Wet
wpe
‘sonyad
'
1
a 4
isi
‘
Ar
isa
rf 7
/
» qf
eee
i
'
\pha af stand
, efi ote tay
ps : sy isdety Tebpbateh Getant
“mee vee - ate enlta geen: sifellay sae
Aci f Tewepheae V ete bewaee!
: is piabey egy tes pnt et yee ts
eee gop aatets PAG ts hao: Haid
ae Ch pate Ld Rg At Ta sats pa et ay!
3 ail Leip tegaeiae CPU eam Bates!
he hae a go Celaya betaine
J ; eaonetads ey
aa i day, sor pglabepprater
A pg Bate ete pea
ya foratgnctbe Heth
: miehitereie &
; aeatetera tare’
: a's whale he ow ‘
‘ VAG be (ete row eters
Leyte ca aka deiaial, gates
a : oe oty at tere the were
vate hae g ttehe a bpei eae ey tae
ae Lied te te eee
a State (bs ale genre
to Hy : pha y
Ce tae 1 rr
us : 1 [ily aha phase ab ee
,
’
t
W
.
73
7 i 4 ds s/s
1 : , ‘ t
F ae ape
{
4 : de F 22 075
74 . aoe
“ ‘ 3
we '
A ae ’
‘ ' bia
4 i :
‘ . 4 thet
a | ia #
’ i ; }
ij :
fie?
bag t
‘
: ; a6
, fo f
‘ »t a
y ‘ta G44
, oo: peug (bos
el { sity fs
7 re
: i :
7 j ; j ,
; ' rot tl,
4 * 5
\ - : j
a ae
i 1%
itp {4
: i
ae |
' Soe
.
; ats rp ,
: . . 4 ag ,
Fi ‘ 7
[eeeda yes
aren
head
be pe, ) bei k
ae ie Pasi
rigs 5 1
; aye
{eres
, § aS
cM eos
a ee
f
eee
(Photograph by Iay-Hart, N. Y.)
AVEN NELSON
Frontispiece to Volume 7, Maprono.
MADRONO
A WEST AMERICAN JOURNAL OF
BOTANY
VOLUME VII
1943 - - 1944
y on Ma
38
& Gy
: National Muse
Mf
&
ae
Published by the
California Botanical Society, Inc.,
4004 Life Sciences Building, University of California, Berkeley
North Queen Street and McGovern Avenue, Lancaster, Pennsylvania
Board of Editors
Hersert L. Mason, University of California, Berkeley, Chairman.
LeRoy Aprams, Stanford University, California.
EpGar ANDERSON, Missouri Botanical Garden, St. Louis.
Lyman Benson, Pomona College, Claremont, California.
Hesrert F. Coperann, Sacramento Junior College, Sacramento, California.
Ivan M. Jounston, Arnold Arboretum, Jamaica Plain, Massachusetts.
Mivprep E. Marutas, 2851 North Lake Avenue, Altadena, California.
Bassett Macuire, New York Botanical Garden, New York City.
Marion Ownsey, State College of Washington, Pullman.
Secretary, Editorial Board
ANNETTA M. Carter, Department of Botany, University of California, Berkeley.
Business Manager
Wititiam Hiesey, Carnegie Institution of Washington, Stanford University,
California.
To Aven Nelson, founder of the Rocky Mountain Herbarium,
President, Emeritus, of the University of Wyoming, now in his
eighty-fifth year—we dedicate this seventh volume of Madrojfio.
His is a life, rich in experience, rich in accomplishment! He
pioneered both in the field of botany and horticulture in the Rocky
Mountain and Great Basin regions. He effected the training and
greatly influenced the lives of many students who have taken their
places among the able botanists of our time.
Apart from his work as an educator his influence on the agri-
culture of the northern Rocky Mountains has been notable. He
wrote many horticultural bulletins issued by the state of Wyo-
ming and was among the first to urge the growing of apples as an
agricultural crop in that area.
His great ability and kindly personality are reflected in his
being appointed to serve as president of the University of Wyo-
ming during the years 1917 to 1922, and, in 1935, being elected
president of the Botanical Society of America. The genus Anel-
sonia Macbride and Payson and such species as Salia Nelsoni Ball,
Carex Nelson Mackenzie, Stipa Nelsoni Scribner, Delphinium Nel-
soni Greene and Cirsium Nelsonii (Pammel) Rydb. are among the
many plants named in his honor.
We wish him good health and many more years of usefulness
in his chosen field of study.
iii
Ryd
yn
CONTENTS
Frontispiece: Aven Nelson
Thalictrum ametrum Greene: an Interesting Nomenclatural Case.
Leon Croizat
Una Nueva Especie de Pinus Mexicano ...... sa ere esas Oe Maximino Martinez
An Investigation of the Presence of Siliceous Rods in the Secondary Wall
Oly WiOOGY, LISSUG: 42a 5 oe ciak Mae, dow dee eh lS See ies Se St eae Walter M. Schall
A New Cliff-Rose from Arizona:.......:............. Thomas H. Kearney
Hr anGcescoylrancesChs @ oe. uae uh. aw 2 oes bd ena Maks Se es John M. Tucker
| ERAS ae IS nt a mE eC 27, 63, 93, 126, 226,
chek Watherine (Crum 6.25% esi. asoans pe Guehhes a daa Ss Herbert L. Mason
The Xerophyllous Species of Philadelphus in Southwestern North America.
C. Leo Hitchcock
A New Species of Phacelia from Saline Valley, California.
Lincoln Constance
Gilia multiflora Nutt. and Its Nearest Relatives.
Thomas H. Kearney and Robert H. Peebles
A New Species of Tauschia from the State of Washington.
Mildred E. Mathias and Lincoln Constance
Notes on the Flora of the Charleston Mountains, Clark County, Nevada.
Wem aChaCedO 9 game eae oe ne Sp Reals A es Ira W. Clokey
Noteworthy Plants from Idaho.. II .................... Arthur Cronquist
The Legitimacy of Thalictrum polycarpum Wats. ....... Arthur Cronquist
Valid and Legitimate Names—and Thalictrum polycarpum S. Wats.
C. A. Weatherby
The Anatomy of Redwood Bark .....................2 04. Iroing H. Isenberg
Walham Albert Setchell’ 52. .2445.4 5) 0.008 de yes ae ands Herbert L. Mason
Vigethia, A New Genus of Compositae Based on Wyethia mexicana Watson.
William A. Weber
A Revision of the Genus Fremontia .................... Margaret Harvey
The Genus Styrax in Central and Western Texas .............. V.L. Cory
The Story of Parthenium alpinum ................... George J. Goodman
Two Types of Broad-leaf Erodium in California.
Kenneth A. Wagnon and Harold H. Biswell
Cytological Evidence for the Taxonomic Position of Schizachne purpurascens.
W.S. Boyle
‘Noteworthy Plants of Texas. II. A New Species of Peltandra.
Fred A. Barkley
A Monograph of the North American Species of Fritillaria.
Dorothy E. Beetle
he Homonym Question . 2.064056. 004.508 dees wes RE i os Leon Croizat
INOTES mala NIG WS: esate es nna ental: et cee Me Ree hee cna, ann, Racine & 160, 192,
The Taxonomy and Cytology of the Subgenus Gormania of Sedum.
Robert T. Clausen and Charles H. Uhl
Jawa Stacey, Caricolocist) -.. so. 264..6hea cu daw Bi. as John Thomas Howell
Dermatitis and Photosensitization Produced by Ptelea angustifolia.
W.C. Muenscher and Babette I. Brown
On the Shoot Apex of Chlorogalum pomeridianum (DC.) Kunth.
Clarence Sterling
Structure and Taxonomy of Taenioma, Including a Discussion of the Phy-
logeny of the Ceramiales ..............5......-.-+ George F. Papenfuss
Notes ‘on the-Algal Genus Paenioma 52 242 6 sec. 2 oe a es C.K. Tseng
Some New Laurenciae from Southern California ..... Elmer Yale Dawson
Nomina Conservanda Proposals for Ten Genera of Trees and Shrubs.
Elbert L. Little, Jr.
Forestiera in Southern and Southwestern Texas ............... V. L. Cory
Me XeCOmNOlUUMIC DV silat ys eit aac stag eh Davamen tet eee LOT De Bene 1 fhe ole as
co
ah a ae
Hen iB
a
ty
VOLUME VII NUMBER 1
MADRONO
A WEST AMERICAN JOURNAL OF
BOTANY
s ~s,
fic Co.
Hi ¢ iD e- |
! il i & be Uj Ij
\ = /
Pe fis eS -
| SA as aw we
b SS VNAL MU.
Pa a —
}
i Contents
THALICTRUM AMETRUM GREENE: AN INTERESTING NOMENCLATURAL CASE,
EEOC OUU LS ae seu ey ea eer cet emer Nard CN, ae Rt ey Sea ORE 1
Una Noeva EspPecie pe Pinus Mexicano, Maximino Martinez ........... 4,
Aw INVESTIGATION OF THE PRESENCE OF SILICEOUS Rops IN THE SECONDARY
Watt or Woopy Tissuz, Walter M. Schall ..............0 0.0.00 ..0.... 8
A New Cuirr-Rose rrom Arizona, Thomas H. Kearney
Francesco FrRANcESCHI, John M. Tucker ........000 00000 ccc ee es 18
Reviews: Maunsell Van Rensselaer and Howard EK. McMinn, Ceanothus.
Part I. Ceanothus for Gardens, Parks and Roadsides. Part II. A
Systematic Study of Ceanothus (J. Francis Macbride); Adriance S.
Foster, Practical Plant Anatomy (Ira L. Wiggins); Andrew Denny
Rodgers, John Torrey, A Story of North American Botany (Mildred
Mathias) ; S. F. Blake and Alice C. Atwood, Geographical Guide to the
Floras of the World (Herbert L. Mason); Harold N. Moldenke, EHrio-
caulaceae, Avicenniaceae, Verbenaceae in Flora of Texas (Ira L.
Wiggins)
Published at North Queen Street and McGovern Avenue,
Lancaster, Pennsylvania
January, 1943 |
MADRONO
A WEST AMERICAN JOURNAL OF BOTANY
Board of Editors
Hersert L. Mason, University of California, Berkeley, Chairman.
LeRoy Asrams, Stanford University, California.
Epcar ANpDERSON, Missouri Botanical Garden, St. Louis.
Lyman Benson, University of Arizona, Tucson.
Hersert F’. Copetanp, Sacramento Junior College, Sacramento, California.
Ivan M. Jonnston, Arnold Arboretum, Jamaica Plain, Massachusetts.
Miuzprep E. Maruias, University of California, Berkeley.
Bassett Macurre, Utah State Agricultural College, Logan.
Marion Ownsey, State College of Washington, Pullman.
Secretary, Editorial Board—EtTHEt Crum
Department of Botany, University of California, Berkeley
Business Manager—Wiut.iam HIeEsEy
North Queen Street and McGovern Avenue, Lancaster, Pennsylvania
or
Carnegie Institution of Washington
Stanford University, California
Entered as second-class matter October 1, 1935, at the post office at
Lancaster, Pa., under the act of March 3, 1879.
Established 1916. Published quarterly. Subscription Price $2.50 per year.
Completed volumes I to V inclusive, $25.00; each volume $5.00; single numbers
$0.75.
Papers up to 15 or 20 pages are acceptable. Longer contributions may
be accepted if the excess costs of printing and illustration are borne by the
contributor. Range extensions and similar notes will be published in con-
densed form with a suitable title under the general heading “Notes and News.”
Articles may be submitted to any member of the editorial board. Manuscripts
may be included in the forthcoming issue provided that the contributor pay the
cost of the pages added to the issue to accommodate his article. Reprints of
any article are furnished at a cost of 4 pages, 50 copies $4.10; 100 copies $4.50;
additional 100’s $0.85; 8 pages, 50 copies $5.95; 100 copies $6.60; additional 100’s
$1.30; 16 pages, 50 copies $8.35; 100 copies $9.35; additional 100’s $2.00. Covers,
50 for $2.75; additional covers at $1.65 per hundred. Reprints should be
ordered when proofs are returned.
Published at North Queen Street and McGovern Avenue, Lancaster,
Pennsylvania, for the
CALIFORNIA BOTANICAL SOCIETY, INC.
President: A. R. Davis, University of California, Berkeley. First Vice-
President: Palmer Stockwell, Institute of Forest Genetics, Placerville, Cali-
fornia. Second Vice-President: Reed C. Rollins, Stanford University, California.
Treasurer: William Hiesey, Carnegie Institution of Washington, Stanford Uni-
versity, California. Secretary: Clarence R. Quick, United States Department
of Agriculture, 26 Giannini Hall, University of California, Berkeley.
Annual membership dues of the California Botanical Society are $2.50,
$2.00 of which is for a year’s subscription to Madrofio. Dues should be
remitted to the Treasurer. General correspondence and applications for
membership should be addressed to the Secretary.
THALICTRUM AMETRUM GREENE: AN INTERESTING
NOMENCLATURAL CASE
Leon CroizatT
The validity of Thalictrum polycarpum S. Wats. against the
earlier 7’. polycarpum Loret, and the later T'. ametrum Greene has
been affirmed by Wheeler (Rhodora 40: 318-320. 19388) in a dis-
cussion which is exceedingly interesting under the standpoint of
nomenclature. In this discussion Wheeler raises two issues, first,
whether the parenthetic author must be indicated in every case;
second, whether an earlier name can be duplicated, and if so
under what conditions.
Lack of space makes it necessary for me to discuss here only
one of these issues. Accordingly, I shall discuss the second,
which is more important.
The first printed mention of Thalictrum polycarpum occurs in a
paper by Loret (Bull. Soc. Bot. France 6: 16. 1859). To spare
the reader the necessity of wading through Loret’s stiff French,
but doing violence to accepted bibliographical standards, I shall
quote here as if in the original my own translation of this publica-
tion. Loret states: “I have collected in a hedge at Barcellonette
(Basses-Alpes) at the end of July 1851 an interesting Thalictrum
which is quite noteworthy on account of its short-ovoid carpels,
9-12 to 14 being borne upon a single receptacle. This plant is
close to, but differs appreciably from T. Jacquinianum Koch and
T. expansum Jord. I believe this Thalictrum to be a new species,
but, fearing to augment the confusion already prevalent in this
genus, I merely bring this plant to the attention of the botanists
who may have the opportunity of collecting at Barcellonette, hop-
ing on my part to see this plant again on the spot. If I were to be
allowed to give this plant a name, I would gladly call it T. poly-
carpum or, better still, T. multiflorum (S’il m’etait réservé de lui
imposer un nom, je lui donnerais volontier celui de Th. polycarpum
ou mieux multiflorum).”
The binomials of Loret have been disregarded by practically
every author, with the exception of Lecoyer. In his monograph
of Thalictrum (Bull. Soc. Bot. Belg. 24: 78-324. 1885), Lecoyer
treats T'. polycarpum as a synonym of T'. multiflorum (op. cit., 304),
which he places in the synonymy of T. minus L. Lecoyer adds (op.
cit., 297) that T. multiflorum is “une forme non décrite” of T. minus,
believing T'.. multiflorum to be a nomen nudum or a nomen semi-
nudum which has T. polycarpum as its synonym.
Wheeler is of the opinion, on the contrary, that T’.. polycarpum
and T. multiflorum “is an illegitimate name of the type known as a
nomen provisorium,”’ without specifying which one of these two
binomials he interprets as the provisional name. I do not believe
it necessary to argue the status of these binomials at this point,
because the matter has little immediate importance. The impor-
ManproNno, Vol. 7, pp. 1-32. January 28, 1943.
2° MADRONO [Vol.7
tant side of Wheeler’s interpretation is in his statement that:
“Provisional names are not only illegitimate but are not validly
published. Since they are not validly published they cannot as earlier
homonyms, invalidate a later name”’ (italics mine).
This statement contains an unmitigated fallacy. Thalictrum
polycarpum is illegitimate because its duplicates T. minus L. (Art.
16, Art. 60[1] Amsterdam Code), and is invalid, in addition, at
least because it is published as a synonym of T. multiflorum (Art.
40). Article 61 in the current Rules states: “Even if the earlier
homonym is illegitimate, or is generally treated as a synonym on
taxonomic grounds, the later homonym must be rejected.” This
is clear enough: T. polycarpum Loret is not only illegitimate and
invalid but has been treated as a synonym of T. minus L. on taxo-
nomic grounds by Lecoyer. Obviously, T. polycarpum Loret,
1859, as an earlier homonym renders illegitimate T. polycarpum
S. Wats., 1879. The text of Article 61 and the status of the bino-
mials both of Loret and Watson precisely and absolutely contra-
dict Wheeler’s affirmations.
Since Wheeler’s error rests upon assumptions which are unfor-
tunately widespread, it is advisable to add here a few words of
comment. Many are the taxonomists who believe that a name
which is not “valid”? may be “ignored.” This belief involves a
fundamental confusion between two different concepts, which can
easily be illustrated by an example. Let us suppose that John
Doe publishes in 1940 Planta una without a Latin description.
This binomial is invalid, because Article 88 of the Amsterdam
Code requires a Latin diagnosis for a valid publication. Accord-
ingly, Jack Roe can freely use the type specimen of P. una and
propose on it in 1942 a new binomial, P. quaevis. Roe can do this
because the publication of Doe does not “‘exist’’ as valid nomen-
clature on account of the lack of a Latin diagnosis. Of course,
Roe, if he so wishes, can honor the earlier invalid binomial pro-
posed by Doe, effectively publishing P. una with a Latin diagnosis.
In this case (Art. 48), the species will be known as P. una J. Doe
in (or ex) J. Roe.
