.8 X 23.1 microns.
The proportions of the grains are by no means constant, and in
examining many grains one gains the impression that the two (or
really three) diameters balance each other even more than is
indicated in these sample averages, that is, that when one diameter
increases beyond the average, the other decreases correspondingly
so as to keep the volume average more constant than appears.
It should be remarked that the above measurements were all
based on material preserved for a year in formalin, and that they
uniformly average less than a small number of measurements made
in August 191 6, of grains taken from fresh flowers and mounted in
water. These few measurements of fresh grains, considered not
enough to be reliable, in general approach more nearly the dimen-
sions of a series made by Halsted* whose report came to my notice
only after my own had been completed. As indicated by
Halsted, dry pollen from fresh anthers is so contracted as to
make its measurement of little significance. It might be expected,
however, that though the amount of swelling of the grains would
be greater when mounted in water than when fixed in formalin,
nevertheless it would be proportional in similar grains, whatever
the medium used; and as a matter of fact the ratio between my
large- and mid-size grains is almost exactly the same in fresh and
preserved material; a greater difference between the size of large
and smallest grains mounted in water as compared with similar
grains in formalin, I assume to be chargeable to the small num-
ber of measurements of fresh grains.

In order to obtain a comparison between our ellipsoid pollen-
grains and those reported by Darwin as spherical, the measure-

* Halsted, B. D. Pickerel weed pollen. Bot. Gaz. 14: 255-57. 1889. In this
brief article, no indication of the number of measurements is to be found; with the
mere statement that " only three prevailing dimensions " occur, and indications of
faulty calculations, it seems worth while to detail my own definite results.

Hazen: Trimorphism and insect visitors of Pontederia 477

ments of each of the two similar sets of grains given above were
combined, and the mean of the three diameters of the average
grain of each kind was then taken; this makes the mean diameter
for all the large grains 43.3 microns, for all the mid-size grains
34.66 microns, and for all the small grains 22.16 microns. On this
basis the ratio of all the large to all the mid-size grains is as 100
to 80, and the ratio of the latge to all the small grains is as 100
to 51, which is a slightly greater difference than that reported by
Halsted, who pointed out the fact that Pontederia cordata shows
the greatest range of pollen size yet recorded for any flower.
This method of averaging, however, is not accurate, and in any
case comparison of volumes would seem to be more significant.
Computing the volume of spheroids with diameters represented
by the measurements detailed above, or, more simply, calculating
the ratios only by use of logarithmic tables, it is found that the
volume average of the two sets of large grains is to that of the mid-
sized grains as 100 is to 53, and the latio of the volume of the large
to that of the small is approximately as 100 to 14. It will be
seen that^these ratios present a much better basis for comparison
with the ratios of style length than the ratios of the diameters.
Darwin's comparison, however, was based on the extremes of size
in single sets of pollen grains. In our plant the largest and smallest
sets of grains are found in the mid-styled form, where the ratio of
mean diameters is as lOO to 50, and the volume ratio about as
100 to 13. In our species, as in Darwin's Brazilian plant, the
pollen grains of both sets of stamens in the short-styled flower are
slightly smaller than those of the stamens of corresponding length
in the other flower forms.

The significance oi these differences in pollen size is a point
of much interest. Delpino regarded the difference in size as a
direct adaptation to the style length, supposing the larger grains
could produce a pollen-tube long enough to penetrate the length
of the long style, and that the tube of the smallest grains would
readily grow only the length of the short style. This view Fritz
Miiller considered confirmed by his experimental work with Eich-
hornia crassipes* where he found that long- and mid-styled pistils

* MuUer found, for example, that flowers on a long-stj-led spike legitimately
fertilized by pollen from the long stamens of mid-styled flowers produced 14 1.7

478 Semi-centennial of Torrey Botanical Club

pollinated with grains from the smallest anthers were less fruitful
than in the case of other illegitimate crosses.

An additional suggestion may be gathered from Halsted's
experiments made to determine whether the small pollen grains-
were fertile, since doubt on this point had been expressed by more
than one previous writer. He found all the grains equally capable
of germination if sufficient time was given, but that the largest
grains germinated much more promptly. In Pontederia such
promptness of germination of the large grains would be of great
importance for the long- and mid-styled plants, inasmuch as the
style withers so early that the pollen-tube of a slow-germinating
grain might be unable to reach the ovule. I have found that large-
and mid-size pollen-grains of P. montevideensis both germinate
very quickly in weak sugar solutions; the only apparent difference
is that the pollen-tube from the mid-size grains has a diameter
about three fourths as great as that of the large grains. I had no
flowers containing small grains for comparison. But this is a
point which can hardly be settled by study of one small group of
species.

In Lythrum and other heterostyled flowers it has been noted
that the stigma of the long-styled form is larger than those of the
mid- and short-styled flowers, and it has been considered that
the longer stigmatic papillae are adapted to receive the large pollen-
grains. In Pontederia cordata there is very little difference in the
length of the papillae in the different stigmas (figs. 8-io) but
the stigma of the long-styled flower is frequently, though not
uniformly, six-parted, and this spreading stigma may be regarded
as directly correlated with its exserted position, for such a stigma
would have a distinctly better chance of being dusted with pollen
by the insect visitor. In the case of the mid- and short-styled
pistils, however, there is no need for such a spreading stigma, since
the perianth-tube would almost certainly guide the pollen-smeared
proboscis of an insect in such a manner as to brush even a narrow
stigma.

Observation of the manner in which the flowers are placed on

seeds per capsule; other flowers fertilized by pollen from the mid-length stamens of
the same spike produced 12 1.3 seeds; while still other flowers fertilized by pollen
from the short stamens of mid-styled flowers produced 113.3 seeds. Kosmos 13:
298. 1883.

Hazen: Trimorphism and insect visitors of Pontederia 479

the axis of the spike in a nearly horizontal, though slightly as-
cending position, indicates that automatic self-pollination is
regularly precluded (fig. ii and PL. 14). In the long-styled flower
the stigma stands out stiffly too far to be reached by pollen from
its own anthers, and these are so nearly included in the perianth-
tube that pollen from them would be very unlikely even to fall on
stigmas lower down on the sanie spike; in the short-styled form
it would be impossible for the pollen to drop down the long, narrow
tube to the low stigma. Only in the mid-styled form does it appear
that pollen from the long-exserted anthers might possibly fall on
the stigma of a lower flower, and here again in all ordinary cases
the erect upper lip of the perianth would protect the lower stigma,
which furthermore hardly projects from its tube far enough to
catch pollen from above.

Among the insect visitors of the pickerel-weed ten species of
Lepidoptera, distributed among five families, and showing nearly
all possible range in size were collected during several excursions
in July and August, 1916, and all of these but two or three were
photographed as they sipped nectar from the flowers, several
species many times. The list includes the least skipper, Ancy-
loxypha numitor Fabr. ; the yellow-spotted skipper, Polites peckius
Kirby; the silver-spotted skipper, Epargyreus tityrus Fabr.; the
variegated fritillary, Euptoieta claudia Cramer; the clouded sul-
phur, Colias philodice Godart; the white cabbage butterfly,
Pieris rapae L.; the viceroy, Basilarchia archippus Cramer; the
tiger swallow-tail, Papilio turnus L.; the black swallowtail,
Papilio polyxenes Fabr.; and the humming-bird moth, Hemaris
thy she Fabr. It is strange that the monarch butterfly, Anosia
plexippus L., which was frequently seen on neighboring plants of
Joe-Pye weed, never visited the Pontederia, a.nd the so-called mimic,
the viceroy, made only one fleeting visit; they evidently prefer
the large flat- topped flower clusters of Eupatorium and milk-weed,
or the nectar found there; or is it possible that blue flowers do not
attract them? Another curious case was that of the pearl crescent
butterfly, Phycioides tharos Kirby, which was the commonest
visitor of the vervain. Verbena hastata L., growing close to the
pickerel-weed ditch, but never came to the Pontederia, though
the latter possesses much the same blue color and has a similar

480 Semi-centennial of Torrey Botanical Club

flower structure. Perhaps a parallel case is that of the honey-bee,
which was the most abundant visitor of the vervain, and is gener-
ally regarded as catholic in its tastes, but which was not taken on
the pickerel- weed. These cases may have a bearing on the view
held by some entomologists, that insects have less color sense than
has been supposed, but are much more keenly attracted by odors
of particular plants, which may not always be perceptible to man.

Of the Hymenoptera, the following species were taken from
the middle to the last of August: the bumble-bees, Bombus fer-
vidus Fa.hr., B. impatiens Cress., B. pennsylvanicus DeGeer; and
a smaller Anthophorid bee, Xenoglossa pruinosa Say (?) perhaps
only gathering pollen. Of Diptera, one specimen of a large
"horse-fly," Tabanus giganteus DeGeer, was perhaps only a
casual caller.*. An aggressive large blue-black bee, observed
several times, but always eluding the pursuing reflex-camera as
well as the net, was probably one of the carpenter-bees of the genus
Xylocopa. Schneckf has reported that Xylocopa virginica regu-
larly slits the lower end of the corolla tube to reach the nectary
in Pontederia and in other plants. In the present case, however,
I feel sure that the bee was sucking nectar from one flower after
another in legitimate fashion.