An entirely different state of affairs obtains if J. Roe attempts
to publish in 1942 a new species, naming it P. una and basing it
upon a type specimen other than the one originally used by J. Doe
in 1940 for his P. una. Such a duplication is expressly forbidden by
Article 61, as it has been seen. Under the Vienna Code (1905) and
Bruxelles Code (1910) it was not permitted to reject a well
known name, “Because of the existence of an earlier homonym
which is universally regarded as non-valid or for any other motive
either contestable or of little import’? (Art. 50, Vienna and
Bruxelles Codes). The motives behind this Article were lofty,
no doubt, but its practical application led to countless contro-
versies and abuses, because the generality embodied in the Article
was not accompanied by an elucidation of what was meant as an
1943] CROIZAT: THALICTRUM AMETRUM 3
homonym “universally regarded as non-valid,’ and what were
“motives contestable or of little import.” Naturally, everybody
thought of his own motives as being true and relevant, and of
those of his opponents as “contestable or of little import.” To
remedy this situation, the text now embodied by Article 61 was
approved by the Cambridge Congress of 1930. Space forbids my
entering into details, but I may at least point out that Miss L.
Green, who is well informed on everything that was proposed and
voted upon at Cambridge, states in her authoritative commentary
on nomenclature (Emp. For. Jour. 10: 68. 19381) that: “All later
homonyms should be rejected even if the earlier homonym is not
an accepted name” (italics in Miss Green’s text).
Much confusion reigns as to the meaning of invalidity as dis-
tinct from illegitimacy in the sense of the Rules, for the very good
reason that the Rules themselves use these terms in a loose and
contradictory manner. Examples of this confusion are rife in the
Amsterdam Code, and one at least may be cited here. Article 2
defines as illegitimate, names or forms contrary to an Article, and
states that such names cannot be maintained. Article 638, on its
part, prescribes that the name of a taxonomic group “must be re-
jected when its application is uncertain’. Since such a name
[nomen dubium] ‘“‘must be rejected,” it stands to reason that this
name is illegitimate under the definition given in Article 2. How-
ever, Recommendation xxxvii which immediately follows Article
63 authorizes the certification of a nomen dubium following
an adequate taxonomic study made on the basis of new evidence
(Art. 17, Rec. iii, Rec. xxxvii). Thus, Article 63 errs in stating
that a nomen dubium “must be rejected,” branding it implicitly as
illegitimate. Such a name is merely invalid, proof of this being
the fact that this name can be used legitimately under
certification.
Since the Rules themselves are not clear as to the proper use
of validity and legitimacy, it would be useless to argue here
Wheeler’s contention that a nomen provisorium is both invalid and
illegitimate. Sooner or later, a fundamental debate is bound to
take place in a Botanical Congress about these concepts. Mean-
while, I may contribute here a brief comment as to the meaning
of validity and illegitimacy, once again using an example.
As it is well known, the law orders that a testament must con-
form with certain specified requirements, a part of the estate of
the deceaésed going automatically to certain parties by reason of
their being related with the author of the will. If the will is
drawn against the law and, for instance, the estate is distributed
in a manner which is forbidden by law, the will is illegitimate, and
as such it cannot be maintained. A will, conversely, may be drawn
according to the law, but before it takes effect it must go through
the procedure of probating, and is not valid until probated. The
probating of a will is exactly the same procedure as the valid publica-
A MADRONO [Vol. 7
tion of a taxonomic name. Neither a will nor a taxonomic name is
valid until it is probated or published according to the laws of
the land or the Articles of the Rules of Nomenclature. Natu-
rally, neither a will nor a taxonomic name is legitimate if it vio-
lates the law of the land or the Articles. A will that violates the
law and a name that violates an Article may be unimpeachable as
to form, but can neither be probated nor maintained because they
are faulty as to substance. This, in a nutshell, is the distinction
that can briefly be made here between the concept of validity and
that of legitimacy. It is high time that the Articles be carefully
revised and amended in order that they be purged of pointless
and confusing abuses of the proper terms ultimately leading to a
flood of mistaken comments in the literature.
The following synonymy is in order:
THALICTRUM AMETRUM Greene in Muhlenbergia 5: 129. 1909.
T. polycarpum S. Wats. in Proc. Am. Acad. Sc. 14: 288. 18793.
Jepson, Fl. Calif. 1: 580. 1922; Munz, Man. South. Calif. Bot.,
173. 19385; Wheeler in Rhodora 40: 318-320. 1938. Non
Loret.
Arnold Arboretum, Harvard University,
Jamaica Plain, Mass.,
March 3, 1942.
UNA NUEVA ESPECIE DE PINUS MEXICANO
MaxiMino Martinez
Pinus Douglasiana sp. nov. Arbor 20 m. alta; diametros 30—
50 cm.; coma densa rotundata. Cortex leviter scabris, rubescens,
2 cm. crassus, squamatus. Rami expansa; ramuli brunneo rubes-
centis, valde scabri. Folia 5, triangularia, crassa, rectiuscula,
pungentia, 25-33 cm. longa, marginibus denso serrulato, claro
virore vel galbinus coloris, fulgentia, intus glauco in folia juniora.
Hypodermo biformis usque endodermo penetrabilis, chlorenquima
partitus; fascies-exterius endodermi incrassatus. Ductus resini-
feri 3 in parenchymatis parte siti; fasces fibrovasculares 2, ap-
proximati, patentibus. Vaginae persistentes, 20-30 mm. longae,
squamatae, castanei rubescens, dein obscuro castanei. Strobilis
junioribus erectis violaceo fuscus, subterminalibus, oblongis at-
tenuatis, obtusis; squamae crassae, apex expansus vel erectus.
Strobili maturi ovoidei, leviter asymetricus, deflexi, paulum in-
curvati, in apex attenuati, fusco rubescens coloris, cadivus, 7.5—
10.5 cm. longis, terni vel quini. Pedunculi 12 mm. incurvi ad
strobili adnatus. Squamae 28-30 mm. longae, 15 mm. latae; apex
irregulariter, obtuso vel rotundato; umbo subquadrangulo vel
polyangulatus, carina transversa patente, carina longitudinali
depressa, fere complanata in basis strobili. Cuspide complanata,
paulum patente, mucro cadivo. Semina obscura fere ovoidea, 5
mm. longa, ala 25 mm. longa 8 mm. lata, brunnea. Lignum mol-
lis, album; resina fere nulla.
1943 | MARTINEZ: PINUS 5
Fic. 1. Distribucién del Pinus Douglasiana.
Typum in Instituto de Biologia, Mexico; isotypi in Arnold
Arboretum, Harvard University, Jamaica Plain, Massachusetts,
United States National Herbarium, Washington, D. C.
He tenido a la vista ejemplares procedentes de: Srinatoa.
Batel, Concordia; Santa Lucia, Concordia, “‘pino real’; Cerro de
Tecoripa, Sierra del Rosario; limites de Sonora y Chihuahua;
Potrero de Bejarano, Badiraguato; Rosario. Nayarit. Juanacata,
Jala, “pinabete.” Jatisco. San Martin Hidalgo; Tecolotlan;
Cerca de Cuale, a 20 km. al E. de Bahia de Banderas; Ameca,
““pino blanco”; Atengo, “pino blanco o pino hayarin” ; Soyatlan, a
2200 m., “pino blanco” ; Concepcién de Buenos Aires. MicHoacan.
Tiripitio. Oaxaca. El Barrio; Santiago Tlaxoyaltepec. Mexico.
Cuautepec, Sultepec.
Es arbol de unos 20 metros de altura, por 30 a 50 cm. de dia-
metro a la altura del pecho; de copa redondeada y densa; con la
corteza algo aspera de 2 cm. de espesor aproximadamente, rojiza
y escamosa, dividida en placas irregulares. Ramas extendidas,
agrupadas en la parte superior del tronco. Ramillas morenas con
tinte rojizo y muy asperas, debido a la persistencia de la base de las
bracteas, las cuales son anchas, salientes y contiguas. Se desca-
man facilmente. Hojas en grupos de 5, triangulares, gruesas, casi
derechas, y agudas, de 25 a 33 cm. de largo, con los bordes fina-
mente aserrados, de color verde claro, algo amarillento, brillantes,
con tinte glauco en las caras interiores, solamente visible en las
hojas tiernas. El] hipodermo es biforme, muy grueso con 5 capas
de células desiguales e irregularmente colocadas, y presenta
entrantes, a veces dobles, que llegan al endodermo seccionando el
6 MADRONO [Vol.7
clorénquima; las paredes exteriores de las células endodérmicas
son muy engrosadas. Tienen dos fasces vasculares contiguos,
bien distintos, rodeados arriba y abajo de células de refuerzo;
los canales resiniferos son medios y en numero de tres. Las
vainas son persistentes, de 20 a 30 mm., escamosas abajo y anil-
ladas arriba, de color castafio rojizo al principio y castafio
obscuro después. Las yemas son coénicas, de color naranjado
rojizo. Los conillos son moreno violadceos, erguidos, oblongos,
subterminales, algo atenuados en ambas extremidades, romos,
generalmente en grupos de tres, con escamas gruesas, armadas de
puntas extendidas o dirigidas hacia el apice. Conos largamente
ovoides, algo asimétricos, reflejados, ligeramente encorvados,
atenuados hacia el apice, de color moreno rojizo, opacos, caedizos,
de7.5a10.5cm. Se presentan en grupos de 8 a 5, sobre pedinculos
de unos 12 mm., siempre encorvados, quedando con el cono cuando
éste cae. Escamas de unos 20 a 30 mm. de largo, por 15 de ancho,
de apice irregular, obtuso o redondeado, umbo irregularmente
cuadrangular o poligonal, rugoso, quilla transversal patente y una
saliente perpendicular poco marcada; apofisis irregular, sub-
piramidal, algo levantada (casi aplanada en las escamas basales),
cuspide aplanada o muy poco saliente, con espina pronto caediza.
Semilla obscura, casi ovoide, de unos 5 mm., con ala de 25 mm. de
largo por unos 8 de ancho, de color moreno. La madera es
blanda, de color blanco; con muy escasa trementina. Se emplea
en construcciones y para muebles.
Este pino, quizé incluido por Shaw en el Pinus pseudostrobus
var. tenuifolia (Benth.) Shaw, coincide con éste en la estructura
de las hojas, pues el hipodermo forma entrantes, a veces dobles,
que llegan al endodermo, pero dichas hojas son gruesas y fuertes,
de 25 a 88 cm., tiesas y derechas, en tanto que las del Pinus
tenuifolia son muy delgadas, flexibles y colgantes. Los conos
en lo general coinciden con los del P. tenuifolia, pero las apofisis
son mas gruesas.
Teniendo en cuenta que las hojas no son delgadas, sino por
el contrario, gruesas y fuertes, no puede convenirle la denomi-
nacién de tenuifolia, ya que la caracteristica de éste, como clara-
mente lo indica el nombre, es que las hojas son delgadas. El
Arbol se encuentra en una zona relativamente amplia (desde
Sinaloa a Oaxaca).
Por tales razones he visto la conveniencia de considerarlo a
parte con rango especifico.
Su zona de vegetacién, como se ve, comprende Sinaloa, Nayarit,
Jalisco, Michoacan, México y Oaxaca (probablemente también
Guerrero), formando masas puras. Se le ve asociado con Pinus
Lumholtziu, Pinus leiophylla y Pinus oocarpa.
Se denominé en honor de la Sefiora Margaret Douglas, dama
norteamericana, entusiasta admiradora de la Flora Mexicana,
y protectora de los estudios de la misma.
1943] MARTINEZ: PINUS 7
f
ay
7
Pirate 1. Pinus Dovetasiana. Fig. 1. Seccidn transversal de la hoja.
Fig. 2. Rama yconos. (Dib. de M. Ornelas C.)
8 MADRONO [Vol.7
SUMMARY
Pinus Douglasiana was perhaps included by Shaw under P.
pseudostrobus var. tenuifolia (Benth.) Shaw, but it differs from
this in its longer, stouter, leaves and larger apophyses of the
cone scale. It occurs from Sinaloa to Oaxaca. It is named in
honor of Mrs. Margaret Douglas.
Morelia 61, México, D. F.
Mayo de 1942.
AN INVESTIGATION OF THE PRESENCE OF SILICEOUS
RODS IN THE SECONDARY WALL OF
WOODY TISSUE*
Watrer M. ScHALL
In 1920, Forrest B. H. Brown (3) proposed an explanation of
differential wood shrinkage by stating that a skeleton of siliceous
rods existed within the secondary wall of wood elements. He
assumed that these rods, acting as a restraining framework,
kept longitudinal shrinkage at a minimum. Since the presenta-
tion of this explanation, many workers in cell wall structure (6,
7, 12) have referred to this siliceous skeleton or have tacitly
assumed its presence, notwithstanding the fact that the micelle
theory advanced by Nageli (13) in 1863 and substantiated by
subsequent workers (1, 2,7, 11, 12, 17) is now generally accepted
as the logical explanation for the shrinkage behavior of wood.
The present study was undertaken not to explain the mechanics of
shrinkage but rather to investigate the procedure employed by
Brown (3), first to check his results and second, if similar results
could be obtained, to interpret them in the light of the accepted
theories regarding wood shrinkage.
The author wishes to give acknowledgment and thanks to Dr.
R. A. Cockrell for his interest and suggestions, both in conducting
the investigation and writing the final manuscript.
In repeating Brown’s (3) work, the following species of wood
were used: Swietenia mahogoni, Trochodendron aralioides, Quercus
alba, Cedrela sp., Pinus strobus, and a lapachol-forming species of
Tecoma. Tecoma, according to Record (15, p. 582), is divided
into four groups, “Prima vera,’ “roble, ~ “ipe peroba,, vand
“lapacho” or “pao d’arco.” The “lapacho” group is charac-
terized by having wood that is very “hard and heavy, has an oily
olive-brown color, and the vessels are more or less completely
filled with yellow crystalline substance (lapachol), which may
give the surface the appearance of having been dusted over with
sulfur. Ripple marks are always present and usually regular.”
Lapachol (C,;Hi,0;) ‘““‘when moistened with ammonia or dilute
1 Contribution of the Department of Forestry, University of California,
Berkeley.
1943] SCHALL: SILICEOUS RODS IN WOODY TISSUE 9
sodium carbonate, turns a deep wine-red, thus providing a
reliable diagnostic feature.’ In order to give c.mparable results
with Brown (3), the results of tests made on Tecoma sp. are com-
pletely reported although the same tests with comparable results
were made on the other material.
If a siliceous skeleton were present in the secondary wall of
woody tissue, a change in dimension might take place upon
desilicification. To test this, following Brown’s (3) procedure,
blocks of the foregoing species were placed in hydrofluoric acid
to dissolve silica thereby eliminating its reputed restraining effect
on shrinkage. In addition, duplicate control blocks were placed
in hydrochloric acid which does not dissolve silica but which in
other respects should have an effect on the cell wall substance
similar to that of hydrofluoric acid. The blocks, ranging in size
from one-half to one inch in linear dimension, with true radial,
tangential, and longitudinal faces, were measured with a microm-
eter and placed in boiling water. After five and one-half hours
of boiling, during which the blocks were measured at one-half
hour intervals, duplicate blocks of each species were immersed in
separate solutions of concentrated hydrochloric and 52 per cent
hydrofluoric acid. Measurements of the change in dimension
were made after eighteen, twenty-three, twenty-eight and thirty-
one hours’ immersion. After washing in running water for ten
hours, and measuring again, the blocks were placed in an oven at
103° C. to remove the excess moisture. The moist blocks were
then allowed to air-dry for eight days. Since the swelling values
for each of the species followed the same general trend, only the
values for Tecoma sp. are included. These are recorded in
Table I.
In order to supplement the block measurements, individual
fibers obtained by maceration with concentrated nitric acid and
potassium chlorate (4) were treated with both hydrofluoric and
hydrochloric acid. The fibers were mounted in alcohol and
measured both in length and width under the microscope. The
alcohol was then allowed to evaporate and the acid was added.
In no case did the fiber length change after contact with either
acid for an hour. If silica in the form of rods was present, the
time in hydrofluoric acid was sufficient to dissolve the silica from
the fiber. Since no measurable change was observed, it must be
concluded that the acids did not remove the restraining force
holding the fiber cell wall together, and the cell wall must have
been fully swollen before the acids were added.
The effect of chemical treatment on the physical characteris-
tics of the wood was marked. In several cases the blocks showed
a decided tendency to check and some even showed some collapse.
These same blocks were brittle and a slight pressure on any of
the faces caused splitting that took place tangentially along the
annual rings as well as radially along the rays. It is not likely
that merely drying the wood would cause such extreme stresses to
10 MADRONO [Vol. 7
TABLE I
Percentage Swelling for Tecoma sp. in Water and Acid*
Longi-
Time Tangential rar
14, hour 5.4 7.0+
1 hour 6.0 11.0+
114 hours 6.3 12.7+
Per cent change 21% hours 61% 127%
in boiling water 3% hours 7.0
41, hours T.0 13:07
51% hours 7.1
Per cent change 1814 hours 12:2 73
in concentrated hydro- 23 hours 12.8 7.6
chloric acid at room 28 hours 12.8 7.5
temperature 31 hours 12.8 thes
Per cent change
at air dry moisture 8 days —4.7 —43
content
Per cent change 1é6hours | 8.2 15.0¢ | 44 11.0¢ | 0.5 0.1+
in 52 per cent hydro- 20%) hours | 7.7 85.0t | 46 43.0+ | 0.5 —18.0+
fluoric acid at room 271% hours T5510} 4.4 29.0+ 0.3 —22.0+
temperature 301% hours 7.5 4.3 0.4
Per cent change
at air dry moisture 8 days 0.1 —13.8¢ | 2.2 —19.0+ | 0.0 -—40.0+
content ; |
* Values represent averages for at least four determinations. Negative
values represent shrinkage.
+ Results as obtained by Brown (3) for similar treatment.
be set up but it is probable that the splitting action was caused
by a material weakening of the cell wall by chemical action.
These checked blocks were disregarded and only apparently
sound blocks were used for measurement.
In determining the air-dry moisture content before and after
acid treatment, it was found that the hygroscopicity had de-
creased. After treatment with hydrochloric acid, the moisture
content of the blocks was reduced to three-fourths of the
original value and after treatment with hydrofluoric acid to
one-third. Again, this would indicate that the acids caused a
chemical and physical change in the minute structure of the cell
wall so that its original equilibrium moisture content was sig-
nificantly reduced. |
Each of the untreated blocks was sectioned on a microtome,
using a jet of a steam to soften the woody tissue. The sections,
mounted in water, were examined at 1300 diameters under a Zeiss
binocular microscope equipped with an oil immersion apochro-
matic objective (N.A. 1.3) and 10x compensating eyepieces.
Likewise, sections mounted in glycerine were examined and even
after contact with the liquid for twelve hours, in no case was there
any indication that small isolated areas, supposedly the cut ends
1943] SCHALL: SILICEOUS RODS IN WOODY TISSUE 11
of siliceous rods, were present. The photomicrograph of a cross
section of Tecoma sp. (fig. 1) taken at 810 diameters shows no
discontinuities in the cell wall.
The index of refrac-
tion of the cell wall, as
found by Brown (3),
was verified using Mc-
Lean’s Solution (4) of
known indices on cross
sections of Tecoma sp.
Since the index of re-
fraction of the secon-
dary wall and silica are
practically the same,
the proximity of the
two indices may have
obscured any possible
difference due to the
presence of silica.
Both the cross sec-
tions and the individual
fibers of Tecoma sp.
were incinerated to see
if visual evidence of the
rods could be found
during the course of in- Fic. 1. Transverse section of a lapachol-
cineration or in the ash. forming species of Tecoma. x 810.
Individual cells were
isolated from a section 14 microns thick and incinerated over an
alcohol flame. The heating time was lengthened for successive
sections so that examination could be made at varying degrees of
incineration. Under the microscope, the ash showed irregulari-
ties but no regular arrangement of bodies was noted. According
to Uber (19), these irregularities are probably a result of a natu-
ral tendency of the ash to check upon shrinkage. Macerated
fibers were incinerated the same way but again no visual evidence
of the localization of silica was obtained.
In addition to determining the effect of the presence of a
siliceous skeleton on the shrinking and swelling of wood, the
actual amount of silica present was determined. Following
Brown’s (3) procedure, this determination was carried out in two
steps. The first was to determine the ash content and the second
to determine the per cent silica in the ash. An oven-dry sample of
Tecoma sp. was accurately weighed and ashed in an electric muffle
at moderate red heat. The weight of ash thus obtained was
0.25 per cent of the dry weight of wood. A silica determination
based on the weight of ash was made using hydrochloric and per-
chloric acid in the quantitative analysis (9). The weight of
12 MADRONO [Vol. 7
silica thus obtained gave an average of 0.85 per cent of the dry
weight of ash.
The values obtained for ash and silica content are consider-
ably below those given by Brown (3). He states that, “after
combustion, 1.8 per cent (of the dry weight of the fiber) of min-
eral matter was obtained and 0.1 per cent (of the dry weight of the
fiber) of silica or silicic acid.” These percentages are of little
value since they are based on the dry weight of fiber which can be
obtained only arbitrarily. The maceration procedure is ex-
tremely variable since neither the time nor the temperature of
reaction is standardized. A trial maceration using concentrated
nitric acid and potassium chlorate (15), gave a dry weight of
fibers for Tecoma sp. equal to 42 per cent of the dry weight of
wood. Based on the dry weight of fibers, this would give values
for ash and silica content approximately twice as large as results
based on the dry weight of wood if the assumption was made that
all the mineral matter was retained by the macerated material.
A comparison of ash and silica content cannot be made with any
marked degree of accuracy, however, since within a single tree,
the values vary from periphery to pith and from the base to the
top of the tree.
It is not outside the realm of possibility that inorganic mate-
rials are centralized within certain areas in the cell wall. Kerr
and Bailey (11, p. 285) state that “the central layer of normal
tracheids, fiber-tracheids, and libriform fibers is composed, in
all cases, of a complex and firmly coherent matrix of cellulose
with elongated, intercommunicating interstices. Within these
interstices more or less ‘lignin’ and other non-cellulosic constit-
uents may be deposited.” Bailey (1) recently states specifically
that minerals may be deposited within these interstices. Since
the mineral content occupies such a small percentage of the
weight of materials within the interstices, it is improbable that
silica would be so localized as to form a continuous rod.
In view of these findings, it is not likely that a highly silicified
skeleton is present in the secondary wall of woody tissue. The
results of swelling tests alone should be conclusive evidence of
this fact since several species of wood were used and the blocks
treated with two mineral acids, one of which would dissolve
silica while each should have similar effects on the cell wall
structure. If siliceous rods were present they would certainly
have been disintegrated after immersion in 52 per cent hydro-
fluoric acid for 80 hours. As indicated in Table I, a small increase
in size was observed but no appreciable difference was noted be-
tween the blocks placed in hydrofluoric acid and those placed in
hydrochloric acid. Furthermore, the individual fibers failed to
give a measurable change when each of the acids was added. In
a schematic drawing of a fiber before and after hydrofluoric acid
treatment, Brown (8) indicates that the diameter increases and
the length decreases with the acid treatment. He explains the
1943 | SCHALL: SILICEOUS RODS IN WOODY TISSUE 13
change by stating that the siliceous rods had been broken, thus
freeing the so-called homogeneous substance which could then
swell without hindrance. If the cell wall is a homogeneous
substance, the fiber should have increased in length and width
in the same proportion and an inverse relationship could not
have resulted.
The acids probably caused a degradation of both lignin and
cellulose as well as attacking any of the minerals in the wood.
This is borne out by the mechanical weakening of the wood and by
the reduced hygroscopicity. Chamberlain (5) and Sacc (17) state
that the action of hydrofluoric acid is to soften wood but they
do not point out the manner of softening. Plowman (14) in 1904
stated that the action of hydrofluoric acid was to remove silica
and other mineral deposits from the wood. Kerr (10) and Har-
low (8) point out that the action of hydrofluoric acid is to attack
the cell wall substance and the removal of silica is a secondary
operation. Harlow (8) explains the softening action as a de-
gradation of lignin since the treated wood does not respond to
the Maulé reaction which is employed as a test for the presence
of lignin. Kerr (10), on the other hand, states that the action is
due to a degradation of cellulose to hydrocellulose and explains
the action by drying other acids into wood in order to soften the
woody tissue. Rudiger (16) points out that swelling precedes
the dissolution of cellulose in liquid hydrofluoric acid and that the
lignin structure of the membranes was destroyed, although the
lignin itself did not swell. Forsaith (6) in his explanation of dif-
ferential swelling of wood with adsorption of water credits the
siliceous skeleton as proposed by Brown (8) as exerting an influ-
ence in conjunction with the micelle theory as proposed by
Nageli (13). A more recent worker, Maby (12, p. 434), in his
explanation of shrinkage and swelling, says, “On the other hand,
the longitudinal siliceous strands in the cell wall, noted and
described by F. Brown, might be expected to exert a binding effect
over dimensional changes in the longitudinal direction.”
Schorger (18, p. 9) states, ‘“Nageli, as a result of his study of
the growth of starch grains and the cell wall, concluded that the
cell wall consists of ultramicroscopic, crystalline, molecular com-
plexes which he called micellae. By this assumption he was able
to explain striation, stratification, swelling, double refraction,
and other properties of the cell wall.’ Subsequent workers have
followed this general idea. Bailey (1, 2) points out that the
cellulose consists of chains of anhydrous glucose residues which
tend to aggregate in a parallel fashion. He also states that the
aggregation of chain molecules is not in separate groups but rather
a part of a continuous system which is held together by over-
lapping chain molecules and perforated by intercommunicating
spaces. It is probably in these intercommunicating spaces that
water is adsorbed and causes the changes in the dimension of
wood. The mineral content is also probably localized here but
14 MADRONO , [Vol.7
evidence of a continuous bond between inorganic material of the
cell wall has not been found.
CoNCLUSIONS
1. Silica in the form of continuous siliceous skeletons is not
present in the secondary wall of woody tissue.
2. The silica content is such a small percentage of the total
weight of wood that it could not have an appreciable effect,
greater than other minerals, on the differential swelling or shrink-
ing of wood.
3. Other than bringing about more rapid degradation of the
wood substance, hydrofluoric acid is similar to hydrochloric acid
in its action on the cell wall.
University of California, Berkeley,
May 28, 1942.
LITERATURE CITED
1. Battey, I. W. The cell and protoplasm. Publ. Amer. Assoc. Advance-
ment Sci. no. 14: 31-43. 1940.
2 . The microfibrillar and microcapillary structure of the cell
wall. Bull. Torr. Bot. Club 66: 201-213. 1939.
3. Brown, F. B. H. The siliceous skeleton of tracheids and fibers. Bull. Torr.
Bot. Club 47: 407-424. 1920.
A. The refraction of light in plant tissues. Bull. Torr. Bot.
Club 47: 243-260. 1920.
5. CHampertain, C. J. Methods in plant histology. Univ. of Chicago Press.
1938.
6. Forsairu, C. C. The technology of New York State timbers. Tech. Publ.
18. N. Y. State Coll. Forestry. 1926.
7. Hawtey, L. F., and L. E. Wise. The chemistry of wood. The Chemical
Catalog Co., Inc., New York. 1926.
8. Hartow, W. H. Contributions to the chemistry of the plant cell wall.
IV. Tech. Publ. 26. N. Y. State Coll. Forestry. 1928.
9. Hittesrann, W. F., and G. KE. F. Lunpextt. Applied inorganic analysis.
John Wiley and Sons, New York.
10. Kerr, I. The action of hydrofluoric acid in softening. Tropical Woods 40:
37-41. 1934.
11. Kerr, I. and I. W. Batrey. The visible structure of the secondary
wall... . Journ. Arn. Arb. 16: 273-300. 1935.
12. Masy. J. C. Micellar structure of the tracheid wall in certain woods in
relation to morphogenetic and mechanical factors. New Phyt. 35: 432-
455. 1936.
13. NAcet1,C. W. Die Anwendung des Polarisationsmikroskops auf die Unter-
suchung der organischen Elementartheile. Beitr. Wiss. Botan. 3: 1-126.
1863.
14. Prowman, A. B. The celloiden method for hard tissues. Bot. Gaz. 37: 456-
461. 1904.
15. Recorp, S. J. Timbers of tropical America. Yale Univ. Press. New
Haven, Conn. 1924.
16. Rupicer, W. Papierfabrickant. 38: 9-13. 1940.
17. Sacc, J. E. Elements of botanical microtechnique. McGraw-Hill. New
York. 1940.
18. Scuorcer, A. W. The chemistry of cellulose and wood. McGraw-Hill.
New York. 1926.
19. User, F. M. Microincineration and ash analysis. Bot. Review 6: 204-226.
1940.
1943 ] KEARNEY: NEW COWANIA FROM ARIZONA 15
A NEW CLIFF-ROSE FROM ARIZONA
Tuomas H. Kearney
It was noted in a recent publication’ that an apparently unde-
scribed Cowania occurs in western Arizona. The writer is now
convinced that this plant merits recognition as a species.
Cowania subintegra sp. nov. Frutex parvus, ramulis ad-
scendentibus-patentibus 80-75 cm. longis, cortice albo-griseo;
folia nec viscida nec distincte glanduloso-punctata, integra vel
nonnunquam apicem versus 1—2-dentata, usque ad 15 mm. longa
et 3 mm. lata, spathulata, apice obtusa, basi attenuata, margine
valde revoluta, supra laetevirentia et parce araneosa, subtus dense
albo-lanata; ramuli, pedicelli, hypanthium, et calycis lobae tomen-
tosi sine glandulis stipitatis ; pedicelli 4-11 mm. longi hypanthium
subaequantes vel valde superantes; hypanthium infundibuliforme
5—7 mm. longum; petala ochroleuca.
A straggling shrub with stems up to 75 cm. long, the branches
ascending-spreading, the bark pale gray, becoming somewhat
shreddy; herbage not viscid; twigs, pedicels, hypanthium tube,
and outer face of the calyx lobes whitish tomentose, without stipi-
tate glands; leaves up to 15 mm. long and 3 mm. wide but usually
shorter and narrower, mostly 1-veined and entire but occasionally
with 1 or 2 subapical rounded teeth, oblanceolate, obtuse at apex,
attenuate at base, the margin strongly revolute, thick, minutely
and obscurely glandular-punctate, bright green and loosely arach-
noid-pubescent over the whole upper surface, the lower surface
densely and conspicuously white-lanate; pedicels 4 to 11 mm.
long, nearly equaling to much longer than the hypanthium;
hypanthium narrowly funnelform, attenuate at base, 5 to 7 mm.
long, in anthesis about 8 mm. wide at summit; calyx lobes about
4 mm. long, broadly ovate, rounded at apex and sometimes ob-
securely apiculate, denticulate-ciliolate, spreading at anthesis,
becoming reflexed; petals ochroleucous, about 10 mm. long and
6 mm. in greatest width, rounded and slightly erose (occasionally
shallowly cleft) at apex, cuneate at base; stamens about 40, the
filaments about 4 mm. long, the anthers 1.25 mm. long, nearly
orbicular; pistils 8 to 7, the ovary short-stipitate, densely se-
riceous, the style 6 to 7 mm. long at anthesis, sericeous on the
lower one-half to two-thirds, naked above; achenes about 6 mm.
long, narrowly obpyramidal, glabrous except near the apex, the
persistent style about 25 mm. long (perhaps longer at full ma-
turity), silky-plumose with long antrorse hairs except the apical
portion, this naked, 2 to 8 mm. long.
The type was collected about two miles west of Burro Creek
crossing on the road from Wikieup to Hillside, southeastern
Mohave County, Arizona, near the Yavapai County line, altitude
1 Kearney, Thomas H., Peebles, Robert H., and collaborators. Flowering
plants and ferns of Arizona. United States Department Agriculture Misc. Publ.
423: 405. 1942.
16 MADRONO [Vol. 7
2,500 feet, April 18, 1941 (Darrow & Benson 10891). The species
is known only from the type locality, where it had been collected
first on April 20, 1938 (Darrow & Crooks 3). On both dates of
collection, only a few late flowers persisted, but the fruit was not
yet fully mature. Dr. Lyman Benson stated (personal communi-
cation) : “The base of the plant is perhaps as much as one to one
and one-half inches thick, but the trunk continues for only a few
inches above ground.” He also reported: ““We found the plant
growing in a rather limited area on disintegrated material of a
peculiar white rock and associated with a vegetation entirely dif-
ferent from that on surrounding territory. However, although
the distribution of the plant was restricted, it was locally abun-
dant.”
Cowania subintegra strikingly resembles C. ericaefolia Torr. of
western Texas in habit, appearance, small stature, and very nar-
row, mostly entire leaves, but the Texas species has linear, sharply
cuspidate leaves not more than 6 millimeters long, stipitate glands
on the hypanthium (often also on the pedicels), and darker
colored bark.
The flowers and fruit of C. subintegra apparently present no
characters that are not within the range of variation of C. Stans-
buriana Torr., but that is a much larger and more erect shrub,
attaining (exceptionally) a height of seven and one-half meters.
The branchlets are more stiffly ascending, the bark reddish brown
or dark gray, and the herbage usually very viscid. The leaves
are much larger, cuneate-obovate in outline, pinnately veined and
deeply 8-cleft with the terminal lobe in turn 38-toothed or 3-cleft,
and they are nearly always conspicuously punctate with few large
glands. With the single exception, so far as the writer knows,
of the specimen noted in the next paragraph, the pedicels and
hypanthium are beset with stipitate glands and are not tomentose,
or but thinly so.
A collection from near Rye Creek, Gila County, Arizona
(Collom 97 in 1933), has the herbage non-viscid, the leaves mi-
nutely and inconspicuously punctate, and the pedicels and hy-
panthium densely pubescent and lacking stipitate glands. In
these characters it resembles C. subintegra but in other respects
it is not distinguishable from C. Stansburiana.
The writer believes that Jepson was justified in reducing C.
Stansburiana to a variety of C. mexicana D. Don (Man. FI. PI. Calif.
498. 1925). The diagnostic characters given by Rydberg in his
key and descriptions (N. Am. Fl. 22: 415, 416. 1913) are as
follows:
C. mexicana: Hypanthium campanulate, abruptly contracted
into the pedicel. Glands of the pedicel sessile and often hidden
in the tomentum.
C. Stansburiana: Hypanthium funnelform, gradually contracted
into the pedicel. Glands of the pedicel stalked.
Standley (Contrib. U.S. Nat. Herb. 23: 326. 1922) mentions
also that the leaf lobes are entire in C. mezicana, whereas at least
1943 | KEARNEY: NEW COWANIA FROM ARIZONA 17
the terminal lobe is cleft or dentate in C. Stansburiana, as was
pointed out by Torrey in his original description (in Stansb. Expl.