LovellJ reports as visitors to Pontederia at Waldoboro, Maine,
July 21 to August 10, 1898, two species of bumble-bees, two species
of small cliff-dwelling bees, one collecting pollen, and four species
of Diptera, all feeding on pollen; butterflies, he states, were com-
paratively rare, and only Colias, Pieris, and Argynnis cybele Fabr.
are mentioned. During one warm but cloudy afternoon in August,
at the New York Botanical Garden, I saw no butterflies visiting
Pontederia, but only bumble-bees. Similarly, during August.
1 91 7, abundant colonies of pickerel-weed growing in the open
border of Lake Cossayuna in Washington County, New York,

* For the identification of these insects I am indebted to Dr. Frank E. Lutz, of
the American Museum of Natural History, whose Field Book of Insects published
in January, 1918, will be a handy guide for the student of floral biology. The
Lepidoptera, identified by myself, were all compared with specimens at the American
Museum.

t Schneck, J. Further notes on the mutilation of flowers by insects. Bot.
Gaz. 16: 313. 1891.

% Lovell, J. H. Three fluvial flowers and their visitors. Asa Gray Bull. 6: 63-65.
1898.

Hazen: Trimorphism and insect visitors of Pontederia 481

were watched from a boat, and no butterflies, but numerous bees
were seen visiting the plants, though one of the photographs taken
there shows clearly a humming-bird moth with extended proboscis
poised before a €ower. Knuth would doubtless place this plant
having a perianth tube of seven or eight millimeters in length
among his groups of bee- or humble-bee-flowers, but the record
at Areola shows clearly that such a classification cannot be rigidly
adhered to in this case, for there certainly the Lepidoptera sur-
passed the bees as visitors of this plant both in number of species
and individuals. This I think was the case throughout the whole
of July and August, though constant pursuit with the camera and
net prevented making an exact record of the number of visits of
any particular species.

Unquestionably the least skipper, Ancyloxypha, was the most
frequent visitor, two or three individuals often being present on
one spike, and often one of them flitted to several flowers on the
same spike in succession; this was always an attractive little
butterfly as the golden-brown scales on the lower surface of the
folded wings caught and reflected the sun. These smaller butter-
flies, using the alighting platform furnished by the spreading
lower Hp of the flower, get the under side of their thorax or abdomen
well dusted with pollen from the longest stamens (plate 15) and
then carrying it to a long-styled flower rub off some of it on the
protruding stigma; at the same visit they may dust the head with
pollen from the mid-length stamens, or thrusting the proboscis
into the tube on the upper side of the flower where there is a
wider space between the perianth and the pistil than on the lower
side, on withdrawing it after sucking the nectar, they drag it
through the row of three inverted anthers of the shortest stamens
(fig. i) to carry the pollen away to other flowers with appropriate
length of style. Larger butterflies, like the silver-spotted skipper,
often stand out farther from the flowers so that only the legs and
proboscis tip become dusted with pollen, though one photograph
of the black swallow-tail shows it grasping the spike with the
abdomen tightly pressed against the flowers. The persistent
visits of these large butterflies furnish the strongest reason for
doubting the rigid applicability of Knuth's classification in the
case of this plant. Standing at the south end of a ditch filled,

482 Semi-centennial of Torrey Botanical Club

like a garden, border, with the bright-flowered plants, I would see
a black swallow-tail alight at the north end and flit, sipping rest-
lessly, the whole length of the bed to the point where the camera
was focussed on the nearest spike; then he w^uld fly without,
stopping straight back to the north end. and repeat the perfor-
mance, until, after three or four exposures in the same place, he
was captured. Query, why did he always proceed in the same
direction, never reversing? Perhaps for some reason he prefers
to sip while flitting toward the sun.

It appears, then, that the pickerel-weed is well supplied with
a varied and constant procession of visitors, which must serve Jt
effectively while supplying themselves with food. Although
illegitimate pollinations may be frequent, at least some kind of
cross-fertilization is undoubtedly the rule after these visits.
Self-pollination, however, is probably possible; the little skippers
might easily thrust some pollen from protruding anthers down to
the stigma of the same flower, or might carry it from one flower
to another of the same spike. But the question arises, would
such pollination be effective, or are the plants self-sterile? And
which of the legitimate crosses are most fruitful? For the purpose
of obtaining some light on these points, plants were brought from
the field and kept in pots placed in tubs of water in the greenhouse.
Numerous crosses and self-pollinations were made, and many of
the pistils so treated apparently set seed, but owing to various
accidents not so many were brought to maturity. It can only be
reported at present that a few mid-styled flowers matured ap-
parently good seed when pollinated from their own long stamens,
and also when pollinated from long stamens of other flowers on
the same spike; owing to further accidents, no seeds were ger-
minated. More definite results will be looked for from future
experimental work.

It is interesting to note that in his second paper* Fritz Miiller
reported having found himself mistaken in his early assumption
that Eichhornia crassipes is unfruitful with its own pollen, for
although only one plant was introduced in his region, his nephews
discovered seeds and young seedlings, and the barrenness had
been due merely to the lack of proper insect visitors.

*Kosmosi3: 297. 1883.

Hazen: Trimorphism and insect visitors of Pontederia 483

In the field one is often inclined to think that one of the three
forms of the species predominates in a particular locality. This
might occur quite as readily as a result of vegetative propagation
by rapid growth of the rhizomes as by greater fruitfulness of one
form. But the pickerel-weed stem regularly becomes geniculate
after flowering, usually just below the insertion of the spathe-like
floral bract (the 'knee' is already indicated in plate 15) and bend-
ing downward, the inflorescence is lowered into the water for ma-
turing the fruits, and they are most apt to fall to the bottom in
the near vicinity of the parent plant ; in this way also extensive
patches of one plant form may be established. Miiller reported
in 1883 ( op. cit. 299) that all the Pontederiaceae known to him,
including Heteranthera renijormis and H. zosteraefolia, the two
species of Eichhornia, and the Pontederia from Curitibanos, have
this habit of bending the flower-stalk down to the marshy ground
or water in which they grow. In 191 2 Hauman-Merck* reported,
as a peculiarity which he thought had escaped previous observers,
this habit of maturing the fruit under water in Pontederia rotiindi-
folia, and stated that P. cordata growing in abundance in the same
places in pools of the banks of the Rio de la Plata matures its
fruits out of the water. This statement is rather surprising, since
our plants are so fixed in this habit of bending down after flowering
that plants kept in the greenhouse with little water develop in a
manner precisely similar to those left in the field. Pontederia
montevideensis shows practically no such tendency when grown here
and it would appear possible that Hauman-Merck was really
dealing with this species rather than with P. cordata.

The final judgment reached at the Areola station was that all
three forms of the plant were about equally numerous there.
During the dry August (in the region of New York) of 191 7, one
visit to the station revealed such a desolate and discouraging
group of plants that no attempt was made to do anything further
during that season. Perhaps clumps of the plant growing in the
borders of the Hackensack River, if they could have been reached,
might have proved more rewarding in such a season, for, contrary
to the usual statement that this family comprises only fresh water

* Hauman-Merck, Lucien. Sur un cas de geotropisme hydrocarpique chez
Pontederia rotundifolia L. Rec. Inst. Bot. Leo Errera 9: 28-32. 1912.

484 Semi-centennial of Torre y Botanical Club

forms, here plants of Pontederia are daily bathed, or even flooded,
by tide water.

explanation of plates 14 AND 15

Plate 14. Pontederia cordata: at the left, a short-styled spike, longest and
mid-length stamens protruding from the perianth tube; in the center, raid-styled
spike, long stamens only exserted, style-tip barely protruding from tube; at the
right, long-styled spike, mid-length stamens only slightly protruding in throat of
tube, styles exserted and about as long as the perianth segments. Photographed
10 Aug. 1916, about three fourths natural size.

Plate 15. Spike of the mid-styled form visited by the least skipper, Ancy-
loxypha numitor; photographed at Areola, N. J., 15 Aug. 1916.

The drawings for all the text figures were made with the aid of the camera
lucida, figures 1-4, 9, 10, and 12 from material preserved in formalin, figure 11 from
fresh flowers.