Great Salt Lake, 386. 1852). The color character of the bark
given by Standley does not hold, many specimens of C. Stansburi-
ana from Utah and northern Arizona having brown bark.
The manner in which the several characters are associated in
specimens from Mexico and from the United States is shown in
Table 1. It is evident that whereas in most of the specimens the
Taste 1. Association of characters in specimens of Cowania in the United
States National Herbarium that have been referred, respectively, to C. mexicana
and C. Stansburiana.
; Stipitate glands
Ea leat Hypanthium on hypanthium,
sel etc.
Species and Campan-
collection ulate, + | Funnel:
Enti Toothed | abruptly 4
ntire Berk Bone attenu- | Present | Absent
tracted ae au
at base a
C. mexicana
Rose 11659, Cusi-
huiriachic, Chih. .. x xX x
E. Palmer 12, Tepe-
huanes, Dur. ...... x xX xX
E. Palmer 71, Papa-
squiaro, Dur. ..... , X xX x
E. Palmer 4669, Papa-
squiaro, Dur. ..... xX xX X
Dugés in 1899, moun-
tains, Guanajuato . xX —- —- — —
C. Stansburiana
Most specimens from
Utah, Arizona, etc.. xX xX x
Hartman 276, Nacori,
SOM te, Ae fhe ge cats xX X X
M. E. Jones 5586c, |
Provo, Utah ....... | X xX x
V. Bailey 1457, near
St. John’s, Ariz. ... xX xX x
Knowlton 238, Grand
Canyon, Ariz. ..... X xX x
Collom 97, Rye Creek,
PNIZ et MO ok a bic xX X X
characters considered by Rydberg and by Standley to be diag-
nostic of C. mexicana and C. Stansburiana, respectively, tend to be
associated as indicated by them, there are several marked excep-
tions. Thus two of the five specimens referred to C. meaicana
have the hypanthium funnelform or intermediate,? rather than
2D. Don, in his description of the genus Cowania, based solely upon C.
mexicana (Trans. Linn. Soc. London 14: 575. 1825), states: “calyx [hypanthium |
obturbinatus basi attenuata tubulosus.” His illustration (Tab. XXII), how-
ever, shows the hypanthium as campanulate and abruptly contracted.
18 MADRONO [Vol.7
campanulate, and three specimens from Utah and Arizona that
are C. Stansburiana in all other characters have a campanulate
hypanthium that is abruptly contracted at base. One of the
Mexican specimens (Dugés in 1899) that has the entire primary
leaf lobes of C. meaicana, is intermediate in shape of the hy-
panthium and the latter is conspicuously glandular, although the
glands are sessile or nearly so. The specimen from Rye Creek,
Arizona, as was noted in a preceding paragraph, although con-
forming to the characterization of C. Stansburiana in shape of the
leaves and of the hypanthium, lacks the stipitate glands. It is
also aberrant in having the leaves obscurely and minutely punc-
tate, not conspicuously and coarsely so, as in all other specimens
of C. Stansburiana and in all specimens of C. mexicana examined
by the writer.
Bureau of Plant Industry,
United States Department of Agriculture,
Washington, D. C.
June 27, 1942.
FRANCESCO FRANCESCHI
JOHN M. TucKErR
This paper is a brief account of the life and work of the man
who stands out above all others in the history of horticulture in
southern California—Dr. Emanuele Orazio Fenzi, known to his
associates in this country in later life as Dr. Francesco Franceschi.
In gathering data I drew upon a number of sources, and take this
opportunity to express my appreciation to the following persons
for the assistance they have given me: to Dr. Emily O. Lamb, who
lived with the Fenzis at Santa Barbara as a member of the family
for fourteen years, to Mr. Peter Riedel and to Mr. H. M. Butter-
field, for much of the information contained in the following
pages; to Mr. Butterfield, Miss Annetta Carter, and Mr. M. Van
Rensselaer, for the loan of horticultural catalogues and journals
—sources of much valuable data; to Dr. H. L. Mason, for placing
at my disposal a collection of Franceschi’s business correspon-
dence (a fund of information of which I have scarcely scratched
the surface), and to Dr. Howard S. Reed, for guiding my efforts
in preparing this paper. Particularly informative also, were the
following two articles: “Una gloria dell’orticoltura italiana. I]
Dott. Emanuele Orazio Fenzi,’ by Mario Calvino—(L’Agricol-
tura Coloniale, 22: 122-128. 1928.) and “Dr. Fenzi’s Contribu-
tions to American Horticulture,’ by F. W. Popenoe (Journ.
Hered. 18: 215-220. 1922.).
Emanuele Orazio Fenzi was born March 12, 1848, in Florence,
Italy. His grandfather was a very wealthy banker of that city
and a senator, a man of an aggressive dominating personality.
The Fenzi family were patrons of the arts and sciences, and fol-
lowed the latest developments in these fields with great interest.
1943 | TUCKER: FRANCESCHI 19
The renowned pianist, Anton Rubinstein, was a friend of the fam-
ily, and on his visits to Florence, used to come to their home to
practice, because their piano was the finest in the city. Thus, the
background of young Fenzi was one of culture and wealth, with
its attendant advantages.
In early youth he lost his parents, and his grandfather took it
upon himself to see that he was properly educated. The grand-
father, seeing his young charge as a future man of affairs, sent
him to the University of Pisa to study law. Although he would
far rather have studied botany, he complied with his grandfather’s
wishes and in 1864 received the degree of Doctor of Laws.
After leaving the university, he turned to botanical and horti-
cultural pursuits, despite the fact that his grandfather wanted him
to start upon a business career. Endowed with a large estate
which made him financially independent, he was able to indulge
his tastes as he chose in the years that followed. At his country
place near Florence, he formed an arboretum of rare trees, and
on the estate of a relative near Rome, he assembled a large collec-
tion of plants from countries all over the world. He was the first
to introduce bamboos to Italy; Genista monosperma was another of
his introductions. He did considerable work in the improvement
of grape and olive culture in Italy, and frequently contributed
horticultural and botanical articles, not only to Italian periodicals,
but to the English journal, “The Gardener’s Chronicle,’ as well.
Fenzi travelled a great deal, and at one time or another visited all
the principal botanical gardens of Europe. In May, 1874, he
served as secretary of the International Agricultural Exposition
at Florence and edited its catalogue. Shortly thereafter, he
became the first secretary of the Royal Tuscan Society of Horti-
culture, an organization of which he was later president.
His activities in later years were not confined to the fields of
horticulture and agriculture, however. When his grandfather
died, he took charge of the affairs of the bank and, in addition,
managed an estate. He was instrumental in establishing an elec-
tric tram line to Fiesole, the first in Italy, and the steam lines at
San Casciano and Greve. But he had so little enthusiasm for
business, that he soon turned over the management of the bank
to a cousin. Then, during the economic crisis of 1889-90 Fenzi
was forced to close the bank. In order to settle accounts with its
creditors he found it necessary to liquidate virtually everything
he owned, so that, finally, he and his family were left with only
a very small fraction of their once large fortune.
Because of his losses, he was no longer hampered by a multi-
tude of business matters, and saw a chance to put his interest in
plants to work. His ambition was to gather together in one area
plants from countries all around the globe. He decided to go to
southern California because the climate was well suited to his
purpose. So, in 1893, he came to Los Angeles, his wife and fam-
20 MADRONO [Vol.7
ily remaining in Italy. He was in California six years before
they joined him. From a strong feeling of family pride he
dropped the surname, Fenzi, lest the stigma of his bank failure
follow him to America, and adopted in its place a family name,
Franceschi. During the twenty years he spent in California, he
was known to all but his intimates, as Dr. Francesco Franceschi.
In Los Angeles, he met J. C. Harvey, the elder Mr. Howard, E. D.
Sturtevant, and other horticulturists and nurserymen. He re-
mained there only a year, however, before he moved to Santa Bar-
bara where he entered into partnership with C. F. Eaton in raising
nursery stock. After a short time, the partnership was dissolved
and Franceschi started a nursery business of his own, calling his
organization the Southern California Acclimatizing Association.
To obtain new species he wrote to botanical gardens, collec-
tors, and plantsmen in all parts of the world, and soon developed
an extensive correspondence. Here the unusual linguistic ability
Franceschi possessed stood him in good stead, for he read, wrote
and spoke not only his native Italian, but also English, German,
French, Spanish, and modern and ancient Hebrew.
Yet his interest was far from being confined to exotics. He
was ever on the lookout for any elements of the native flora that
might possess striking ornamental qualities. Indeed, in Novem-
ber, 1894, having been in Santa Barbara less than a year, he made
a week’s trip to Santa Cruz Island, the largest of the Santa Bar-
bara Channel Islands. While there he obtained seeds of several
different species which were at that time unknown in the horti-
cultural trade. The most noteworthy of these was Lyonothamnus
floribundus var. asplenifolius, the Santa Cruz Island ironwood.
Unable to find any seedlings of this tree, he laboriously dug up a
living stump and gathered some seed. With considerable dif-
ficulty, he managed to carry his prize back to camp, and on his
return to the mainland planted it in his lathhouse in Montecito,
a few miles from Santa Barbara. In five or six months it had
started to sprout. The next year, when he moved his nursery to
Santa Barbara, he transplanted it to his new location where, in a
few years, it developed into a fine tree. From the seed, Fran-
ceschi obtained several trees, one of which is the fine specimen to
be seen today in the grounds of the old botanical garden north of
the library, on the campus of the University of California at
Berkeley.
He had been in Santa Barbara only a year, when, in 1895, he
published a small book entitled, ‘““Santa Barbara Exotic Flora.”
This book contains a good deal of meteorological and climatic
data, and observations on the soil and native flora of the region.
It includes notes on the history of plant introduction in the region,
and mentions the two oldest introduced trees of the town, Casi-
miroa edulis, the White Sapote, and Prunus Capuli, the Capulin
Cherry, both natives of Mexico. This volume gives an appar-
ently comprehensive review of the exotic plants then cultivated in
PLatTeE 2. FRANCESCO FRANCESCHI.
:
Ni Uh sare t
4 nate
/ : han } : if
x : fi
So ‘ }
1943] TUCKER: FRANCESCHI 21
Santa Barbara and evidences the keenness of Franceschi’s obser-
vations, and the breadth of his botanical knowledge.
David Fairchild, in his recent book, “The World Was My
Garden,” tells of meeting Franceschi at Santa Barbara in 1898.
He speaks of Franceschi’s enthusiasm in the following words:
“Santa Barbara in 1898 was but a simple, small town. Residents
of the beautiful hillside villas today would not credit their eyes
could they visualize the bare, sparsely settled roads where I drove
with Dr. Franceschi. . . . Santa Barbara was so undeveloped
that I considered him visionary and over-optimistic. However,
he foresaw the future more clearly than I, and lived to see Santa
Barbara become a great winter resort containing hundreds of
beautiful villas like those on the Riviera.’ Fairchild, at that time
with the United States Bureau of Plant Industry, was impressed
with the work Franceschi was doing, and had numerous new plant
introductions of the Bureau sent to him from time to time for trial
in Santa Barbara.
In 1904 Franceschi acquired forty acres of land on Mission
Ridge, at that time a dry, barren hillside entirely outside the
bounds of the city. Here he built the house in which. he lived
until he left Santa Barbara, a place he named “‘Montarioso.” He
established another nursery here, and, in order to have an ample
water supply, built a small reservoir near the top of the hill. A
perusal of his business correspondence leaves one with the impres-
sion that he was not blessed with much financial success during
the following years. In June, 1904, his propagating house
burned and he appears to have been considerably in debt in the
months that followed. In 1907 he went into partnership with
Mr. Peter Riedel, and incorporated the Southern California
Acclimatizing Association, hoping thereby to be relieved of some
of the burden of routine business matters. This arrangement did
not work out well, however, and after little more than a year, in
1909, they decided to dissolve their partnership. Following this
incident, Franceschi continued his business independently on Mis-
sion Ridge, calling it the Montarioso Nursery. By offering his
services in landscaping and maintenance of the grounds of a num-
ber of estates in Montecito, he was able to bolster his income.
However, the continued cost of introducing new plants, a work
Franceschi carried on despite his reverses, and the limited demand
for his exotic rarities, made financial disaster inevitable.
It is not surprising, therefore, that upon receiving an offer
from the Italian government in 1912, to take a post in the African
colony of Libya, he decided to accept. He was to introduce new
plants having agricultural and horticultural value, and to do what
he could to develop agriculture in the colony, the government
furnishing the land and facilities for his introduction grounds.
Accordingly, on July 21, 1918, he bade farewell to Santa Barbara,
and with his wife and daughters, started out for his native Italy.
22 MADRONO [Vol. 7
His two sons remained in California, the older carrying on the
business as manager of the Montarioso Nursery. For the next
year and a half, Franceschi lived in seclusion on the Italian
Riviera, preparing the manuscript of a book, “Frutti Tropicali e
Semi-Tropicali,” which was published at Florence in 1915. A
short time later, he made a reconnaisance trip to Libya, accom-
panied by his nephew, Guido Corsini, and in February, 1915, laid
out his establishment in the city of Tripoli. Thus, at an age when
most men would have been long since retired, Franceschi, at 72,
was setting out on a fresh venture. He held this post for several
years, and then retired to carry on the same work at his own
expense.
In 1922, he was awarded the third Meyer Memorial Medal by
the council of the American Genetic Association. This is a medal
presented periodically to persons who have accomplished out-
standing work in introducing plants to American horticulture.
The medal was sent to Tripoli, in 1923, by the governor, Count
Volpi.
Franceschi continued his literary activity to the end of his life,
contributing articles on agriculture in Tripoli to the Italian jour-
nal, “L’Agricoltura Coloniale.” He died in Tripoli on November
5, 1924, at the age of 81. Franceschi’s oldest daughter has car-
ried on his work in Tripoli, and today maintains a successful
nursery there. She, apparently, had, in addition to a deep inter-
est in things botanical, considerable business acumen. Probably
the most important accomplishment of the two in Tripoli, has
been the importation and propagation of large numbers of euca-
lyptus trees of various species.
HortTicuLtTURAL ACHIEVEMENTS
During the years that he lived in Santa Barbara, Franceschi
wrote numerous articles for the local newspaper, and was a regu-
lar contributor to such journals as “Pacific Garden” and “Rural
Californian.”’ He did much toward developing an appreciation
of the beauty as well as the economic value of the large number
of ornamental plants and tropical and subtropical fruits that could
be grown in the region. His reputation spread throughout horti-
cultural circles in this country, and every botanist or plantsman
who visited Santa Barbara, was certain to pay him a call to see
his collection of rare plants. The esteem in which his judgment
and opinions were held is shown by the frequency with which he
is quoted in the literature of tropical and subtropical plants.
Among the multitude of new plants that he brought to the
gardens of Santa Barbara, the following are some best suited to
the climate there. A number of these have attained the popular-
ity they deserve, although some, none the less fine ornamentals,
are still rather rare.
1943] TUCKER: FRANCESCHI 23
Acacia obliqua Harpephyllum caffrum
Acacia podalyriaefolia Hibiscus heterophyllus
Aglaia odorata Jasminum simplicifolium
Aleurites Fordii Lippia repens
Alée Salm-Dyckiana Lithraea Gilliesii
Anthyllis Barba-Jovis Lyonothamnus floribundus var.
Asparagus decumbens asplenifolius
Asparagus scandens var. de- Metrosideros tomentosum
flexus Myoporum acuminatum
Bauhinia grandiflora Myoporum tomentosum
Bauhinia tomentosa Pithecoctenium clematidium
Bauhinia variegata Pithecoctenium muricatum
Benthamia fragifera Pittosporum heterophyllum
Bocconia frutescens Pittosporum rhombifolium
Buddleia madagascariensis Pittosporum viridiflorum
Carica quercifolia Psidium lucidum
Convolvulus florida Rhynchosia minima
Dioclea glycinoides Schinus terebinthifolius
Dombeya natalensis Schotia latifolia
Dombeya punctata Solanum Guatemalense
Erythrina tomentosa Sterculia discolor
Eugenia edulis Stigmaphyllon littorale
Feijoa Sellowiana Taxodium mucronatum
Ficus altissima Tecoma garrocha
Ficus infectoria Tipuana speciosa
Ficus retusa Tricuspidaria dependens
Genista monosperma _ Vitis capensis
Of all the new plants Franceschi introduced, none has be-
come better known than Lippia repens. Its popularity is probably
due in large part to the publicity given it by Franceschi, but the
importance he attached to this particular introduction appears to
have been based on an erroneous idea that he held. In 1904, in
an article that he wrote for the Los Angeles Times, Franceschi
states that he first introduced this species from Italy in 1898.
He says: “From the Director of the Botanic Garden in Rome I
obtained by mail a small tin box of Lippia plants, less than 12
ounces weight. Now, after six years, there are hundreds and
hundreds of acres planted with Lippia between California, Ari-
zona, Mexico and Australia, and it all came out of that small
time DOX.)... |
In contrast to this remarkable statement, H. N. Moldenke, an
authority on the Verbenaceae, in a personal communication of
May 26, 1941, makes the following comment: “You can be very
sure that the plants (of “Lippia repens’’) of Texas, New Mexico,
and Arizona, and central, eastern, and southern United States
have nothing to do with the plants introduced into California by
Franceschi, but it is my belief that most of those of southern
California ‘(at least all that I have seen so far) are descendants
of the ones he introduced.”