Mem. Torrey Club

Volume 17, plate 14

Mem,

ToRREY Club

Volume 17, plate 15

INDEX

Abena jamaicensis, 153
Abies balsamea, 297

Abutilon abutiloides, 147; hirtum, 146,
147; indicum, 147; lignosum, 147;
permolle, 147
Acacia Farnesiana, 117; Greggii, 310
Acer pennsylvanicum, 298; rubrum, 276,
277

Acidity in plant tissues, Determination
of, 8. 241

Acids and alkalies on the growth of the
protoplasm of pollen-tubes, The effect
of, 8, 84

Achillea Ptarmica, 339

Achlya, 57; apiculata, 57

Achras Sapota, 169

Achyranthes, 95

Aconitum Napellus, 350

Acrostichum gorgoneum, 96

Actinothyrium, 70; Hopeae, 70

Adams, J. F., B. O. Dodge and. Some
observations on the development of
Peridermium Cerebrum, 7, 253

Adams, J. F. Origin and development
of the lamellae in Schizophyllum
commune, 8, 326

Adenocalymna, 155

Aecidium, 56, 98, 99; circumscriptum,
160; Cissi, 161; desmium, 113; Fara-
meae, 163; Gossypii, 76; Nymphaea-
rum, 106; Nymphoidis, 140; passi-
floriicola, 162; Pisoniae, 161; Puerariae,
56; punctatum, 119; Rivinae, 106,
142; roseum, 159; simplicius, 162;
Tournefortiae, 162; tubulosum, 162

Aeschynomene americana, 112

Agalinis acuta, 274, 283; purpurea, 278

Agaricus, 326; campestris, 179, 181, 187;
Frostianus, 251; vernus, 247

Agave Cantula, 69

Ageratum maritimum, 158

Agrostis alba, 274, 279

Aletris farinosa, 274, 278, 283

Aleurites, 95; moluccana, 93

Algae, Bermuda, 2

Alkalies on the growth of the protoplasm
of pollen-tubes. The effect of acids and,
8, 84

AUodus megalospora, 152; Podophylli, 48
Alocasia indica, 66
Alphitonia, 95

Alternaria, 72; Brassicae, 72, 73
Alyxia, 95

Amanita, i76» 178, 246, 249, 250, 252; bac-

cata, 248; bisporigera, 246; citrina,
177; cothurnata, 247, 248, 249; crenu-
lata, 177; fiavoconia, 251, 252; Fros-
tiana, 251, 252; jonquillea, 248-251;
Morrisii, 177; muscaria, 177, 178, 199,
250-252; pantherina, 247, 248, 250,
251; phalloides, 177, 246; spreta, 177;
velatipes, 248, 250, 251; verna, 246,
247

Amanita, Six misunderstood species of,
2, 246

Amanitopsis, 249; albocreata, 248-250;

vaginata, 177, 249; volvata, 177
Ambrosia, 139; artemisiifolia, 279
Amelanchier canadensis, 296
Amphisphaeria, 65; hesperidum, 65
Amphitrichum Hibisci, 59; Sacchari, 59
Amygdalus Persica, 120
Ancyloxypha, 481; numitor, 479, 484
Andropogon, 44, 269; furcatus, 274, 276,

283, 286; halepensis, 135; scoparius,

274-276, 283
Anemone, 120; nemorosa, 347
Aneura, 92
Aneura maxima, 92

Annona, Cherimolia, 167; reticulata, 166;

squamosa, 167
Anoda hastata, 146, 147
Anoectochilus, 95
Anosia plexippus, 479
Antennaria neglecta, 274, 283; planta-

ginifolia, 275, 283
Anthemis arvensis, 339, 347
Anthephora elegans, 137; hermaphrodita,

137

Anthostomella, 66; mirabilis, 66

Antidesma, 95

Arachis hypogea, 73, 167

Aralia hispida, 299

Areca Catechu, 70; sapida, 70

Argynnis cybele, 480

Argyroxiphium, 93

Armillaria mellea, 50

Arnica montana, 343;. unalaschensis, 346
Aronia nigra, 277

Arthur, J. C, and J. R. Johnston.

Uredinales of Cuba, 2, 97
Artocarpus communis, 164, incisa, 164
Arundo Donax, 60
Aschersonia, 70; sclerotioides, 70
Asclepias curassavica, 133, 150; nivea,

133; pulchra, 278
Ascophanus, 58; verrucosporus, 58

486 Semi-centennial of Torrey Botanical Club

Aspergillus, 71; niger, 287, 288, 290;
periconioides, 71

Aspergillus niger, Some factors influenc-
ing the stimulative action of zinc sul-
phate on the growth of, 2, 287

Aspidium aristatum, 96; caryotideum,
96; squamigerum, 96; terminans, 96;
truncatum, 96

Asplenium Adiantum-nigrum, 96; cauda-
tum, 96; contiguum, 96; ebenoides,
81;' fragile, 96; horridum, 96; Nidus,
96; normale, 96; polyphyllum, 96;
spathulinum, 96; varians, 96

Astelia, 95

Aster dumosus strictior, 274, 283; eri-
coides, 279; prenanthoides, 345, 347;
salicifolius, 278

Asterina, 62; Elmeri, 63; Lawsoniae, 63;
pelliculosa, 63; pemphidioides, 62

Astrantia major, 343

Astrocystis, 64; mirabilis, 64, 66

Atkinson, George F. Six misunder-
stood species of Amanita, 2, 246

Austin, Coe Finch, A sketch of the
life of, 2, 31

Avena sativa, 138

Baccharis, 158
Bacterium malvacearum, 75
Bambos vulgaris, 165
Bambusa, 64, 66
Banisteria laurifolia, 144
Baptisia, 269; tinctoria, 271, 274, 280,
281, 283

Barnhart, John Hendley. Historical
sketch of the Torrey Botanical Club, 12;
History of the Torrey Botanical Club, i

Bartonia virginica, 275, 280

Basilarchia archippus, 479

Bauhinia divaricata, 128; heterophylla,
167; rubiginosa, 167

Baumea, 95

Bean-pod as compared with that of the
bean-plant. The individuality of the, 2,
207

Befaria, 46

Bellis perennis, 338, 339
Bermuda algae, 2
Bermudiana, 82

Betula lenta, 296; lutea, 296; populifolia,
273

Bidens leucantha, 134; pilosa, 134
Bignonia aequinoctialis, 155
Bjerkandera adusta, 49
Blakeslee, a. F., and P. A. Warren.

Parthenocarpy in cucumbers, 9
Blechnum, 469
Blechum Brownei, 156
Bletia patula, 166

Blodgett, F. H. Weather conditions

and crop diseases in Texas, 9, 74
Boas, Helene M. The individuality of

the bean pod as compared with that of

the bean plant, 2, 207
Boas, Helene M,, A. B. Stout and.

Statistical studies of flower number

per head in Cichorium Intybus: Kind

of variability, heredity and effects of

selection, 334
Boletus, 177, 275, affinis, 177; bicolor,

177; chromapes, 177; felleus, 177;

miniato-olivaceus, 177; miniato-oliva-

ceus sensibilis, 177; ornatipes, 177;

paluster, 177; Ravenelii, 178; Roxanae,

178; separans, 177
Boitonia latisquama, 343
Bomarea, 45

Bombus fervidus, 480; impatiens, 480;

pennsylvanicus, 480
Borreria laevis, 156

Botanical Club, History of the Torrey, i
Botanical collecting in the Republic of

Colombia, Recent, 8, 39
Botrychium lanceolatum, 297
Botryodiplodia, 70; Elasticae, 70
Botryorhiza Hippocrateae, 161
Botrytis, 202-206; vulgaris, 203
Bridelia, 68

Britton, N. L. The flora of the Isle of
Pines, Cuba, 7; Torrey Botanical Club
reminiscences, 2, 24

Bromus, 138

Burgess, Edward S. A method of
teaching economic botany, 3, 52

BuRLiNGHAM, GERTRUDE S. A pre-
liminary report on the Russulae of
Long Island, 8, 301

Byronia, 94

Byrsonima coccolobifolia, 114; crassi-
folia, 114

Caesalpinia bahamensis, 118; pulcher-

rima, 118; Rugeliana, 118
Cajan Cajan, 129
Cajanus indicus, 129
Calliospora Farlowii, 119
Calocera cornea, 49
Caltha palustris, 349

Campbell, Douglas Houghton. The

origin of the Hawaiian flora, 2, 90
Camptosorus rhizophyllus, 297
Canna, 141; indica, 141
Cantharellus aurantiacus, 176
Cardiospermum microcarpum, 144
Carex lurida, 278; pennsylvanica, 274,

283; sterilis, 278
Carica Papaya, 71
Carpinus caroliniana, 296
Cassia emarginata, 105, 118, 119; occi-

dentalis, 68, 72, 73; robinaefolia, 118
Castalia odorata, 278
Castilla elastica, 70, 164
Catskills, Two months in the southern, 2,

294

Index

487

Cayaponia racemosa, 134

Cenchrus echinatus, 136; viridis, 136

Centaurea 465; alba, 344; atropurpurea,

344. 346, 349; Cyanus, 339; Jacea, 339
Centrosomes during early fertilization

stages in Preissia quadrata, 2, 323
Cephalanthus occidentalis, 170
Cercidium, 310; Torreyanum, 310
Cercospora, 72; Gliricidiae, 72; Litseae-

glutinosae, 72; occidentalis, 72; per-

sonata, 73

Ceriomyces communis, 50; crassus sep-

arans, 50; viscidus, 50
Cerotelium Fici, 112; Gossypii, 113
Cerrena unicolor, 49

Chaetochloa geniculata, 137; imberbis,

137; onurus, 137; purpurascens, 137;

setosa, 137; verticillata, 137
Chaetomium, 64; stercoreum, 64
Chamaesyce hirta, 131; hypericifolia, 131
Champeria manillana, 63
Chanterel cinnabarinus, 50; minor, 50
Chiogenes hispidula, 299
Cholisma ligustrina, 277
Chrysanthemum inodorum, 338; Leucan-

themum, 338, 344, 347; segetum, 338,

354. 453.

Cichorium 339, 354; Intybus, 356-358
Cichorium Intybus: Kinds of variability,

heredity, and effects of selection,

Statistical studies of flower number per

head in, 334
Cichorum, Statistical studies in, 8
Ciliciopodium, 73; Grayanum, 73
Cionothrix, 115; Cupaniae, 115
Cirsium discolor, 274
Cissus rhombifolia, 114; sicyoides, 161
Citrus, 70, 72; Aurantium, 65, 72; de-

cumana, 65; nobilis, 70
Cladonia symphicarpa epiphylla, 267, 275
Cladosporium, 72; personatum, 73; sub-

fusoideum, 72
Claoxylon, 95

Clavaria coronata, 49, cristata, 49
Clintonia borealis, 298; umbellulata, 298
Clitocybe, 178; dealbata sudorifica, 177,

178; illudens, 177-179; infundibuli-

formis, 50; lactariiformis, 50; multiceps,

177-179, 188; nebularis, 176; virens, 50
Clitoria rubiginosa, 128
Coastal locations, Comparative cultures

of seed-plants in desert valley, desert

mountain, and, 9
Coccolobis Uvifera, 164
Cocos flexuosa, 68; nucifera, 68, 71
Coelastrum, 229
Colchicum autumnale, 459
Coleosporium, 107, 108; Elephantopodis,

109; Eupatorii, iii; Ipomoeae, no;

Plumierae, no; Vernoniae, no
Coleus, 451

Colias, 480; philodice, 479

Collecting Fungi at Delaware Water

Gap, 7, 48
Colocasia, 93

Colombia, Recent botanical collecting

in the Republic of, 8, 39
Coltricia cinnamomea, 49
Comandra, 275; umbellata, 275
Commelina, 127; nudiflora, 127, longi-

caulis, 127

Comparative cultures of seed-plants in
desert valley, desert mountain, and
coastal locations, 9

Comptonia peregrina, 274

Coniothyrium 69; melasporum, 69

Contact and pressure reactions in Pedias-
trum simplex, 2,210

Convolvulus, 105, 145; dissectus, 169;
nodiflorus, 133

Coprinus atramentarius, 50; micaceus,
50, 331; Spraguei, 50

Coprosoma, 95

Cordyline, 95

Coriolellus sepium, 49

Coriolus abietinus, 49; molliusculus, 49;
nigromarginatus, 49; versicolor, 49

Cornus canadensis, 300

Cortinarius corrugatus, 50

Cortinellus multiformis, 50; rutilans, 50

Corylus americana, 274

Cotton-rust epidemic in Texas, A, 2

Cracca virginiana, 274, 283

Craterellus cornucopioides, 49

Crinipellis zonata, 50

Crocanthemum, 274;dumosum, 280, 283

Cronartium Byrsonimatis, 114; notatum,
114; Wilsonianum, 114

Crop diseases in Texas, Weather con-
ditions and, 9

Crotolaria stenophylla, 62

Crucibulum vulgare, 50

Cuba, The flora of the Isle of Pines, 7

Cuba, Uredinales of, 2, 97

Cucumbers, Parthenocarpy in, 9

Cupania americana, 115; glabra, 115;
macrophylla, 131

Cyathodes, 95

Cyathus striatus, 50

Cydista, 155; aequinoctialis, 155

Cyperus, 139; ferax, 139

Cypholophus, 95

Cyrtandra, 95

Daedalea confragosa, 49; quercina, 49
Dalbergia, 125; Amerimnum, 125
Daldinia concentrica, 48
Dalea domingensis, 119
Darluca melaspora, 69
Daucus Carota, 279

Delaware Water Gap, Collecting Fungi
at, 7, 48

Demarest, Mrs, Abraham (Sarah

488 Semi-centennial of Torrey Botanical Club

Elizabeth Austin). A sketch of the
Hfe of Coe Finch Austin, 2, 31
Dendroceros, 92

Dendroecia, 117; Lysilomae, 117

Dennstaedtia punctilobula, 297

Denslow, Herbert McKenzie, Rem-
iniscences, 2, 22

Desert valley, desert mountain, and
coastal locations, Comparative cul-
tures of seed-plants in, 9

Desmodium, 60; capitatum, 62; gangeti-
cum, 60; Scorpiurus, 62, 130; tortuo-
sum, 130; trifiorum, 62; virgatum, 60

Determination of acidity in plant tissues,
8, 241

Dianella, 95

Dictyophora duplicata, 50
Didymosphaeria, 66; striatula, 66
Diffiugia, 451; corona, 210
Dimorphotheca pluvialis, 209
Diodia, 156

Dioscorea, 165; convolvulacea, 166

Dipholis salicifolia, 149

Diplodia, 69; Agaves, 69

Diseases in Texas, Weather conditions

and crop, 9, 74
Dodge, B. O., and J. F, Adams. Some

observations on the development of

Peridermium Cerebrum, 7, 253
Dolichos, 129; Lablab, 129
Doodya media, 96
Dothichiza populea, 48
Dothidea grammodes, 62; perisporioides,

62

Dothidella grammodes, 62
Dracaena, 95

Drosera, 26; intermedia, 278

Dryopteris Braunii, 297; Goldieana, 297,

spinulosa, 297, Thelypteris, 297
Dulichium arundinaceum, 277
Dumortiera, 91, 92; hirsuta, 91; trichoce-

phala, 91; velutina, 91

Early horticultural journalism in the

United States, 8, 79
Echium vulgare, 298

Economic botany, A method of teaching,
3. 52

Effect of acids and alkalies on the growth
of the protoplasm of pollen-tubes.
The 8, 84

Eggleston, W. W. The route taken by
Capt. Nathaniel J. Wyeth and Mr.
Thomas Nuttall from the Mississippi
River to the Columbia River in 1834, 8

Eichhornia, 461, 466, 483; azurea, 460,
463, crassipes 460, 477, 482; monte-
vidensis, 466

Elaeocarpus, 94

Elfvingia fomentaria, 49; megaloma, 49
Eleocharis capitata, 140; geniculata, 139;
melanocarpa, 278

Elephantopus mollis, 109

Eleutheranthera ruderalis, 104, 159, 160

Embelia, 95

Emilia sonchifolia, 159

Endophyllum, 163; circumscriptum, 160;

Rivinae, 142, 143; singulare, 114
Endothia parasitica, 48
Entoloma, 177; cuspidatum, 177; nidoro-

sum, 177; rhodopolium, 177; sal-

moneum, 177; sericiceps, 50; sinuatum,

177; strictius, 50, 177
Entomophthora Anisopliae, 57
Epargyreus tityrus, 479
Eragrostis, 126; tephrosanthos, 126
Eriocaulon septangulare, 277
[ Eriophorum gracile, 277; virginicum, 278
Eriosporangium evadens, 157; tucuma-

nense, 154;
Ernodia, 156

Erysiphe Alni, 61; Coryli, 61; guttata, 61;
suffulta, 61

Erythrina glauca, 167

Erythroxylon havanense, 168

Eugenia, 62; Jambos, 148; malaccensis, 93

Eupatorium, 140, 479; album, 343, 344;
hyssopifolium, 274, 283, 286; macro-
phyllum. III; perfoliatum, 277; pur-
pureum, 278; villosum, 158, 159

Euphorbia heterophylla, 131; hirta, 131;
hypericifolia, 131; pilulifera, 131

Euptoieta claudia, 479

Eurya, 94

Euthamia tenuifolia, 274, 278
Evolution of cell types and contact and

pressure responses in Pediastrum, The,

210

Exidia glandulosa, 49
Exocarpus, 95

Exploration in Colombia, Recent bo-
tanical, 8, 39

Faramea occidentalis, 163

Farwell, Oliver Atkins. Sisyrinchium
Bermudiana, 2, 82

Ferns of tropical Florida, The, 2

Fertilization stages of Preissia quadrata,
Centrosomes during early 2, 323

Ficus, 55, 63, 69; Carica, 69, 112; chry-
solepis, 63; Combsii, 112; heterophylla,
64; Minahassae, 64; odorata, 63; ulmi-
folia, 63; validicaudata, 63

Fimbristylis ferruginea, 165

Fischeria crispiflora, 151

Fistulina hepatica, 50

Flammula betulina, 177

Flora of the Isle of Pines, Cuba, The, 7

Flora of the Rocky Mountains and adja-
cent plains, 7

Florida, The ferns of tropical, 2

FossH plants from Porto Rico, 2

Fraxinus, 313, 314; attenuata, 311, 313

Freycinetia, 95

Index

489

Fuirena simplex, 140; umbellata, 165
Fumaria officinalis, 466
Fungi at Delaware Water Gap, Col-
lecting, 7, 48