24 MADRONO [Vol.7
The efforts Franceschi made to secure new plants, to make
certain of their identity, and to obtain accurate information re-
garding their culture and optimum growth conditions, were often
great, and show a truly scientific spirit. Scattered throughout his
correspondence are letters to and from such well-known botanists
as Joseph Burtt-Davy, William Trelease, J. H. Maiden of
Australia, Charles Sprague Sargent, Harvey Monroe Hall, Miss
Alice Eastwood, and T. S. Brandegee, requesting and receiving
identifications of specimens which he had submitted. As another
illustration, his letter of October 80, 1908, to Mr. C. Wercklé at
San Jose de Costa Rica, might be cited. Wercklé was the dis-
coverer of Hidalgoa Wercklei, the “Climbing Dahlia,’ which Fran-
ceschi had introduced at Santa Barbara. Having had difficulty
in bringing it to flower, we find him, in this letter, requesting
information regarding its mode of growth and climatic require-
ments from the man who, logically, would know most about such
matters—its discoverer. This thoroughness was typical of
Franceschi.
A good example of his persistence and determination in effect-
ing the introduction of a desirable species, is the case of Tarodium
mucronatum, the ““Montezuma Cypress.” Franceschi had often
seen the tree in the Botanic Garden at Naples, planted by Tenore,
the botanist who described the species, and its beauty had made
a lasting impression upon him. In 1898 he decided to obtain
seeds and try it out in Santa Barbara, and accordingly, sent to
Naples for seed, which failed to germinate. Year after year,
seeds from there and elsewhere persistently refused to germinate,
but at last, in 1908, after ten years of failure, his patience was
rewarded. Through his friend, Professor C. Conzatti of Oaxaca,
he obtained, from the Federal Park at Chapultepec, Mexico, seeds
which, much to his gratification, germinated successfully. Today,
many fine trees grown from these seeds may be seen in the parks
and gardens of Santa Barbara.
The culture of tropical and subtropical table fruits was a sub-
ject of special interest to Franceschi. He carried on much
correspondence on aspects of this subject with men of the United
States Bureau of Plant Industry, and horticulturists and fruit
growers (the Popenoes of Altadena, California, particularly),
throughout the warmer parts of this country and of many foreign
countries. He introduced several new species of Anona, and from
a superior tree of A. Cherimola growing in Altadena, he raised
plants in 1910 which he subsequently sent out under the name of
A. Cherimola mammillaris. His best known fruit introduction was
the “Pineapple Guava,” Feijoa Sellowiana. This he introduced in
1901, obtaining seeds from France, where it had been introduced
previously from its native South America. Although it was given
much publicity at the time, Feijoa has not gained the popularity
that Franceschi had hoped for it, and certainly has not attained
any economic importance as a fruit in southern California.
1943 | TUCKER: FRANCESCHI 25
Of much greater consequence in this field, however, were the
frequent articles he wrote for newspapers and _ horticultural
journals. His book, “Frutti Tropicali e Semi-tropicali,” probably
embodies the findings of his long years of experience in this
field. It is a work of some 260 pages, contains numerous illus-
trations, and descriptions of 727 species. The majority of these
descriptions are rather brief, although for those which are of more
economic importance he gives much more detailed information.
An English translation of this book from the Italian would be a
real contribution to the literature of subtropical horticulture in
this country.
The total number of different kinds of plants that Franceschi
grew during his stay in Santa Barbara has not yet been fully
worked out. Among the miscellaneous papers in the collection of
his business correspondence, stored at the Herbarium of the
University of California, is a typewritten list of plant names, con-
tained on 114 sheets, with the following inscription pencilled on
the first sheet: “List of Seeds and Plants Tried Out in Santa
Barbara, California, by Dr. F. Franceschi.” The number of
genera is approximately 796, of species, varieties, and horticul-
tural forms, approximately 2,129. Exactly what significance may
be attached to this list is a questionable matter, however. There
is nothing to indicate who compiled it or when this was done. It
appears to have been prepared without the exercise of very much
care, and in no case is information given as to the results of the
trial introductions.
The total number of his new plant introductions is a debatable
matter. In the catalogues of the Southern California Acclima-
tizing Association which he published at irregular intervals, and
later, in the price lists of the Montarioso Nursery, he points out
the plants that he claims were first offered by him in the horticul-
tural trade in this country. The total number of these—species,
varieties, and horticultural forms—mentioned in his catalogues
and price lists from 1896 to 1914 (see bibliography) reaches
nearly 900.
These claims, however, cannot all be accepted at their face
value. There are occasional instances where species that he
claimed to have introduced (“plants first offered by us in the
United States’’) can be shown actually to have been offered by
nurserymen prior to his coming to California. A single example
is Phoeniz canariensis, the “‘Canary Island Date Palm.” Franceschi,
in his “Condensed Catalogue and Price List’? for 1908, indicates
that this species was one of his introductions, but, according to
Mr. H. M. Butterfield this palm was offered by John Rock at San
Jose as early as 1877. Moreover, the lack of consistency with re-
gard to some of his claims, in catalogues of different years, casts
a shadow of doubt on their validity. For instance, in his cata-
logue for 1908, he claims to have been the first in this country to
26 MADRONO [Vol.7
offer the following species (among numerous others): Cinnamo-
mum Camphora, Euphorbia pulcherrima, Leucadendron argenteum,
Magnolia grandiflora, and Persea gratissima. In his catalogues for
the years 1896, 1897, and 1900, however, he does not make this
claim for any of these species, although they are all listed. How-
ever, the number of errors of this sort is probably small, although
a careful check of his catalogues against earlier horticultural
literature is necessary before their extent can be ascertained.
After making due allowance for these errors, the remainder is
still truly remarkable. F. W. Popenoe, said of his work (Journal
of Heredity, 13: 215, 1922) “His introductions are more numer-
ous than those of any one man (in the United States), and many
of them are now widely grown in the land of their adoption.”
No evaluation of Dr. Franceschi’s contributions to horticulture
in this country would be complete without a consideration of the
influence of his personality on those of his contemporaries with
whom he came in contact. His wealth of horticultural and
botanical knowledge, gained from years of travel, observation,
and practical work, coupled with his untiring interest in plants,
were a constant source of inspiration to plant lovers wherever he
went.
The plants he introduced stand as living reminders of his un-
tiring work, and his memory will long be perpetuated at Santa
Barbara in his old home, “‘Montarioso,” which today is a city park,
named in his honor, “Franceschi Park.”’
BIBLIOGRAPHY
The following list, although not complete, probably contains a substantial
part of the literary work of Franceschi.
Bamboos. Gard. Chron. Ser. 2, 6: 773-774. 1876.
Bamboos in California. Pac. Gard. July, August, September, 1908.
Bauhinias. Pac. Gard. November, 1908.
Behavior of alien plants at Santa Barbara. Ann. Rep. Am. Breeder’s Assoc. 6.
1911. .
Colletia cruciata. Gard. Chron. Ser. 2, 9: 243, fig. 43. 1878.
Dasylirion longifolium. Soc. Tose. Ort. Bull. 15: 112-113, pl. 6. 1890.
Erythea armata. Gard. Chron. Ser. 3, 20: 424, fig. 74. 1896.
Erythea edulis at home. Gard. Chron. Ser. 3, 13: 507-508. 1893.
Erythea edulis at Santa Barbara. Gard. Chron. Ser. 3, 22: 157. 1897.
Eugenias for California. Rural Californian. April, 1910.
Euphorbia pulcherrima at Santa Barbara, California. Gard. Chron. Ser. 3, 21:
124, fig. 35. 1897.
Fifteen years experience in southern California. Pac. Gard. December, 1908,
January, February, March, 1909.
Fioritura del Rhododendron longifolium. Soc. Tose. Ort. Bull. 8: 63-65. 1883.
Frutti Tropicali e Semitropicali. Instituto Agricolo Coloniale Italiano, Florence.
1915.
Introductions from Australasia. Pac. Gard. April, 1909.
Japanese art of gardening. Pac. Gard. April, 1908.
La Domenica delle Palme. Soc. Tose. Ort. Bull. 3: 116-121, fig. 12. 1878.
Le Palmier de Vile de Guadalupe. Rev. Hort. 297-299. 1893.
New and little known trees suitable for southern California avenues. Pomona
Coll. Journ. Econ. Bot. 1: No. 4. December, 1911.
1943] REVIEWS 27
Notes on hardy bamboos. Gard. Chron. 1228-1229. 1872; Gard. 2: 316-318.
1872; (Ueber die harteren Bambus-Arten. Hamburg Gart. Blumenzeit 28:
417-423. 1872).
Olea fragrans. Soc. Tose. Ort. Bull. 1: 273-275. 1876.
Passato, presente e futuro della olivicultura in Tripolitania. L’Agricoltura
Coloniale. 19: 201-204. 1925.
Pittosporum. Pac. Gard. October, 1908.
Pritchardia filamentosa Wendl. Soc. Tosc. Ort. Bull. 1: 116-118, fig. 1. 1876.
Progressi di frutticultura a Tripoli nel 1923-24. L’Agricoltura Coloniale. 18:
309-313. 1924.
Santa Barbara exotic flora. Santa Barbara. 1895.
Trees from South Asia acclimatized in southern California. Forester. 4: 76-77,
129-130. 1898.
Tropic and semitropic fruits for southern California. Pac. Gard. December,
1907, January, 1908.
Weeds. Santa Barbara.
Yucca angustifolia Pursh. Soc. Tose. Ort. Bull. 11: 10-11, pl. 1. 1886.
Yucca filifera Chabaud. Soc. Tosc. Ort. Bull. 14: 278-280, pl. 9. 1889.
CATALOGUES
General descriptive catalogue. May, 1896, no. 4.
General catalogue. May, 1897, no. 5.
General catalogue. Easter, 1900, no. 6.
Condensed catalogue and price list. 1908.
Montarioso Nursery, descriptive price list. November, 1910, no. 2.
Handbook and price list. 1911.
Montarioso Nursery, descriptive price list. March, 1912, no. 3.
Supplementary price list of the Montarioso Nursery. 1914.
Department of Botany,
University of California, Berkeley,
May, 1941.
REVIEWS
Ceanothus. Part I. Ceanothus for Gardens, Parks and Roadsides,
by Maunseryt Van Renssevarr. Part II. A Systematic Study of
Ceanothus, by Howarp E. McMinn. Pp. xii+ 808. A Publication
of the Santa Barbara Botanic Garden, Santa Barbara, California.
Gillick Press, Berkeley, California. 1942. $2.50.
If there is anyone who admits the place of decorative plants
in the general scheme of things but thinks that botanists—above
all taxonomists and their tools, herbaria and botanical gardens—
have little or “nothing to do with the case,’ may he be given a
copy of “‘Ceanothus”’!
And may it be opened at the preface which, as G. B. S. long
ago maintained is apt to be the most important part of a book.
William Lassiter, Major General, United States Army, Retired,
has written this one which not only reviews the work and outlines
its purpose but places anyone who has anything to do with orna-
mental plants in the mood to desire greater use, understanding
and appreciation of them.
The book is divided into two parts: the first by Van Rens-
selaer is devoted to a consideration of the species, alphabetically
arranged, known in cultivation; the second, by McMinn is a
28 MADRONO [Vol.7
detailed taxonomic account of the entire genus. The former is
followed by an evidently carefully considered exposition of the
propagation and cultivation of these shrubs prepared by A. J.
Stewart, Horticulturist of the Santa Barbara Botanical Garden,
and the latter by a chapter of twenty-odd pages, entitled “‘Dis-
tributional History and Fossil Record of Ceanothus,’ by Herbert
L. Mason. Even those primarily interested in the living plants
will read Dr. Mason’s lucid account of the probable evolution of
the group and its species with interest because he gives a glimpse
of the relationship of old floras with those of today, and by his
thoughtfully interpreted observations throws light on the prob-
able development of a number of the forms treated with doubt as
distinct entities.
McMinn precedes his key to the fifty-five species and about
half as many varieties (and the descriptive text) by a “General
Discussion”; incidentally this includes the formal description of
the genus. The discussion itself comprises about ten pages not
counting some twenty more occupied by distributional maps,
probable relationship charts, natural hybrid charts, chromosome
“pictures” and a list of the specimens from which they were
drawn (this commendable). There is also a plate depicting well
the two sections into which the genus is readily divisible. The
seven distributional maps are clear and conveniently show on
each one several species and their variants. With the aid of these
various charts the author explains, apparently with considerable
confidence, his taxonomic conclusions and his reasons for arriv-
ing at them. The reviewer has no knowledge of Ceanothus what-
soever but anyone with general experience in classification will, it
seems to him, have the impression from the author’s discussion,
and that of Mason, that the taxonomy even with present knowl-
edge could have been on a sounder basis. Perhaps a number of
the entities recognized as species of equal merit could have been
differently evaluated considering the fact that a simple geographi-
cally limited, variable, often solitary character has frequently
been accepted as “specific.” However the author believes that
species can be proved by experiment. Granting that this is so,
it is my impression that experimental taxonomic methods, so to
speak, have more often proved than disproved the fundamental
soundness of specific lines and apparent relationships as they were
previously suggested by the more capable botanists of yesterday
and entirely from morphological and geographical data. Mc-
Minn’s interest ...I almost wrote enthusiasm ... in “‘ex-
perimental methods” has, I cannot help but sense, inhibited his
own expression of what he believes is actually happening (or has
happened) in the history of these fascinating and plastic plants.
This unfortunately (from the standpoint of practical simplicity)
has resulted in his not anticipating facts (as he hints himself, for
example page 191 and elsewhere) that the methods he advertises
will probably at least in many cases be able to prove. In any
1943] REVIEWS 29
case his keys and remarks are clear, and others can follow his
reasoning. His obvious knowledge of the group is little short
of amazing even with due appreciation of his indebtedness to the
work of previous students, notably Jepson and the latter’s pupil,
J.T. Howell. To the former he pays the compliment of using his
method of citation of references and specimens. In this connec-
tion one may remark the excellent, distinctive typography. There
is a lovely color plate of C. purpureus and innumerable photographs
uniformly of exceptional beauty and value as well as a number of
good drawings.
Finally let us turn to the descriptive account of the seventy-
odd species and a number of varieties distinguished by Van Rens-
selaer in cultivation. He describes them in the idiom of the
horticulturist, usually adding some remarks as to distribution,
where cultivated and the growing conditions required. It would
have been desirable if, besides the index to the entire book, page
references after the descriptions had been given in each part to the
other part. In this case some discrepancies in names used for the
same plants would have been discovered. For example on page
14 we find the name C. austromontanus instead of C. foliosus in
which it is included by McMinn, page 223. On page 30 the name
C. exaltatus is given as “‘a new horticultural designation” while
McMinn ignores it except as a variety of C. gloriosus, et cetera.
Some of these slips, or they may be differences in opinion between
the two authors, are going to confuse if not anger bibliographers,
not to mention certain professional botanists who, of course, are
almost God-like in the perfection of their own work!
Above the rare mistakes, here is a living work, jointly con-
ceived, jointly prepared, inspiring to everyone whether amateur
or professional in the garden, herbarium or laboratory, and
creating a closer bond of understanding, of friendship between all
who have to do with plants, as Major General Lassiter has hap-
pily phrased it. The contributors who made the book possible
are to be congratulated and thanked for supporting so worthwhile
a project that is destined to become a classic of its kind. Human-
ity needs many similar books and from them will be born the
realization that adequately financed herbaria and gardens must
always be the basis for them.—J. Francis Macsrinz, Field Museum
of Natural History, Chicago.
Practical Plant Anatomy. By Avriance S. Foster. Pp. 1-155.
D. Van Nostrand Company, Inc., New York. 1942. $2.50.
This compact book of fourteen chapters or “Exercises” is
spirally bound in flexible fabrikoid. Each exercise consists of a
brief but accurate résumé of both early and recent papers dealing
with the subject of the chapter, some discussion of the subject
matter and different points of view thereon, and suggestions for
study of selected materials and drawings to be made by the
student. A short but well-chosen bibliography completes each
30 MADRONO [Vol. 7
exercise. The book is primarily a laboratory guide for a course
in plant anatomy, although the discussions contain much factual
material not usually included in a strictly laboratory guide.
The suggestions for study, the material recommended, and
the diagrams and drawings which the student is supposed to
make, are all thoughtfully handled. The author has succeeded
in reducing the number of drawings required to a minimum, a
feature that will be welcomed by the immature students who want
to confine their laboratory work to a definite, set period. This
feature is not as reprehensible as it might seem to some pro-
ponents of many drawings, for diagrams are substituted for the
tedious, time-consuming detailed drawings to show relationships
among various tissues. Of course, any student interested in mak-
ing numerous detailed drawings of cellular types will find ade-
quate hints throughout the text if he looks for them! The total
absence of figures and illustrations mitigate against the use of the
book as a guide to a study of plant anatomy by those who are
unable to work under the direction of a trained instructor or
where the library facilities do not afford extensive reference
works.
No attempt was made to give directions for the preparation
of permanent microscope slides. But a far better point of view
has been taken in that free-hand, temporary mounts of fresh
materials are called for in nearly every exercise. No student
using this method will get the idea that xylem is always stained
red and that phloem and parenchyma cells are characterized by
an affinity for a blue or green stain! Brief directions for macer-
ating woody tissues and for the use of a few special reagents are
included in the short appendix. The index is gratifyingly com-
plete.
The method of approach is analytical and classificatory rather
than phylogenetic. The author’s reason for thus avoiding contro-
versy is well stated in the following excerpt from Exercise IV
(p. 89): “Since all methods for classifying plant tissues are open
to objection, the writer has adopted a non-committal and ‘practi-
cal’ attitude in this book. Instead of following any one scheme
of classification, the emphasis is placed first of all upon the salient
morphological features of the principal types of plant cells. These
cell types recur in various regions, ‘tissues’ and organs of the
higher plants, and a thorough knowledge of their form, structure,
development, and presumable function(s ) must constitute the
necessary analytical approach to anatomy.”