Gahnia, 95
Galera tenera, 177

Galerula crispa, 50; hemisphaerica, 50

Ganoderma Tsugae, 49

Gardenia, 95

Garnotia, 95

Gaya occidentalis, 147

Gaylussacia baccata, 274

Genea, 59; Thwaitesii, 59

Gentiana Saponaria, 277

Geopetalum candidissimum, 50

Geranium, 142; maculatum, 206

Gerardia decemloba, 274

Gilia, 464

Gleichenia longissima, 96

Gliricidia sepium, 72

Gloeophyllum trabeum, 49

Gloeoporus conchoides, 49; Palmarum, 70

Glomerella Gossypii, 74

Gossypium, 113; acuminatum, 113

Gouania domingensis, 145, 146; lupuloides,
^145, 146; polygama, 105, 145, 146;
'tomentosa, 145

Graff, Paul Weidemeyer. Philippine
Micromyceteous Fungi, 3, 56

Graham, Margaret. Centrosomes dur-
ing early fertilization stages in Preissia
quadrata, 2, 323

Gratiola aurea, 278

Guepinia spathularia, 49

Gymnoconia interstitialis, 48

Gymnogramma calomelanos, 164

Gymnogramme javanica, 96

Gymnopus dryophilus, 50; lachnophyllus,
50; platyphyllus, 50; radicatus, 50;
velutipes, 50

Gymnosporangium effusum, 48; ger-
minale, 48; globosum, 48; guaraniticum,
132; Juniperi-virginianae, 48; Nidus-
avis, 49

Gynandropsis, 93

Habenaria maculosa, 166
Hapalopilus gilvus, 49; rutilans, 49
Haplographium, 71; echinatum, 71
H/rper, R, a. The evolution of cell
types and contact and pressure re-
sponses in Pediastrum, 2, 210
Harper, Roland M. The vegetation

of the Hempstead Plains, 9, 262
Harris, J. Arthur. On the osmotic
concentration of the tissue fluids of
desert Loranthaceae, 8, 307
Hawaiian flora. The origin of the, 2, 90
Hazen, Tracy E. The trimorphism and

insect visitors of Pontederia, 9, 459
Helianthemum, 274

Helianthus angustifolius, 277; annuus,

158, 342, 345
Helotium citrinum, 48
Helvella esculenta, 178; lacunosa, 48
Hemaris thysbe, 479
Hemidiodia ocimifolia, 156
Hempstead Plains, The vegetation of the,

9, 262
Hepatica, 120
Hesperocnide, 95

Heteranthera reniformis 483; zosterae-

folia, 483
Hevea brasiliensis, 70
Hewittia sublobata, 60
Hexagona alveolaris, 49
Hibiscus, 348; S3^riacus, 113, 348
Hippocratea volubilis, 161
Holcus halepensis, 135; Sorghum, 135
HoLLiCK, Arthur. Fossil plants from

Porto Rico, 2; Torrey Botanical Club

reminiscences, 29
Hopea Pierrei, 70

Horne, W. T., F. J. Seaver and. Life-
history studies in Sclerotinia, 9, 202

Horticultural journalism in the United
States, Early, 8, 79

Houstonia longifolia, 275

Howe, Marshall A. Bermuda Algae, 2

Hydnocystis Thwaitesii, 59

Hydnoporia fuscescens, 49

Hydnum ochraceum, 49

Hydrocotyle australis, 148

Hydrodictyon, 211

Hygrophorus conicus, 177; hypothejus,
177; marginatus, 177; parvulus, 177;
pratensis albus, 177; pratensis cinereus,
177

Hymenaea Courbaril, 167
Hypericum adpressum, 277; canadense,
278

Hypholoma, 177; appendiculatum, 50;

Candolleanum, 50; cernuum, 177;

fasciculare, 176; instratum, 177: rugo-

cephalum, 50; sublateritium, 176
Hypomyces hyalinus, 48
Hypoxis, 280; hirsuta, 274, 283
Hypoxylon, 67; anthracodes, 67; coc-

cineum, 48; efifusum, 67; marginatum,

67; rubiginosum, 67
Hyptis, 106; capitata, 154; lantanifolia,

154; pectinata, 153; radiata, 154;

suaveolens, 153, 154
Hysterium confluens, 64; rufulum, 58, 64

Ibidium cernuum, 278; gracile, 275, 280

Ichthyomethia Piscipula, 116

Ilicioides mucronata, 299

Indigofera suffruticosa, 116

Individuality of the bean-pod as com-
pared with that of the bean-plant,
The, 2, 207

Inheritance of height in peas, 9, 316

490 Semi-centennial of Torrey Botanical Club

Inheritance studies in Pisum, iii. The

inheritance of height in peas, 316
Inocybe, 50, 178; decipiens, 178; infehx,

177; infida, 178, 179
Insect visitors of Pontederia, Trimor-

phism and, 9, 459
lonactis, 283; hnariifolius, 274
Ipomoea, 60; acuminata, 110, 151;

Carolina, 152; cathartica, no, 151;

Learii, no; mutabihs, no; stolonifera,

no; triloba, 151
Iresine angustifolia, 142; Celosia, 127,

142; celosioides, 142; elatior, 142;

paniculata, 127, 142
Irpiciporus lacteus, 49; mollis. 49
Isaria Anisopliae, 57
Isle of Pines, Cuba, The flora of the, 7
Isnardia repens, 148

Jacquemontia, 95; nodifiora, 105, 133;

tamnifolia, no
Jambosa vulgaris, 148
Jambos Jambos, 148

Jatropha Curcas, 164; gossypifolia, 164;

Manihot, 130
Johnston, J. R., J. C. Arthur and.

Uredinales of Cuba, 2, 97
Joinvillea, 95

Journalism in the United States, Early

horticultural, 8, 79
Juglans cinerea, 296

Juncus acuminatus, 277; canadensis, 277;
Greenei, 274, 278, 283

Klebahnia Bidentis, 134
Kneiffia, 277; linearis, 277
Kniphofia, 470

Kuehneola Fici, 112; Gossypii, 113;

malvicola, 113
Kyllinga, 139

Lablab vulgaris, 129
Laccaria laccata, 50; striatula, 50
Lachnea scutellata, 48
Lachnocladium Micheneri, 49; Schweinit-
zii, 49

Lachnorhiza piloselloides, no

Lactaria glyciosma, 50; hygrophoroides,

50; piperata, 50; rimosella, 50; rufa,

176; scrobiculata, 50; subdulcis, 50;

torminosa, 177; uvida, 177; varia, 50;

vellerea, 176
Lactuca intybacea, 170
Lamellae in Schizophyllum commune.

The origin and development of the,

8, 326

Lantana Camara, 152; involucrata, 123,
152; odorata, 152; reticulata, 152, 153;
trifolia, 152

Lasiacis, 126

Lawsonia inermis, 63

Lecanium, 70

Lechea maritima, 274; villosa, 274
Lentinus strigosus, 50
Lenzites betulina, 49
Leonotis, 106; nepetaefolia, 155
Leotia lubrica, 48
Lepiota, 50

Leptochloa domingensis, 165
Leptoniella, 50

Lespedeza angustifolia, 274, 283; capitata,
280; capitata sericea, 274, 283

Levine, Michael. The physiological
properties of two species of poisonous
mushrooms, 7, 176

Libertella Fici, 69

Life-history studies in Sclerotinia, 9, 202
Life of Coe Finch Austin, A sketch of the,
2, 31

Limnanthemum, 104; Grayanum, 106, 140
Linaria vulgaris, 279

Linum fioridanum, 274; intercursum,
274, 283; medium, 274; striatum, 277

Lippia dulcis, 123, 152; stoechadifolia, 152

Litsea, 61; glutinosa, 72

Lloyd, Francis E. The effect of acids
and alkalies on the growth of the pro-
toplasm of pollen-tubes, 8, 84

Lobelia Nuttallii, 278

Lonchocarpus campestris, 116; latifolius,
116

Long Island, A preliminary report on the

Russulae of, 8, 301
Long Island, The vegetation of Mon-

tauk, 9

Long Island, The vegetation of the
Hempstead Plains, 9, 262

Loranthaceae, The osmotic concentra-
tion of the tissue fluids of desert, 8, 307

Lucuma nervosa, 169

Ludwigia alternifolia, 278

Lycium carolinianum, 155

Lycoperdon gemmatum, 50

Lycopodium adpressum, 278; lucidulum,
297; Phlegmaria, 96; serratum, 96;
volubile, 96

Lycopus, 277

Lygodium, 27

Lysiloma bahamensis, 117; tergemina, 117
Lysimachia terrestris, 277
Lythrum, 478, Salicaria, 459, 462

MacDougal, D. T. Comparative cul-
tures of seed-plants in desert valley,
desert mountain, and coastal locations, 9

Macrophoma, 68; Musae, 68

Macropodia fusicarpa, 48

Macrosporium Brassicae, 72

Malache scabra, 113

Malva rotundifolia, 351

Malvastrum corchorifolium, 148; coro-
mandelianum, 148

Malvaviscus Sagreanus, 113

Mangifera indica, 60

Index

491

Manihot Manihot, 130; utilissima, 68, 130

Marasmius archyropus, 50; caryophyl-
leus, 50; confluens, 50; elongatipes,
50; insititius, 50; perforans, 50; prae-
acutus, 50; resinosus, 50; Rotula, 50;
siccus, 50; subnudus, 50

Marattia Douglasii, 96

Marchantia polymorpha, 324

Matteuccia Struthiopteris, 297

Medicago sativa, 128

Medsger, Oliver P. Two months in
the Southern Catskills, 2, 294

Megaceros, 92

Meibomia Scorpiurus, 130, tortuosum,
130

Melampsora, 107
Melampsoridium, 107
Melanoleuca, 50

Melanthera brevifoHa, 135; hastata cuben-
sis, 135

MeHola, 59; amphitricha, 50; Arundinis,
, 60; Desmodii, 60; Litseae, 61; Mangif-

erae, 60; quadrispina, 60; substenos-

pora, 60
Melosira, 340

Melothna guadalupensis, 134
MeruHus, 326

Mesophaerum capitatum, 154; lantani-
foHum, 154; pectinatum, 153; rugosum,
154; suaveolens, 153, 154
Metarrhizium, 57; AnisopHae, 57
Metastelma, 151; penicillatum, 105,
151

Method of teaching economic botany,

A, 3, 52
Metrosideros, 95
Mezoneuron, 95

Microlepia strigosa, 96; tenuifoHa, 96
Micromycetous Fungi, PhiHppine, 3, 56
Mimosa asperata, 119; pigra, 119
Mitracarpum, 156
Monadelphus illudens, 50
Montauk, Long Island, The vegetation
of, 9

Morchella esculenta, 48
Morus alba, 62

MuRRiLL, William A. Collecting Fungi
at Delaware Water Gap, 7, 48

Musa paradisiaca, 69; sapientum, 66

Mycosphaerella, 65; Alocasiae, 66; Fra-
gariae, 65; Musae, 66; Pericampyli, 66

Myoporum, 95

Myrica carolinensis, 274

Naemospora, 69; Fici, 69
Nama, 95

Naucoria firma, 177
Nephlyctis transformans, 123
Neurolaena lobata, 159, 160
Nigella, 351, 354; hispanica, 351
Nigredo Caladii, 49; columbiana, 135;
Houstoniata, 49

Nummularia, 67; anthracodes, 67
Nuttall, from the Mississippi River to
the Columbia River in 1834, The route
taken by Capt. Nathaniel J. Wyeth
and Mr. Thomas, 8
Nyssa, 276

Observations on the development of

Peridermium Cerebrum, 7, 253
Ochrosia, 95
Odontoloma repens, 96
Oenothera, 277, 346; biennis, 279
Oidium, 71; Oxalidis, 71
Olive, E. W. A cotton-rust epidemic

in Texas, 2
Olyra latifolia, 105, 136
Omphalopsis campanella, 50; Fibula, 50
Onoclea sensibilis, 277
On the osmotic concentration of the

tissue fluids of desert Loranthaceae, 307
Oospora Destructor, 57
Operculina dissecta, 169
Ophioglossum pendulum, 96
Oplismenus compositus, 60
Opuntia, 28
Orbilia chrysocoma, 48
Origanum vulgare, 298
Origin and development of the lamellae

in Schizophyllum commune, 8, 326
Origin of the Hawaiian flora. The, 2, 90
Osmotic concentration of the tissue fluids

of desert Loranthaceae, The, 8, 307,
Osmunda, 469; cinnamomea, 277; re-

galis, 277

Oxalis, 45, 139; corniculata, 71; repens, 71
Oxy coccus macrocarpus, 277

Pallavicinia, 92

Panaeolus, 183-189, 191-193, 195, 196,

198, 199; papilionaceus, 179; retirugis,
50, 177, 181, 192-195, 198, 199, 201;
venenosus, 179-182, 187, 188, 190-197,

199, 201
Pandanus, 95
Panellus stypticus, 50

Panicum, 126; barbinode, 125; fascicu-
latum, 136; sanguinale, 137; trichoides,
136; virgatum, 277

Panus, 326

Papilio polyxenes, 479; turnus, 479
Parodie la, 62; grammodes, 62; perispori-

oides, 62; pumila, 62; puncta, 62
Parosela domingensis, 119
Parthenocarpy in cucumbers, 9
Paspalum conjugatum, 165; virgatum,

136

Passifiora rubra, 162
Patouillardiella guaranitica, 132
Paxillus involutus, 50
Peas, Inheritance of height in, 9, 316
Pediastrum, 210, 212-214, 222, 223, 235,
239; angulosum, 228; asperum, 211-

492 Semi-centennial of Torrey Botanical Club

213, 223-227, 234, 236; Boryanum,
211, 212, 214, 218, 222-225, 228, 234;
clathratum, 211, 214, 223-227, 233,
234, 236, 237; diodon, 223; diodon
reticulatum, 223; duodenarium, 214,
216; duplex, 223, 225; duplex reticu-
latum, 223, 224; Ehrenbergii, 211, 229,
230;'incisum, 230; incisum Rota, 230;
integrum, 213, 214, 229, 234; lobatura,
230; muticum, 213; pertusum, 223,
228; pertusum asperum, 224; pertusum
clathratum, 223; Rotula, 211, 230, 231;
simplex, 211, 212, 214-222, 227, 228,
231, 234, 237; Sturmii, 214, 220, 221;
triangulum, 211, 214-218, 231, 234;
triangulum angustum, 216, 217; trian-
gulum latum, 216, 217; tricornutum,
229

Pediastrum, The evolution of all types
and contact and pressure responses in,
210

Pediastrum simplex. Contact and pres-
sure reactions in, 2
Pellaea ternifolia, 96
Pellia epiphylla, 324

Penicillium Anisopliae, 57; echinatum, 71
Peniophora cinerea, 49
Pericampylus incanus, 66
Peridermium, 253, 259; Cerebrum, 253,

256-258, 260, 261; fusiforme, 260
Peridermium Cerebrum, Observations on

the development of, 7, 253
Perrottetia, 95

Persicaria, 279; punctata, 141
Pestalozzia, 71; palmarum, 71
Peziza, 59; badia, 48

Phakopsora Aeschynomenis, 112; Vignae,

112; Vitis, III
Pharbitis cathartica, 110
Phaseolus, 129; odoratus, 85; vulgaris, 129
Phegopteris punctata, 96
Philibertella clausa, 150
Philippine Micromycetous Fungi, 3, 56
PhoUota autumnalis, 178; Johnsoniana,

50; mutabilis, 50; praecox, 50
Phoma, 68; herbarum, 68; Musae, 68
Phoradendron, 307, 311, 313-315; cali-

fornicum, 310, 315; Coryae, 311, 315;

macrophyllum, 311, 314, 315; mac-

rophyllum Jonesii, 313, 314
Phragmidium, 107, 121, 122
Phragmites, 60
Phycioides tharos, 479
Phyllachora, 63; circinata, 63; Fici-

fulvae, 63, 64; Fici-minahassae, 63;

Kaernbachii, 63, 64; Merrillii, 63;

Minahassae, 64
Phyllactinia, 61; antarctica, 61; Can-

dollei, 61; guttata, 61, suffulta, 48, 61
Phyllostegia, 95

Phyllosticta, 68; Brideliae, 68; cocophila,
68

Physalis peruviana, 46

Physiological properties of two species

of poisonous mushrooms. The, 7, 176
Physopella Aeschynomenis, 112; Arto-

carpi, 164; concors, 112; Fici, 112;

ficina, 112; Vitis, 11 1
Piaropus, 461, 464, 466, 467; azureus,

459, 466, 468; crassipes, 460
Picea rubra, 297

Pieris, 480; Mariana, 273, 277, 283;

Rapae, 479
Pinanga, 71

Pinus Banksiana, 253, 256; divaricata;

254, 256; rigida, 253, 256, 261, 271, 273,

276; virginiana, 253, 256, 261
Pipturus, 95
Pisonia aculeata, 161
Pissodes Strobi, 296
Pisum, Inheritance studies in, 316
Pithecoctenium echinatum, 123
Pithecolobium apoense, 60; tortum, 117
Pittosporum, 94
Pityrogramma calomelanos, 164
Plant tissues, Determination of acidity in,

8, 241
Plasmopora viticola, 48
Plectranthus, 95
Plectronia, 95
Pleuropus unitinctus, 50
Plumiera emarginata, no; obtusa, no;

rubra, no

Pluteus cervinus, 50; granularis, 50

longistriatus, 50
Poa pratensis, 181
Poinciana pulcherrima, 118
Poinsettia heterophylla, 131
Poisonous mushrooms. The physiological

properties of two species of, 7, 176
Polites peckius, 479

Pollen-tubes, The efifect of acids and
alkalies on the growth of the proto-
plasm of, 8, 84

Polydesmus exitiosus, 72

Polygala cruciata, 277; Nuttallii, 274;
polygama, 274, 283; viridescens, 274

Poh'gonum punctatum, 141; sagittatum,
277

Polypodium Hookeri, 96; lineare, 96;

samoense, 96; Spectrum, 96; tamaris-

cinum, 96
Polyporus elegans, 49; Polyporus, 49
Polythelis fusca, 49
Polytrichum juniperinum, 267, 275
Pontederia, 459, 461, 466, 467, 469, 470,

474, 478-480, 483, 484; azurea, 461;

cordata, 461-463, 466, 473, 475. 477.