The format and typography is good and errors are extremely
few. The fabrikoid cover helps to protect the book from damage
where liquids may be spilled on the laboratory table and the
spiral binding permits the book to lie flat when opened to any
page. The numerous references in the text and the bibliogra-
phies following each exercise hold valuable keys to voluminous
literature on plant anatomy.—Ira L. Wieains, Stanford University.
1943] REVIEWS 31
John Torrey, a Story of North American Botany. By ANDREW
Denny Ropeers III. Pp. 1-352. Princeton University Press,
1942. $8.75.
Mr. Rogers has included in one volume an amazing wealth of
information on the history of botanical exploration in North
America. He is to be congratulated on presenting such a com-
prehensive picture of the life of John Torrey. Torrey, professor
of chemistry at Columbia and Princeton, was by avocation a bota-
nist. Through his pioneering efforts, the systematization of the
flora of this continent was begun. He arranged and described
the collections of numerous individuals and expeditions—Fre-
mont, Emory, Owen, Whipple, Ives, the Mexican Boundary Sur-
vey, the United States Exploring Expeditions, and many others.
He was founder of the Torrey Herbarium, now housed at the
New York Botanical Garden, and of the United States National
Herbarium. His influence was extended by his association with
the foremost botanists of his time, one of his first proteges being
Asa Gray. The large amount of material from original sources
which is made generally available for the first time in this book
is intensely interesting as well as invaluable to a study of the his-
tory of botanical exploration.—Mitprep Maruias, Department of
Botany, University of California, Berkeley.
Geographical Guide to the Floras of the World. Part I. By S.F.
BuakE AND Atice C. Atwoop. United States Department of Agri-
culture, Miscellaneous Publication 401, pp. 1-336. Washington,
DAC.) June, 1942. $0.75.
Here is the first part of a catalogue, unique in its field, that is
destined to become one of the most useful books in the entire
science of botany. Its value has been adequately demonstrated
to the reviewer many times during the short time that the book
has been in his possession. The work is a bibliographic catalogue
listing in geographic order the floras and floristic accounts of the
various geographic units of the world. In general, only complete
works are included, but for little known regions collector’s lists
are sometimes cited. Both general and local floras are included
and most of the entries are annotated to indicate the content of
the work. Part one deals with Africa, Australia, insular areas,
North America and South America. To the authors are due the
thanks of the entire botanical profession for performing so well
a task that doubtless entailed much drudgery.
It is to be hoped that this work may serve as a stimulus to
inspire some bibliographically inclined individual to compile and
publish a companion volume which will guide students to the
literature dealing with taxonomic accounts, in whole or in part,
of the families and genera of the flowering plants——Hersert L.
Mason,
32 MADRONO [Vol. 7
Eriocaulaceae, Avicenniaceae, Verbenaceae. By Haroitp N. Mot-
DENKE, in Flora of Texas, edited by C. L. Lundell. Vol. 3, pt. 1,
pp. 1-87. University Press in Dallas. Southern Methodist Uni-
versity. 1942. $1.50.
This is the first part to be published of a proposed “
10-volume work, each volume to contain approximately "700
pages. Volume 1 and 2 will contain the history of botanical
exploration, the key to the families, a catalogue of all species,
and maps showing distribution according to counties.”” The pro-
posed work is unlike most modern floras in that each part will be
a complete monographic unit in itself and will not, apparently,
be arranged in taxonomic sequence by families.
This first part by Dr. Harold N. Moldenke, of the New York
Botanical Garden, is notable for the length of the generic and
specific descriptions, complete synonomy, citation of numerous
specimens, and extensive (probably complete) citation of refer-
ences to publications where the species have been treated by
earlier authors. The number of specimens cited for each family,
together with an enumeration of the herbaria where they are
deposited, is given in a footnote below each family description.
The keys to genera and species are dichotomous and separate
varieties as well as species. In most instances they use qualita-
tive rather than quantitative characters, but sometimes make use
of geographic ranges to supplement comparatively weak morpho-
logical characters.
Brief but interesting comments deal with vernacular names,
geographic ranges, relationships, earlier misidentifications and
uses by aborigines and the early settlers.
The paper is of good quality and the printing clean-cut. Typo-
graphical errors are few. Both the taxonomic and editorial work
seem to have been done thoroughly, although Dr. Moldenke
appears to draw specific lines pretty finely.
The publication of this monograph of the three families as
they are represented in Texas initiates an ambitious project which
will be watched with interest by all plant taxonomists. Dr. Lun-
dell is to be congratulated for undertaking to see through the
tedious processes of writing and printing a work that will fill a
long felt need—an adequate and exhaustive flora of the Lone Star
State. It will be a source of great satisfaction to taxonomists if
the high standards established in this initial publication hold
throughout the entire work. It is the hope of the reviewer that
Dr. Lundell may be able to secure the aid of enough collaborators
to push the series through to completion within a reasonable
length of time.—Ira L. Wicearns, Stanford University.
MADRONO
A West American Journal of Botany
A quarterly journal devoted to important
and stimulating articles dealing with
plant morphology, physiology, taxonomy,
and botanical history. These volumes
should be a part of every botanist’s li-
brary and should be made accessible to
students of all universities and colleges.
Volume I, 1916-1929... $5.00
Volume II, 1930-1934 .. 5.00
Volume III, 1935-1936 . 5.00
Volume IV, 1937-1938 . 5.00
Volume V, 1939-1940 . . 5.00
Volume VI, 1941-1942 . 5.00
Single numbers....... 0.75
The subscription price of MADRONO
is $2.50 per year. We solicit your pat-
ronage,
Address all orders to:
William Hiesey Bus. Mgr.
Carnegie Institution of Washington
Stanford University, California
VOLUME VII NUMBER 2
MADRONO
A WEST AMERICAN JOURNAL OF
BOTANY
APR 3 0 i843,
‘.. :
NE =
Contents
——————
ay ges XEROPHYLLOUS SPECIES OF PHILADELPHUS IN SOUTHWESTERN NortTH
_ AMERICA, ORE CON WERLECICOGIR iUi (a aa Gna tdcs ai ey tee eG ULI) Biya Ts Ado 35
i New Species oF PHACELIA FROM SALINE VALLEY, Ca tirornia, Lincoln
OTT EAE) URLS RICOEUR AUD ES Oe a A 56
Giza mourririora Nurr. anp Irs Nearest Revatives, Thomas H. Kearney
AGU EE EL SCE CCOLESIVON SA Uh sa CON ku eee cs ABE Pome pay Won Uh a enablers byaeNeey 59
Review: Viola Brainerd Baird, Wild Violets of North America (Ethel
Crum), 46424. Cie RP IORI RANE UNS To Col oe A veg Eek CELL DR Sea SP 63
Achaea at North Queen Street and McGovern Avenue,
Lancaster, Pennsylvania
April, 1943
MADRONO
A WEST AMERICAN JOURNAL OF BOTANY
Board of Editors
Hersert L. Mason, University of California, Berkeley, Chairman.
LeRoy Asrams, Stanford University, California.
Epcar AnpeERsoN, Missouri Botanical Garden, St. Louis.
Lyman Benson, University of Arizona, Tucson.
Hersert F. Copetanp, Sacramento Junior College, Sacramento, California.
Ivan M. JoHnston, Arnold Arboretum, Jamaica Plain, Massachusetts.
Muprep EF. Marutias, University of California, Berkeley.
Bassett Macume, Utah State Agricultural College, Logan.
Marion Ownsey, State College of Washington, Pullman.
Secretary, Editorial Board—ANNeEtTTA CARTER
Department of Botany, University of California, Berkeley
Business Manager—Wutiiam Hirsey
North Queen Street and McGovern Avenue, Lancaster, Pennsylvania
or
Carnegie Institution of Washington
Stanford University, California
Entered as second-class matter October 1, 1935, at the post office at
Lancaster, Pa., under the act of March 3, 1879.
Established 1916. Published quarterly. Subscription Price $2.50 per year.
Completed volumes I to V inclusive, $25.00; each volume $5.00; single numbers
$0.75.
Papers up to 15 or 20 pages are acceptable. Longer contributions may
be accepted if the excess costs of printing and illustration are borne by the
contributor. Range extensions and similar notes will be published in con-
densed form with a suitable title under the general heading “Notes and News.”
Articles may be submitted to any member of the editorial board. Manuscripts
may be included in the forthcoming issue provided that the contributor pay the
cost of the pages added to the issue to accommodate his article. Reprints of
any article are furnished at a cost of 4 pages, 50 copies $4.10; 100 copies $4.50;
additional 100’s $0.85; 8 pages, 50 copies $5.95; 100 copies $6.60; additional 100’s
$1.30; 16 pages, 50 copies $8.35; 100 copies $9.35; additional 100’s $2.00. Covers,
50 for $2.75; additional covers at $1.65 per hundred. Reprints should be
ordered when proofs are returned. |
Published at North Queen Street and McGovern Avenue, Lancaster,
Pennsylvania, for the
CALIFORNIA BOTANICAL SOCIETY, INC.
President: H. S. Reed, University of California, Berkeley. First Vice-
President: N. T. Mirov, California Forest and Range Experiment Station, Uni-
versity of California, Berkeley. Second Vice-President: Jack Whitehead, Uni-
versity of California Botanic Garden, Berkeley. Secretary: Clarence R. Quick,
United States Department of Agriculture, 26 Giannini Hall, University of Cali-
fornia, Berkeley. Treasurer: William Hiesey, Carnegie Institution of Wash-
ington, Stanford University, California.
Annual membership dues of the California Botanical Society are $2.50,
$2.00 of which is for a year’s subscription to Madrofio. Dues should be
remitted to the Treasurer. General correspondence and applications for
membership should be addressed to the Secretary.
1943] MASON: ETHEL CRUM 33
ETHEL KATHERINE CRUM
It is with regret that we announce the death of Miss Ethel
Crum, Secretary to the Editorial Board of Madrofio and Assistant
Curator of the Herbarium of the University of California. She
passed away on January 5, 1943, at her family home in Lexing-
ton, Illinois. Her botanical career began with the attainment
of the degree of Master of Arts at the University of California in
1929 and her appointment as research assistant to Dr. W. L. Jep-
son at the time that the manuscript for volume two of “A Flora
of California” was in preparation. She assisted in the organiza-
tion of the material for many of the larger genera in this volume
but adopted as her own, the genus Potentilla, the manuscript of
which was accepted with some revisions. Further evidence of
her work may be noted in the Capparidaceae and in the Legu-
minosae. In 1933 she accepted a position on the curatorial staff
of the Herbarium of the University of California where she re-
mained until her death. Here, in addition to her duties, she in-
terested herself in the genera Monolopia and Pseudobahia and for
the Editorial Board of Madrofio she assumed the burden of edit-
ing and preparing manuscript for the printer. It is in this capac-
ity that she made her most significant contribution to botany.
Her early training in the classics, her command of the English
language and her experience in handling manuscript combined to
make her particularly suited to these tasks.
Her ideas of editorial procedure were well conceived and
fairly consistently adhered to. She believed that the first func-
tion of an editor of a botanical journal is to protect the author
from himself by preventing him where possible from making
hasty and unconsidered statements, from being misinterpreted or
misunderstood because of his lax use of grammar and from in-
jecting personalities into an article to the point of creating
offense. She edited to reduce the hazards of future embarrass-
ments to the author.
She had a positive concept of where the rights of the author
leave off and the rights and duties of the editor begin. These
rights fluctuated only with the age and experience of the author.
She believed, for instance, that the text of a manuscript, while
subject to such editing as is necessary to make it conform to the
style of the journal and the standard of fitness set by its editors,
is the concern of the author and when once the article is accepted
by the editors any changes affecting the argument or conclusions
may have the effect of impeaching the author and are wholly
unjustified. On the other hand, the precise wording of the title
of an article should be left to the editor subject to suggestion and
approval by the author. This concern stemmed from experience
in the use of bibliographies wherein a tremendous amount of time
is wasted investigating the subject matter of articles whose titles
ManroXo, Vol. 7, pp. 33-64. April 28, 1943.
apR 29 ‘43
34 MADRONO [Vol. 7
are inadequate or misleading. She insisted that a title should be
sufficiently specific to indicate not only the subject of an article
but something of its scope. She insisted that it should be ex-
pressed in language that can be readily translated into any
foreign tongue and still convey its precise meaning. This entails
the elimination from the title of all abbreviations and all forms
of speech that are colloquialisms or professionalisms.
As to the use of footnotes, it was her firm conviction that any-
thing worth saying in a footnote was worth incorporating into
the text. Footnotes, she maintained, interrupt the continuity of
thought of the reader and in addition spoil the appearance of the
printed page. They should, therefore, be avoided or reduced to
emergency use. | }
She believed that articles on isolated nomenclatorial problems
were largely a waste of time and space and constantly urged the
Editorial Board to adopt a rule refusing to print them unless they
were written in conjunction with a monograph of the group con-
cerned or were being used to urge the necessity of changes in or
additions to the rules of botanical nomenclature. It was her
further conviction that the available space in any journal is too
precious in these days to waste it on preliminary taxonomic revi-
sions and that no revision is worthy of publication until the author
has completed his work upon it. She deplored the word “‘pre-
liminary” because it is often taken as a promise of a more com-
plete treatment to follow, whereas the author is content to allow
his decisions to rest with the preliminary account. She argued
that if the author intended that the account would be final then
his use of the word “preliminary” in the title becomes a form of
dishonesty and can only be construed as an excuse to cover mis-
takes and decisions resulting from inadequate study.
She continually urged upon young authors the avoidance of
making in print promises as to the publication of future re-
searches. Too often circumstances do not permit the fulfillment
of such promises and the fact that they have been made may cause
some investigator to make fruitless search of the literature. On
behalf of the training of youthful and inexperienced authors she
was a crusader; she would spare no effort to see to it that they
were at least shown the paths of righteousness. Most of them
responded in good taste and evidenced a sincere appreciation of
her efforts to help them. A few have become soured on editors
for life.
Her work on Madrofio has set a standard of excellence in
editorship the maintenance of which will serve as a challenge to
her successors. She, perhaps more than any other person, has
influenced its recent editorial policy..
Miss Crum was born in Lexington, Illinois, on March 18, 1886.
She was educated at the University of Illinois and graduated with
the class of 1907. Her education stressed English literature and
1943] HITCHCOCK: PHILADELPHUS 35
the classics. From 1909 until 1929 she taught in the public
school systems of America, an experience of which she spoke as
“traversing an intellectual desert of the most barren sort.” This
may serve to explain better than more precise words her reasons
for shifting her career to botany. During the years from 19382
to 1988 she served as Secretary of the California Botanical Soci-
ety, relinquishing these duties to devote more time to her work
as Secretary to the Editorial Board.
Upon acquaintance one soon became impressed with the fact
that Miss Crum was a woman of outstanding intellectual bril-
liance, that she had an engaging personality and a ready flow of
wit and humor. She was devoted to her work and her several
hobbies crowded one another for her attention. Her passing
ends a career of uncommon usefulness; her life was a milestone
in the history of a journal HersBert L. Mason.
THE XEROPHYLLOUS SPECIES OF PHILADELPHUS
IN SOUTHWESTERN NORTH AMERICA
C. Leo HircucocK
There are, in southwestern United States and northern
Mexico, several species of Philadelphus which are xerophytic.
These plants have a rather heavy indumentum on the lower sur-
faces of the leaves as well as on the calyces and on the epidermis
of the twigs. Their leaves are small (1-3 cm. long), rather thick
and leathery, and entire. Whereas the flowers of Philadelphus
are usually borne in cymes or panicles of from three to many blos-
soms, the flowers of these members of the genus usually occur
singly (rarely in two’s or three’s) at the ends of short leafy lat-
eral branches. They have been placed in the group Microphylli
by Rydberg (No. Amer. Fl. 22: 168. 1905) with no indication
whether this group is of sectional or subgeneric rank. Since the
precise taxonomic status of this and corresponding groups of the
genus is not pertinent to this paper, the term “group” which was
frequently used by Rydberg will be used.
Floristically, these small-leaved species are particularly inter-
esting, since they occur chiefly in the lower levels of the larger
mountain ranges from Texas to California. Since localized
populations of each of the species are isolated by intervening
deserts, geographical races have become differentiated from one
another. As a unit they are readily distinguished from all other
North American species of Philadelphus, yet it is quite apparent
that they have been derived from, and are very closely related
to, certain species of the Mewicant (Rydb. op. cit.). A general
idea of the relationship of these sections as well as the geographic
distribution of the Microphylli is expressed in the accompanying
diagram (fig. 1).
Of the various species in the Meaicani, Philadelphus affinis
36 MADRONO [Vol. 7
Se =— ae
SONORA
1. RAFFINIS
2. MEXICANUS
3, COULTERI
4, PURPUS!}
5. MICROPHYLLUS
SuBsP ARGENTEUS
MACULATUS
CRINITUS
STRAMINEUS
F. ZIONENS!IS
PUMILUS
ARGYROCALY X 5
TYPICUS oe 3
: OCCIDENTALIS oe
6 SERPYLLIFOLIUS
7, MEARNSII
A. Sv8SP BIFIDUS
LO MNMOODS
Fic. 1. Phylogeny and distribution of Philadelphus (Microphylli).