478, 483, 484; cordata lancifolia, 466,

468; crassipes, 459-461; montevideen-

sis, 461, 466-469, 471, 478, 483; rotun-

difolia, 461, 483
Pontederia, The trimorphism and insect

visitors of, 9, 459

Index

493

Populus, 314, 315; grandidentata, 296;
tremuloides, 273/ 296; Wislizenii, 314

Poria medullapanis, 49; vaporaria, 49

Poronidulus conchifer, 49

Porto Rico, Fossil plants from, 2

Potamogeton, 278

Potentilla canadensis, 274

Preissia quadrata, 323

Preissia quadrata, Centrosomes in fer-
tilization stages of, 2, 323

Preliminary report on the Russulae of
Long Island, A, 8, 301

Pritchardia, 95

Priva lappulacea, 152

Prosopis velutina, 310

Prospodium, 122, 149, 169; Amphilophii,
123; appendiculatum, 120, 156; Lip-
piae, 123; plagiopus, 98, 104, 120;
tuberculatum, 123

Protoplasm in pollen-tubes. The effect of
acids and alkalies on the growth of the,
8, 84

Prunus Persica, 120
Psathyrella disseminata, 50
Pseudomorus, 95
Psoralea tenuiflora, 281
Pteris excelsa, 96

Puccinia, 107, 108, 121, 122, 166, 170;
abrupta, 158; Adenocalymnatis, 155;
aequinoctialis, 155; Amphilophii, 123;
Anthephorae, 104, 137; appendiculata,
120; Arechavelatae, 144; Asteris, 104;
barbatula, 144; Blechi, 156; canalic-
ulata, 139; Cannae, 141; Cenchri,
136; Chaetochloae, 137; Chaseana,
137; compacta, 150; concrescens, 150;
Conoclinii, 158; coronata, 138; cras-
sipes, 151, 152; cuticulosa, 156; Cynan-
chi, 106, 151; Cyperi, 139; deformata,

105, 136; elegans, 124; Eleocharidis,
139; Eleutherantherae, 159; evadens,
157; exitiosa, 124; Fuirenae, 140;
fuscella, 157; globosipes, 155; Gonolobi,

106, 150, 151; Gouaniae, 105, 145;
graminis, 104, 138; Helianthi, 158;
heterospora, 105, 146; Huberi, 136;
Hydrocotyles, 148; Hyptidis, 154;
Impatientis, 49; inflata, 143; insititia,
154; invaginata, 146; Johnstonii, 149;
Lantanae, 152; lateripes, 156; lateritia,
156; Leonotidis, 155; Lippiae, 123;
Ludwigiae, 148; macropoda, 142; mal-
vacearum, 148; medellinensis, 153, 154;
megalospora, 152; obliqua, 105, 151;
Osmorrhizae, 49; plagiopus, 104, 120,
121; poculiformis, 104, 138; Polygoni-
amphibii, 141; Psidii, 148; purpurea,
135; Raunkiaerii, 142; Rhamni, 138;
Rivinae, 97, 106, 142, 143; rosea, 159;
Ruelliae, 156; salviicola, 153; Scirpi,
106, 140; Scleriae, 162; scleriicola,
140; Smilacis, 141; solida, 104, 159,

160; Sorghi, 139; sphaerospora, 151;

striolata, 142; substriata, 137; Syne-

drellae, 104, 159; transformans, 123;

Tridacis, 159; tuberculata, 123; Ur-

baniana, 153; urticata, 49; valida, 166;

Vernoniae, 157; Violae, 49; Xanthii,

156; Zcrniae, 143
Pucciniastrum, 107
Pucciniosira pallidula, 160
Pueraria, 56; sericantha, 56; Thunbergii,

56

Pycnoporus cinnabarinus, 49
Pyropolyporus igniarius, 49

Quercus, 296, 311; arizonica, 311; Emoryi,
311; heterophylla, 261; hypoleuca,
311; ilicifolia, 260, 274, 282; mary-
landica, 260, 270, 273; oblongifolia,
311; prinoides, 273, 282, 283; stellata,
270, 273

Raillardia, 93

Rajania cordata, 165

Ramularia Fragariae, 65; Tulasnei, 65

Ravenelia, 107, 108, 116-118, 167;
cubensis, 118; Humphreyana, 118;
Indigoferae, 116; Lonchocarpi, 116;
Lysilomae, 117; Piscidiae, 116; Pithe-
colobii, 117; portoricensis, 105, 118;
pulcherrima, 118; siliquae, 117

Recent botanical exploration in Colom-
bia, 8, 39

Reminiscences, 2, 22

Reminiscences, Torrey Botanical Club,
2, 29

Report on the Russulae of Long Island,

A preliminary, 8, 301
Reynoldsia, 95
Rhexia virginica, 277
Rhus copallina, 274, 280, 283; radicans,

277; Vernix, 277
Rhynchospora alba, 277; glomerata, 278
Rhytidhysterium guaraniticum, 59; ja-

vanicum, 59
Riccardia, 92

Richards, Herbert M. Determination

of acidity in plant tissues, 8, 241
Riella Clausonis, 323

Rivina humilis, 142; octandra, 97, 106,

142
Robinia, 44

Rocky Mountains and adjacent plains,

Flora of the, 7
Rosa palustris, 277
Rosellinia Bambusae, 64, 66
Rubus cuneifolius, 274; hispidus, 277;

strigosus, 300
Ruellia, 156

RusBY, H, H. Recent botanical col-
lecting in the Republic of Colombia, 39

RusBY, H. H. Recent botanical explo-
ration in Colombia, 8

494 Semi-centennial of Torrey Botanical Club

Russula, 301-303, 305, 306; aeruginea,
305; albella, 301, 303; albida, 301, 303;
anomala, 302, 303; betulina, 302, 303;
blanda, 302, 303; cinerascens, 305;
compacta, 301, 303; crustosa, 301, 304;
decolorans, 302, 305; delica, 305; densi-
folia, 305; Earlei, 301, 304; emetica,
305; flava, 50, 306; flaviceps, 302, 304;
flavida, 301, 304; foetens, 50, 306; fra-
giliformis, 306; fragilis, 306; hetero-
phylla, 306; humidicola, 304; lepida,
306; magnifica, 302; Mariae, 301, 302,
304; modesta, 304; obscura, 302, 306;
olivascens, 306; pectinata, 302, 306;
purpurina, 306; pusilla, 302, 304;
Queletii, 176; rubescens, 302, 306;
rubra, 306; sanguinea, 176; serissima,
302, 304; sororia, 306; sphagnophila,
301, 302, 305; squalida, 177; subolivas-
cens, 306; subvelutina, 301, 305; un-
cialis, 301, 305; variata, 301, 305;
vinacea, 305; virescens, 50

Russulae of Long Island, A preliminary
report on the, 8, 301

Rydberg, p. a. Flora of the Rocky
Mountains and adjacent plains, 7

Rynchospora, 127; distans, 126

Saccharum, 60; officinarum, 69; spon-

taneum, 69
Sagittaria latifolia, 278
Salix humilis, 273; tristis, 273
Salvia, 106; occidentalis, 153
Sambucus canadensis, 277
Santalum, 95
Sapindus Saponaria, 60
Sapota Achras, 169
Sarcoscypha occidentalis, 48
Sarothra gentianoides, 275
Savia sessilifiora, 168
Scaevola, 95
Scenedesmus, 211
Schizaea, 26; australis, 96
Schizophyllum, 326, 331, 332; commune,

327. 332

Schizophyllum commune. The origin and
development of the lamellae in, 8, 326
Schizophyllus, 326; alneus, 50
Schizostachyum, 67
Scirpus capitatus, 140; lacustris, 140
Scleria lithosperma, 127; pauciflora, 274,

283; verticillata, 26, 140
Scleroderma aurantium, 50
Sclerotinia, 202, 206; Geranii, 205
Sclerotinia, Life-history studies in, 9, 202
Sclerotium Erysiphe corylea, 61; sufful-
tum, 61

Scott, James G. Early horticultural
journalism in the United States, 8, 79

Seaver. F. J., and W. T. Horne. Life-
history studies in Sclerotinia, 9, 202

Securidaca, 45

Seed-plants in desert valley, desert
mountain, and coastal locations, Com-
parative cultures of, 9

Sempervivum, 350

Senecio alpinus, 343; Fuchsii, 339;

Jacobaea, 339; nemorensis, 339
Septocylindrium suspectum, 57
Sericocarpus, 281; linifolius, 274, 282, 283
Setaria setosa, 137
Sicyos angulata, 469

Sida angustifolia, 147; glutinosa, 147;

procumbens, 147; spinosa, 147
Sideroxylon foetidissimum, 149
Sisyrinchium, 82, 95, 274; angustifolium,

82; Bermudiana, 82, 83; iridioides, 82,

83

Sisyrinchium Bermudiana Linnaeus, 2, 82
Six misunderstood species of Amanita,

2, 246
Skier ka, 133

Small, John K. The ferns of tropical

Florida, 2
Smilax havanensis, 141
Smithia ciliata, 62
Solanum torvum, 162

Solidago nemoralis, 274; puberula, 274,

280; rugosa, 278; Virgaurea, 339
Some factors influencing the stimulative

action of zinc sulphate on the growth

of Aspergillus niger, 2
Some observations on the development of

Peridermium Cerebrum, 253
Sorbus americana, 297; sambucifolia, 297
Sorghastrum, 276; nutans, 274, 283
Sorghum halepense, 135; vulgare, 135
Sparganium, 277
Spermacoce, 156
Sphacele, 95

Sphaerella Fragariae, 65; Musae, 66
Sphaeria amphitricha, 59; anthracodes,

67; Fragariae, 65; fragariaecola, 65;

grammodes, 62; marginata, 67; rubi-

ginosa, 67

Sphaerophragmium, 108, 125; Dalbergiae,
124

Sphaeropsis Musarum, 68
Sphagnum, 278

Spiraea latifolia, 277; tomentosa, 277

Spirodela, 16

Sporobolus indicus, 126

Stachytarpheta jamaicensis, 153

Staphylea trifolia, 348

Statistical studies in Cichorium, 8

Statistical studies of flower number per
head in Cichorium Intybus; Kinds of
variability, heredity, and effects of
selection, 334

vSteinberg, R. a. a study of some fac-
tors influencing the stimulative action
of zinc sulphate on the growth of
Aspergillus niger, i ; The effect of the
presence of zinc in the cultural flasks.

Index

495

287; Some factors influencing the
stimulative action of zinc sulphate on
the growth of Aspergillus niger, 2

Stellaria media, 347

Stenolobium stans, 120, 124

Stereum complicatum, 49; frustulosum,
49; hirsutum, 49; lobatum, 49

Stigmaphyllon lingulatum, 144; peri-
plocifolium, 143; reticulatum, 144;
Sagraeanum, 144

Stigmatea Fragariae, 65

Stout, A. B., and Helene M. Boas,
Statistical studies of flower number
per head in Cichorum Intybus: Kinds
of variability, heredity and effects of
selection, 334

Stout, A. B. Statistical studies in
Cichorium, 8

Strobilomyces strobilaceus, 50, 178

Stromatinia, 206; Geranii, 206

Strongylodon, 95

Stropharia semiglobata, 50

Strophostyles, 129

Studies in Cichorium, Statistical, 8, 334
Studies in Sclerotinia, Life-history, 9,
202

Study of some factors influencing the
stimulative action of zinc sulphate on
the growth of Aspergillus niger, i.
The effect of the presence of zinc in the

■ cultural flasks. A, 287

Symphogyna, 92

Synchytrium, 56, 57; Puerariae, 56
Synedrella nodiflora, 159
Syntherisma sanguinalis, 137, 279

Tabanus giganteus, 480
Tamarindus indica, 59
Tanacetum corymbosum, 349
Taxus minor, 298

Taylor, Norman. The vegetation of

Montauk, Long Island, 9
Tecoma, 163; lepidota, 120, 122; pen-

taphylla, 120, 162; stans, 120, 124
Tetraplasandra, 95
Texas, A cotton-rust epidemic in, 2
Texas, Weather conditions and crop

diseases in, 9, 74
Thalictrum, 120

Thelephora multipartita, 49; regularis, 49
Tissue fluids of desert Loranthaceae,
The osmotic concentration of the 8, 307
Tithymalus, 128

Torrey Botanical Club, History of the, i
Torrey Botanical Club reminiscences,
2, 24, 29

Tournefortia hirsutissima, 162; peru-
viana, 162; volubilis, 134
Tranzschelia punctata, 49, 119
Tremella lutescens, 49
Tremellodon gelatinosum, 49
Triadenum virginicum, 277

Trichobasis euphorbiaecola, 105, 131;

labiatarum, 106
Tricholoma ustale, 177
Trichomanes meifolium, 96; parvulum, 96
Tridax procumbens, 160
Trillium undulatum, 298
Trimorphism and insect visitors of Pon-

tederia. The, 9, 459
Triticum sativum, 138; vulgare, 138
Triumfetta semitriloba, 160
Tryblidiella, 58,64; Balansae,59;rufula,58,

64

Tryblidium guaraniticum, 59
j Tuberculina, 77, 78

; Two months in the southern Catskills, 2,294
Tylopilus felleus, 50

Tyromyces chioneus, 49; lacteus, 49;
semipileatus, 49; semisupinus, 49;
Spraguei, 49

United Statis, Early horticultural journ-
alism in the, 8, 79
Uredinales of Cuba, 2, 97
Uredo, 98, 99, 107, 108, 163, 164, 167;
Adenocalymnatis, 155; Aeschynomenis,
112; Anthephorae, 137; Arachidis, 167;
Artocarpi, 164; bauhiniicola, 167;
Cabreriana, 167; Cephalanthi, 170;
Cherimoliae, 166; Coccolobae, 164;
concors, 112; cristata, 131, 132; cutic-
ulosa, 156; Dalbergiae, 125; Des-
modii-tortuosa, 129, 130; Dioscoreae,
165, 166; Erythroxylonis, 168; Fici,
112; ficicola, 112; ficina, 112; Fuirenae,
165; gemmata, 133; Gossypii, 113;
Gouaniae, 146; Gymnogrammes, 164;
gynandrearum, 166; Hibisci, 113;
Hymenaeae, 167; ignobilis, 126; jatro-
phicola, 164; Lucumae, 169; malvicola,
113; moricola, 112; nigropunctata, 166;
notata, 114; Operculinae, 169; paspali-
cola, 165; proximella, 170; Sapotae, 169;
Saviae, 168; Sissoo, 124, 125; Steven-
siana, 165; striolata, 142; superior, 165;
Vignae, 112; Zorniae, 143
Uromyces, 107, 108, 132, 166, 167; ap-
pendiculatus, 105, 129; bidenticola,
134; Bidentis, 135; Celosiae, 127;
columbianus, 135; Commelinae, 127;
cristatus, 108; Cupaniae, 131; Doli-
choli, 129; dolichosporus, 134; Erag-
rostidis, 126; Euphorbiae, 131; gem-
matus, 105, 133, 145; Hedysari-pani-
culati, 129; Hellerianus, 134; Howei,
133; ignobilis, 126; insularis, 128;
jamaicensis, 128; Janiphac, 130; lep-
todermus, 125; Lucumae, 169; major,
126; Medicaginis, 128; Medicaginis-
falcatae, 128; Neurocarpi, 128; pro-
eminens, 106, 131; Puerariae, 56; Rhyn-
cosporae, 126; Scirpi, 149; Scleriae,
127, solidus, 129

496 Semi-centennial of Torre y Botanical Club

Uromycladium, io8, 119; cubense, 119
Uropyxis, 107, 121, 122, 169
Utricularia, 55

Vaccinium corymbosum, 277
Vachellia Farnesiana, 117
Vaginata agglutinata, 50; plumbea, 50
Valerianodes jamaicensis, 153
Vallesia, 95

Vegetation of Montauk, Long Island,
The, 9

Vegetation of the Hempstead Plains,

The, 9, 262
Venenarius . cothurnatus, 50; Frostianus,

50; muscarius, 50; phalloides, 50;

rubens, 50
Verbena hastata, 479
Vernonia menthaefoHa, 157; novebora-

censis, 277
Viburnum alnifolium, 298; dentatum,

277; odoratissimum, 60 *
Vigna, 129; repens, 129; vexillata, 129
Viguiera helianthoides, 158
Viola fimbriatula, 274; lanceolata, 275,

278; pedata, 274, 280, 283; primuli-

folia, 277
Viscum, 95
Vitis vinifera, iii
Vittaria elongata, 96

Volvaria gloiocephala, 179; speciosa, 179

Warren, P. A., A. F. Blakeslee and.

Parthenocarpy in cucumbers, 9
Weather conditions and crop diseases in

Texas, 9, 74
White, O. E. Inheritance of height in

peas, 9; Inheritance Studies in Pisum

III. The inheritance of height in

peas, 316
Wiesnerella, 91
Wikstroemia, 95
Wilkesia, 93

Wissadula periplocifolia, 147

Wyeth and Mr. Thomas Nuttall from
the Mississippi River to the Columbia
River in 1834, The route taken by
Capt. Nathaniel J., 8

Xanthium, 139, 157; longirostre, 156,

157; saccharatum, 157
Xenoglossa pruinosa, 480
Xylaria Hypoxylon, 48; polymorpha, 48
Xylocopa, 480; virginica, 480
Xyris, 278

Zea Mays, 139

Zinc sulphate on the growth of Asper-
gillus niger, Some factors influencing
the stimulative action of, 2, 287

Zornia diphylla, 143