Schlect., a species of central Mexico, has leaves that are large,
thin, and denticulate, 6-10 cm. long, and rather sparsely pubes-
cent. Its petals are 1.5—2 cm. long, its styles are densely hairy,
1943 | HITCHCOCK: PHILADELPHUS 37
and its inflorescence is a five- to eleven-flowered cyme. In gen-
eral appearance it is very suggestive of the species of Californici,
Coronarii, and Grandiflori. Another species of the same group,
P. mexicanus, is quite similar to P. affinis in most particulars, but
the plant is more pubescent and the inflorescence is greatly re-
duced, usually consisting of one, or of three, flowers at the ends
of short lateral branches. A third species, P. Coulteri, has much
smaller leaves (3-6 cm. long) which are more leathery and,
although denticulate, are very densely hairy; the petals are con-
siderably smaller (12-17 mm. long) and the inflorescence is
usually one-(three-) flowered. The styles are densely hairy as is
the upper part of the ovary.
Philadelphus Purpusi (Microphylli) although known from only
one collection, is a well-marked species that is apparently closely
related to P. Coulteri, having leathery, hairy leaves, and flowers
that are quite similar to those of its supposed relative, but differ-
ing in that it has entire, small (2—3.5 cm. long) leaves, petals that
are but 10-15 mm. long, and sparsely hairy styles. Philadelphus
microphyllus, with its various subspecies, has apparently been de-
rived from P. Purpusii, probably through the subspecies argyro-
calyx, argenteus, and crinitus, all of which have the pubescence and
leaves characteristic of P. Purpusi. In the first two of these sub-
species there is even a trace of the pubescence of the styles and
ovary that characterizes P. Purpusw and P. Coulteri.
The relationship of P. serpyllifolius and P. Mearnsii to the
other species of Microphylli is not so apparent, but it is possible
that they have been derived from P. microphyllus.
From the map it can be seen that the two subspecies of Phila-
delphus Mearnsit apparently are widely separated geographically.
It seems probable, however, that more thorough collecting in
northern Mexico will disclose that the plants are much more
widely distributed than extant collections indicate. The typical
form of the species was collected in 1892 and it was not until 1924
that the next collection was made, and that in a region remote
from the type locality. Only two collections have been seen that
were made since 1924, yet it seems certain that the plant is to be
found in other parts of New Mexico, Texas, or Mexico.
The presence of what I consider to be a local population of
P. microphyllus subsp. typicus in the Chisos Mountains of Texas is
dificult to explain. As mentioned under the treatment of that
subspecies, however, it is strongly suspected that these plants are
really genetically different from the plants in central New
Mexico, but aside from the fact that they are much greener in
aspect, distinctive taxonomic characters are not discernible from
the herbarium material examined.
Until the treatment of the genus by Rydberg in 1905, but few
specific names had been proposed in the group. Rydberg not
only recognized five species that had been described previously
38 MADRONO [Vol. 7
but proposed seven additional specific entities. Since his work
one or two additional names have been proposed. It was with
the hope of evaluating the validity of these various species that
this study was undertaken.
Herbarium material has been made available to me from the
following institutions, their abbreviations, as I have used them in
citing material, being shown in parentheses. University of Cali-
fornia (C), Gray Herbarium (G), New York Botanical Garden
(NY), Pomona College (P), Stanford, Dudley Herbarium (S),
United States National Museum (US), University of Washington
(Wash), State College of Washington, Pullman (WSC). To the
curators of these herbaria I express my sincere appreciation.
Key To SPECIES
A Lower surfaces of leaves grayish with short matted
tomentum and (usually) longer straight ap-
pressed hairs; stamens mostly 28 (26-32 or more
?)s ‘stylesless tham 1 mm. long? > 2... ee 3. P. serpyllifolius
AA Lower surfaces of leaves green or grayish but never
with tangled matted tomentum; stamens and
styles various.
B Base of styles and adjacent ovary densely pilose;
leaves coarsely strigose on both surfaces;
CalyCeSceray 2 oo nec ee ahr eee eee 4. P. Purpusii
BB Base of styles and adjacent ovary glabrous or
(very rarely) with few hairs; leaves and
calyces various.
C Stamens 16-24, filaments distinct; styles less
than 1 mm. long; pubescence of leaves
very coarse, the hairs thick in cross
SECTION \ pinch Se aunt ee eae ee 2. P. Mearnsii
CC Stamens more than 25, filaments often united
at the bases; styles usually at least 1 mm.
long; pubescence of leaves of slender
DAIS. 4-2 soties til cae as eee ee eee 1. P. microphyllus
1. PHILADELPHUS MICROPHYLLUs Gray, Mem. Am. Acad. Sci. I,
4:54. 1849.
Much branched, rounded shrub 1-2 m. tall; young branches
densely pubescent with appressed hairs, often silvery, bark more
or less reddish-brown to tan, usually exfoliating the second year,
older branches grayish to grayish-brown; petioles 1-3 mm. long,
leaf-blades ovate, ovate-lanceolate, or lanceolate, to lanceolate-
elliptic, 8-35 mm. long, 83-15 mm. broad, entire, often slightly
revolute, the apices rounded to acute, 3-nerved from base, pubes-
cent with hairs that are minutely papillate their entire length, the
upper surfaces either strigose with closely appressed hairs or hir-
sute with short erect hairs, but sometimes both strigose and hir-
sute, rarely glabrate, lower surfaces usually more densely hairy,
the hairs either fairly long and partially appressed, or shorter,
straight, and closely appressed; flowers borne singly or in threes
(twos) on pedicels 0.5—3 mm. long, at ends of short leafy shoots;
1943] HITCHCOCK: PHILADELPHUS 39
calyx-tube 2—3 mm. long in flower, enlarging to 3—5 mm. in fruit,
sparsely to densely strigose (glabrate), less commonly quite sil-
very with dense indumentum of tangled somewhat curled hairs,
lobes 3-5 mm. long, acute to acuminate, outer surface glabrous to
pubescent like the tube, inner surface always lanate; petals white
to cream, 6-17 mm. long, usually rounded but often emarginate ;
stamens (30) 32 to about 70, the filaments usually partially united
at least at base, but sometimes entirely free; styles (0.5) 1-2 mm.
long, free above, or more commonly, united to tips; stigmas 1.5—
3 mm. long, usually partially united.
Philadelphus microphyllus has been treated taxonomically in
various ways. Many authors, like Rydberg, considered the plant
as it occurs in central New Mexico to be specifically distinct from
the many closely related forms of California, Nevada, Utah, and
Arizona. It is true that it is largely a matter of interpretation
whether these other entities be considered varieties, subspecies,
or species, but certain characteristics upon which Rydberg based
his species appear to me to be but normal variations largely ac-
countable for by habitat. Still other differences which were used
to characterize some of these segregates are not at all constant
even though they may be genetic in origin, so that the mainte-
nance of some of Rydberg’s species, even as subspecific entities,
does not seem feasible. One such characteristic is that of the
degree of union of styles and stigmas. Although there is some
variation in the length of the styles and in the degree to which
they are united, I can find no consistent variation which can be
correlated with other distinctive morphological characters or
with geographical distribution. In some cases the styles of dif-
ferent flowers on the same specimen may be “united” and “‘par-
tially free’ (Hitchcock et al. 4148, C).
Although Rydberg listed definite and distinctive numbers of
stamens for several of his closely related species (P. stramineus 30-
40, P. microphyllus ca. 40, P. minutus ca. 60, P. pumilus ca. 30, and
P. argyrocalyz, P. ellipticus and P. occidentalis “‘many’’), the sta-
mens are so inconstant in number that species cannot be distin-
guished in this manner. The normal number of stamens is seem-
ingly a multiple of four, P. stramineus, for example, may have any-
where from thirty to forty, but usually has thirty-two, whereas
P. occidentalis, P. argenteus, and P. argyrocalyx usually have from
forty to sixty. There is often some variation in the stamens of
flowers of the same branch, so that neither P. pumilus nor P. stra-
mineus can be maintained on the basis of this character alone.
Leaf size, too, varies greatly, but the variation appears to be
due chiefly to the age of the branch on which the leaves are pro-
duced or to the amount of shade in which the plant grows. As
would be expected, plants which have grown in partial or com-
plete shade have large leaves and rather sparse pubescence. The
variation in leaf size due to branch age is best shown, perhaps, on
40 MADRONO [Vol. 7
a plant collected by Rydberg and Garrett (no. 9608). The leaves
of a young shoot are well over 3 cm. in length, those of an older
shoot are scarcely 1.5 cm. long. Ina Heller and Heller collection
(no. 3792) nearly all plants have leaves from 2.2 cm. to 8.5 cm.
in length; however, the sheet at Stanford bears a plant with
leaves less than 2 cm. long. Another collection (Harrison 6604)
made from a plant with leaves 6 to 8 mm. long had a two to three
year old branch with leaves 9 mm. long. The three twigs of this
branch bore new growth with leaves all 12 to 19 mm. long.
I can see no significant variations in the color, pubescence, or
degree of exfoliation of the bark. It seems, therefore, that the
characters that are dependable for subspecific delimitation are
the following, listed in order of significance: type and amount of
pubescence, number of stamens, leaf size, and flower size.
Key To SUBSPECIFIC VARIATIONS OF P. MicropHyLLus
A Petals with large purple spot at base ......... lb. P. microphyllus
subsp. maculatus
AA Petals not spotted with purple.
B Calyx-tube silvery, completely covered with
long hairs.
C Pubescence of calyx matted, consisting
of long straight hairs mixed with
more slender intertwined hairs;
leaves rather densely pubescent,
the hairs of lower surfaces rather
long and often not appressed, the
upper surfaces much less pubes-
cent and usually greenish ........ 1d. P. microphyllus
subsp. argyrocalyx
CC Pubescence of calyx usually not mat-
ted, but if so, the hairs all of same
type, often the pubescence too
sparse to be tangled.
D Calyx shaggy with long matted
slender hairs; both surfaces
of leaves grayish with long
slender hairs, these sometimes
somewhat matted on the lower
SUPPACES UI. taco e ene ee ne Ic. P. microphyllus
subsp. crinitus
DD Calyx with appressed pubescence
chiefly, the hairs not tangled;
leaves mostly distinctly green-
ish above, if grayish the hairs
not at all matted.
E Upper surfaces of leaves with
hirsute pubescence, the
hairs erect or nearly so.
F Stamens usually 40 or
more; leaves mostly
over 15 mm. long, the
pubescence of lower
surfaces not tightly
appressed ".......... ld. P. microphyllus
subsp. argyrocalyx
1943 | HITCHCOCK: PHILADELPHUS Al
FF Stamens usually 32; leaves
often less than 15
mm. long, the pubes-
cence of lower sur-
faces rather tightly
appressed.
G Hairs of upper sur-
faces of leaves all
essentially of the
same length ....
GG Hairs of upper sur-
faces of leaves
mainly short and
erect, the remain-
der less numer-
ous, longer and
more appressed .
EE Upper surfaces of leaves with
strigose pubescence, the
hairs appressed.
H Stamens usually 40 or
more, some usually
united half their
length; leaves mostly
over 15 mm. long ..
HH Stamens usually 32, free
or united only at
base; leaves seldom
over 15 mm. long ...
BB Calyx-tube not silvery, or at least not com-
pletely covered by pubescence.
I Leaves hirsute above, the hairs erect or
nearly so.
J Leaves usually 8-12 mm. long, the
pubescence of lower surfaces of
appressed hairs, that. of the
upper surfaces consisting of
many short erect hairs mixed
with less numerous, longer, sub-
appressed thairs’ <.4 9020: 4045 254
JJ Leaves 8-30 mm. long, the hairs of
lower surfaces not appressed, or
those of the upper surfaces all
of one type.
K Stamens usually 40 or more;
leaves commonly at least
15 mm. long, the hairs of
lower surfaces not always
tightly appressed ........
KK Stamens usually about 32 (to
40); leaves 8-15 mm. long,
the hairs of lower surfaces
APPLESSEG. che cea. eo eee
If.
ig:
la
P. microphyllus
subsp. stramineus
forma zionensis
. P. microphyllus
subsp. pumilus
. P. microphyllus
subsp. argenteus
le. P. microphyllus
lg
Id
subsp. stramineus
. P. microphyllus
subsp. pumilus
. P. microphyllus
subsp. argyrocalyx
le. P. microphyllus
subsp. stramineus
42 MADRONO [Vol. 7
II Leaves not hirsute above, the hairs
appressed.
L Petals 11-17 mm. long; hypanthium
with but little pubescence, this ;
chiefly at the angles ......... lh. P. microphyllus
subsp. typicus
LL Petals seldom over 10 mm. long;
hypanthium various but often
quite densely pubescent.
M Filaments usually united at
least one-third their
length; calyx-tube - sil-
very; leaves mostly 20-30
mm. long, not hirsute ... la. P. microphyllus
subsp. argenteus
MM Filaments free or united at
base only; calyx-tube not
silvery, or if so, leaves
usually less than 20 mm.
long.
N Stamens usually about
32; entire outer sur-
face of calyx quite
densely pubescent
with stiff straight ap-
pressed hairs, the
outer surface almost
completely covered
by shairs ee le. P. microphyllus
subsp. stramineus
NN Stamens usually 40 or
more; the outer sur-
face of calyx sparsely
pubescent or _ even
partially glabrous, in
any case not hidden
by JNaibS eo. ar eee li. P. microphyllus
subsp. occidentalis
la. PHILADELPHUS MICROPHYLLUS subsp. argenteus (Rydb.)
comb. nov. P. argenteus Rydb. N- Am. Bl 22: 17109 1905" ak:
argyrocalyx var. argenteus (Rydb.) Engl., Engl. & Prantl, Nat.
Pflanzenf. II, 18a: 193. 1930. P. microphyllus var. argenteus
(Rydb.) Kearney & Peebles, Journ. Wash. Acad. Sci. 29: 480.
1939. P. Palmert Rydb. N. Am. FI. 22: 178. . 1905; type: sierra
Madre, 40 miles south of Saltillo, Mexico, Palmer 2122. P. ma-
drensis Hemsl. Kew. Bull. 251. 1908; type: Sierra Madre,
Durango, Mexico, Seemann 2167.
Leaves mostly 20-35 mm. long, ovate-lanceolate or lanceolate,
pubescence of lower surfaces grayish-strigose, upper surfaces
sometimes glabrate; calyx more or less canescent with long
straight appressed hairs which are often mixed with shorter more
slender ones; petals 8-11 (12) mm. long, often pubescent on
outer surfaces near the base; stamens 40 or more, filaments
usually united at least at the base and often united to near the
1943] HITCHCOCK: PHILADELPHUS 43
tip, usually grouped in irregular phalanges; styles sometimes
sparsely hairy near base.
Type. Fort Huachuca, Arizona, Dr. T. E. Wilcox in 18938.
Range. Mountains of southeastern Arizona, south into So-
nora, Chihuahua, Coahuila, and Durango, Mexico.
Material seen. Arizona. Coronado Mts., Goldman 2379
(US); Santa Rita Forest Reserve, Griffiths 4194 (US). Santa
Cruz County: near Patagonia, Harrison 7180 (C, NY). Cochise
County: Bisbee, Carlson in 1915 (US); Fort Huachuca, Wilcox in
1898, type collection (NY), Wilcoz in 1873 (NY), probably
should read 1893, surely part of type collection; Wilcox in 1892
(NY), and Wilcor 2384 (US); Huachuca Mts., Ramsey Canyon,
Jones 24918 (P) and Millers Canyon, Goodding 147 (G, NY, US) ;
between Fort Huachuca and San Pedro River, Mearns 1540
(US); Chiricahua Mts., Barfoot Park, Blumer 1303 (G, NY, S,
US); Barfoot Fire Station, Eggleston 10766 and 10794 (US);
Upper Pine Canyon, Burrall in 1906 (US) ; Dos Cabezas, Lemmon
in 1881 (C). Graham County: Mt. Graham, Peebles, Harrison, &
Kearney 4449 and 4450 (US). Pima County: Santa Catalina
Mts., Lemmon & Lemmon in 1881 (C); Santa Rita Mts., Pringle in
1881 (G, NY, P, US) and Peebles & Harrison 3000 (US) ; Rincon
Mts., Spud Range, Blumer 3565 (C). Mexico. Cuinuanua:
Guayanopa Canyon, Sierra Madre Mts., Jones in 1903 (P); San
Luis Mts., Goldman 1432 (US). Sonora: San Jose Mts., Mearns
1617 (US). Coauuiza: Sierra Madre, 40 miles south of Saltillo,
Palmer 2122, type collection P. Palmeri (G); Sierra de Parras,
Purpus 4592 (G, US).
Philadelphus microphyllus subsp. argenteus approaches subsp.
stramineus and subsp. pumilus in the pubescence of the calyx, but
is larger-leaved than either and the pubescence of the leaves is all
appressed. It is also quite similar to subsp. argyrocalyx but dif-
_ fers because the leaves are not hirsute, their lower surfaces have
more appressed pubescence, and the pubescence of the calyx is
more uniform.
There is scarcely any difference between the type of P.
Palmeri and the type of subsp. argenteus except that the filaments
are more completely united in the former. Pubescence, leaf size
and shape, and flower size are alike in the two. Rydberg surely
had seen only the type specimen when he described P. Palmeri, but
the Purpus collection, made in 1910, is annotated by Rydberg as
P. Palmeri and shows that he must have had to change his concept
of the species considerably, since the flowers of that collection
have petioles that are but 1.5 mm. long (original description of
P. Palmeri specified 1.5 cm., although this is probably a typo-
graphical error), petals that are 9-12 mm. long (12 in original
description), styles that are not distinct, and a hypanthium that
is as long as it is in most of the other subspecies of the complex.
The only possible basis for maintaining P. Palmeri even as a sub-
AA MADRONO [Vol. 7
species would be the union of the filaments. However, much of
the material from Arizona that is surely P. argenteus has filaments
as completely united as are the filaments of the flowers of the two
collections of P. Palmeri (Pringle in 1881, Santa Rita Mts., and
Blumer 1303).
Philadelphus madrensis is being reduced to synonymy even
though the type (Seemann 2167, Sierra Madre, Durango) has not
been seen. That plant, it will be noted, was collected very near
the type locality of P. Palmeri, and the original description of P.
madrensis fits P. Palmeri quite well. The ‘1—3-flowered inflores-
cence, and subsessile, acute, entire, 8-nerved, sericeous-pilose
leaves” are similar to those of the type of the latter species and
therefore to those of subsp. argenteus. There is nothing distinc-
tive about the flowers, the stamens being more or less connate into
four phalanges and the ovary either glabrous or slightly pilose on
top. In fact one could scarcely compose a description that would
apply more closely to Philadelphus microphyllus subsp. argenteus
than does the description of P. madrensis except that the stamens
are said to be about thirty-two, whereas in the specimens of P.
argenteus I have seen there have been forty or more. This dis-
crepancy notwithstanding, it seems safe to assume that P.
madrensis is identical with P. Palmeri and therefore with subsp.
argenteus.
There is the possibility that this is the same plant as P. asperi-
folius (Koern. Gartenflora 16: 78. 1867) but in the absence of
proof of this identity, that name is not being used. Rydberg’s
separation of P. asperifolius on the basis “‘bark of previous year’s
growth not exfoliating” is inconclusive.
1b. PHILADELPHUS MICROPHYLLUs subsp. maculatus subsp. nov.
Foliis lanceolatis vel ellipticis-lanceolatis, 1.5—2.5 cm. longis,
ca. 4 (3-8) mm. latis, viridis, appresso-pubescentibus; calicibus
purpureis, cum pilis adpressis tenuibusque pubescentibus, cinera-
ceis, tubo ca. 3.5 mm. longo, lobis 4-5 mm. longis, abrupto-acumi-
natis; petalis oblongo-ovatis, ca. 12 mm. longis, albis, petalo
inferiore purpureo-maculato; staminibus 32-40, plus minusve in
phalangibus ex 2-8 filamentis conjunctis, interdum ad proximum
apicem connatis; stigmatibus ca. 8 mm. longis, saepe absoluto-
connatis; stylis ca. 1.5 mm. longis, ad basum sparse pilosis.
Leaves lanceolate to elliptic-lanceolate, 1.5—2.5 cm. long, ca.
4 (38-8) mm. broad, both surfaces greenish, the pubescence ap-
pressed; calyces purple, finely appressed pubescent and rather
cineraceous, the tube ca. 3.5 mm. long, the lobes 4-5 mm. long,
abruptly acuminate; petals oblong-ovate, ca. 12 mm. long, white
with distinct purplish blotches at base, sometimes with this purple
spot extending to upper half or even to tip of petal; stamens 32—
40, irregularly united into phalanges of 2—7 or 8 filaments, these
sometimes connate to near tip; stigmas ca. 3 mm. long, usually
1943] HITCHCOCK: PHILADELPHUS 45
completely united, styles ca. 1-5 mm. long, sparsely strigose at
base.
Type. In forest of large pines, forest floor of low vegetation,
mountain top, 7 kilometers southwest of Miquihuana, Tamaulipas,
Mexico, lat. 23° 42’ N., long. 99° 45’ W., elev. 3430 m., August 5,
1941, Stanford, Retherford, and Northcraft 690 (Wash).
This showy-flowered form of Philadelphus microphyllus is
closely related to subsp. argenteus, but differs not only in the
purple-spotted petals, but also in the narrower, greenish leaves.
This greenness may be due, in part at least, to the fact that the
plants were probably partially shaded.
It is known only from the type collection but the collectors re-
port that it was fairly abundant, that all plants seen had the
spotted petals, and that their collection, of about ten herbarium
sheets, was made from about four different shrubs.
lc. PHILADELPHUS MICROPHYLLUS Gray subsp. crinitus subsp.
nov.
Folia 15-25 mm. longa, supra alba, adpressa-strigosa, subtus
simplicatio-strigosa, calicibus extra albis implicatis-strigosis ;
petalis ca. 11 mm. longis; staminibus 48-68, filamentis basi con-
natis; stylis 1-1.5 mm. longis, connatis; stigmatibus connatis.
Leaves mostly 15-25 mm. long, pubescence of the upper sur-
faces grayish, strigose, the hairs appressed, pubescence of the
lower surfaces densely long strigose, the hairs often somewhat
tangled; calyces gray with long more or less matted slender
hairs; petals ca. 11 mm. long; stamens 48-68, united at base;
styles 1-1.5 mm. long, united; stigmas 3—4 mm. long, united.
Type. Rocky ground near top of Mount Livermore, Davis
Mountains, Jeff Davis County, Texas, June 4, 1928, E. J. Palmer
34347 (NY).
Other material seen. Texas. North side Mount Livermore,
Palmer 34364 (NY); Livermore Peak, Ferris & Duncan 2529 (NY,
S
The status of this entity is puzzling. By some workers it has
been mistaken for P. serpyllifolius, but it differs so markedly from
that species in pubescence, stamen number, style and stigma
length, and general appearance, that a close relationship between
the two seems doubtful. It is probable that it is much more
closely related to P. argyrocalyz or even to P. argenteus, both of
which it strongly resembles in general aspect, but the pubescence
of the calyx and leaves is so unique that it cannot be combined
with either. Since the flowers are so like those of the other sub-
species of P. microphyllus there is not an adequate basis for ac-
cording the plant specific recognition.
1d. PHILADELPHUS MICROPHYLLUs subsp. argyrocalyx (Wooton)
comb. nov. P. argyrocalyx Woot. Bull. Torrey Bot. Club 25: 452.
1898. P. serpyllifolius var., Gray, Pl. Wright. 2: 64. 1852. P.
ellipticus Rydb. N. Am. Fl. 22: 172. 1905; type: Mesilla Park,
46 MADRONO [Vol. 7
New Mexico, Tinsley in 1896, probably wrong locality (Woot.
& Standl., Contr. U.S. Nat. Herb. 19: 300. 1915).
Much like subsp. argenteus in general appearance but leaves
hirsute on upper surfaces and usually with less appressed, but
somewhat longer, looser, hairs on lower surfaces; calyces typi-
cally grayish with matted mixture of long straight and slender
curled hairs, but pubescence sometimes less abundant and calyces
then scarcely grayish; petals 8-11 (17) mm. long, often pubes-
cent near base on the outer surfaces; stamens usually 40 or more,
filaments commonly partially united, often united to near tips;
styles united, sometimes sparsely hairy near base; stigmas at least
partially united.
Type. Eagle Creek, White Mountains, Lincoln County, New
Mexico, Wooton 524.
Range. Mountains of south central and western New Mexico
into southeastern Arizona, south into Chihuahua, Mexico.
Material seen. New Mexico. Lincoln County: White Mts.,
Wooton 524, type collection (NY, US) and Wooton in 1901 and
1905 (US) ; West Peak, El Capitan Mts., Earle & Earle 220 (NY);
Mesilla Park, Tinsley in 1893, type collection P. ellipticus (NY,
US) ; Ruidoso Creek, Wooton in 1895 (US). Otero County: head
of Rio Fresnal, Barlow in 1911 (C); Cloudcroft, Sacramento Mts.,
Rehder 372 (US), Eggleston 14541 (US), Wooton in 1899 (US),
and Orcutt 1351 (US). Without definite locality, Wright 1101,
the plant which Gray called P. serpyllifolius var., therefore from
“mountain sides at the Copper Mines” (G, NY, US).
When Wooton described P. argyrocalyx he included therein
the Wilcox collection from Arizona. That plant was later made
the type of Rydberg’s P. argenteus. The two entities are similar
and undoubtedly they are closely related; in many areas they
grade into one another. Yet subsp. argyrocalyx, as it occurs in
Lincoln County, New Mexico, is very unlike any collections of
Philadelphus from Arizona, the pubescence of the calyx being
much more curled and matted, consisting not only of matted
curled trichomes but also of long straight hairs. The leaves are
conspicuously hirsute with short erect hairs, a condition which is
characteristic of nearly all material from southwestern New
Mexico but not of the plants from Arizona. However, material
becomes more and more like subsp. argenteus near the Arizona
border until, in the mountains of extreme southeastern Arizona,
the material is intermediate in nature between the two subspecies.
The following collections therefore are not quite identical with
the type collection and some are at least as similar to subsp.
argenteus as to subsp. argyrocalyz. All have hirsute leaves. NEw
Mexico. Sierra County: Mimbres Mts., Hillsboro Peak, Diehl
626 (P). Soccoro County: San Mateo Peak, Goldman 1745 (US).
Grant County: Burro Mts., Metcalfe 173" (€, G, NY, P)s, US),
quite typical as to leaves, the calyx almost the same as in material
1943] HITCHCOCK: PHILADELPHUS AT
from the type locality, the petals pubescent; Animas Peak, Gold-
man 1870 (US). Valencia County: mountains west of Grants Sta-
tion, Wooton 1109 (US). Arizona. Oak Creek, Goldman 2174
(US). Pinal County: Superstition Mts., Harrison 6604 (NY).
Cochise County: Chiricahua Mts., Price in 1894 (S) and Jones
283827 (C, P), Barfoot Park, Blumer 1291 (G, NY, S), very close
to P. argenteus, but calyx very sparsely hairy, the leaves hirsute,
however; White River Canyon, Chiricahua Mts., Toumey in 1894
(C); Upper Pine Canyon, Burrall in 1906 (US). Pima County:
Mt. Lemmon, Santa Catalina Mts., Livingston & Thornber in 1906
(NY); Santa Catalina Mts., Lemmon & Lemmon 186 (C) and Lem-
mon 170 (G); near Mud Springs, Harris C16344 (US). Graham
County: Pinaleno Mts., Bonita, Munz 1248 (P). Mexico. Cui-
HUAHUA: Colonia Juarez, Sierra Madre Mts., Jones in 1903 (NY,
Bs US).
Other collections are more intermediate with subsp. typicus:
Hillsboro Peak, Sierra County, New Mexico, Metcalfe 1323 (C, G,
NY,P,US). The plants of this collection have silvery leaves and
united stamens (characters of subsp. argyrocalyx but the calyces,
as in subsp. typicus, are nearly glabrous.
le. PHILADELPHUS MICROPHYLLUS Gray subsp. stramineus
(Rydb.) comb. nov. P. stramineus Rydb. N. Am. Fl. 22: 172.
1905.
Much the same as subsp. occidentalis, but the calyces usually
canescent with stiff straight appressed hairs; leaves mostly 10-25
(probably averaging about 15) mm. long, pubescence of both
surfaces usually appressed, if hairs more erect, then all approxi-
mately the same length; petals 7-10 mm. long, their outer sur-
faces sometimes pubescent at base; stamens 30—50 but usually 32,
the filaments from scarcely united to united to near the tips.
Type. White Mountains, Mono County, California, Shockley
in 1888.
Range. Foothills and lower levels of the Charleston Moun-
tains of Nevada, White Mountains of California, and San Pedro
-Martir Mountains of Lower California. Probably more wide-
spread than present collections indicate.
Although it is believed that subsp. stramineus is a valid entity,
if it is to be maintained it must be on a different basis than that
used by Rydberg. It was keyed by him on the following charac-
ters: “Leaf blades equally grayish on both surfaces; sepals acute;
bark of the old stems straw colored,’ whereas P. microphyllus, in
contrast, had “leaf blades paler beneath; sepals acuminate; bark
of the old stems gray.” The leaf-blades of the Shockley collec-
tion are paler on the lower surfaces than on the upper; the sepals
are just as sharp-pointed as are those of much of the material of
P. microphyllus subsp. typicus; lastly, although it is true that the
older bark of the type collection of P. stramineus is not grayish, in
other collections of that entity from the White Mountains (Cassel
48 MADRONO [Vol. 7
262), the bark is as weathered and gray as that of any material
from New Mexico.
The unique characteristics of the entity are as follows: the
calyces of subsp. stramineus are grayish with dense stiff straight
hairs, in this respect being more similar to subsp. argenteus than
to either subsp. occidentalis or subsp. typicus. Usually there are
fewer stamens (thirty-two being the commonest number) per
flower than in subsp. argenteus, subsp. occidentalis, or subsp.
typicus, but since there are occasional plants with as many as forty
stamens per flower (Percy Train 2178, from the Charleston Moun-
tains), too much importance should not be ascribed to this pecu-
liarity of subsp. stramineus.
Material seen. Nevapa. Clark County: Charleston Mts.,
Clokey & Clokey 7133 (NY,S, Wash), Clokey 5458 (NY,S, Wash),
Clokey 5490 (NY,S, Wash), Heller 10998 (C, NY, S, US), Clokey
70940 (G), Percy Train 2178 (NY), Jones in 1927 (P), and Jaeger
in 1925 (P). Caxirornia. Inyo County: White Mts., Black Can-
yon, Cassel 262 (S) and Duran 540 (C, NY, P, S, US), Wyman
Canyon, Ferris 6966 (S); Crag Canyon, Grapevine Mts., Gilman
3262 and 3259 (P). Mono County: White Mts., Shockley in 1888
(NY, type) and Shockley 454, probably type collection also (C,
G, S, US). Mexico. Lower Catirornia. San Pedro Martir
Mts., Vallecitos, Goldman 1223 (US); San Pedro Martir, Brande-
gee in 1898 (C) and Heller in 1902 (C); La Encantada, Wiggins &
Demaree 4947 (C, G, NY, P,S, US) ; east of La Encantada, Melinge
in 1981 (P,S, US). Some of these collections from Lower Cali-
fornia approach subsp. pumilus in that the upper surfaces of the
leaves tend to be hirsute with short erect hairs; further collections
may prove that they more properly belong with that entity.
Material from Zion National Park is unique in that the pubes-
cence of the upper surfaces of the leaves is short hirsute rather
than strigose. In this respect the plants resemble subsp. pumilus,
but that entity has long, appressed hairs mingled with the short
erect ones, whereas the material from Zion Canyon is hirsute with
hairs of uniform length. Also, the leaves of the plants from Cali-
fornia are smaller. This variant from Utah is therefore accorded
minor status as follows.
1f. PHILADELPHUS MICROPHYLLUS Gray subsp. sTRAMINEUS
(Rydb.) C. L. Hitche. forma zionensis forma nov.
Planta a subsp. stramineus differens: folia supra pilosula.
This form differs from subsp. stramineus, in the leaves which
are hirsute above with short semi-erect hairs of uniform length.
Type. Near summit of Lady Mountain, Zion National Park,
Washington County, Utah, June 19, 1928, T’. Craig 1439, Pomona
College Herb. no. 184362.
Range. Known only from Zion Canyon, Utah.
Material seen, in addition to the type. Zion Canyon, Jones in
1923 and in 1925 (P), Garrett R2669 (NY), and Woodbury 19
(US); “Utah,” Ward 699 (US).
1943] HITCHCOCK: PHILADELPHUS 49
lg. PHILADELPHUS MICROPHYLLUS subsp. pumilus (Rydb.)
comb. nov. P. pumilus Rydb. N. Am. Fl. 22: 173. 1905.
Much like subsp. stramineus, especially as to pubescence of the
calyces, but leaves usually averaging no more than 8 mm. long,
their upper surfaces hirsute with many short stiff erect hairs
mixed with fewer and longer more appressed hairs; petals 6-10
mm. long, usually slightly pubescent on the outer surfaces near
base; stamens 32 (to 50), the filaments free or united at base
only; styles from completely united to almost entirely free.
Type. Tamarack Valley, San Jacinto Mountains, Riverside
County, Hall 2500.
Range. Apparently restricted to the San Jacinto and Santa
Rosa Mountains of southern California at 7000-8500 feet eleva-
tion. Possibly also in Lower California (see under subsp. stra-
mineus ).
Material seen. Catirornia. Riverside County: Santa Rosa.
Mts., Munz 15392 (Wash) ; San Jacinto Range, Hall S00 (S); San
Jacinto Mts., near Tamarack Valley, Hall 2500, type collection
(C, NY,S, US); Dark Canyon, Munz 8762 (P) and Munz & John-
ston 8738 (P); Long Valley to Palm Springs Trail, Jaeger 1015 (C,
P) ; above Chino Canyon, Jaeger in 1922 (S, US).
The variation in the degree of union of the styles is easily seen
in this phase of the species. The flowers of the type collection
have styles completely united, even the stigmas being united over
half their length in a few cases. The collection made by Munz
and Johnston is so similar to the type that Rydberg surely would
have called them conspecific, yet the styles of this collection are
free nearly to the ovary. It is because of such instances that the
union or non-union of the styles is considered of secondary impor-
tance as a diagnostic character.
lh. PHILADELPHUS MICROPHYLLUS Gray subsp. typicus nom.
nov. P. microphyllus Gray, l.c.; Rydb. N. Am. Fl. 22: 172. 1905.
Leaves (12) 17-35 mm. long, ovate-lanceolate to lanceolate,
strigose on both surfaces to glabrate above, the hairs all appressed
but the upper surfaces usually greenish; calyces glabrate to mod-
erately strigose, the tube 2—4 mm. long, usually strigose on the
angles, the lobes often glabrous on the outer surfaces, moderately
lanate on inner surfaces; petals 11-17 mm. long, not pubescent on
the outer surfaces; stamens mostly 40-52 (32), practically free
except for few that are geminate.
Type. Eleven miles above Santa Fe, on Santa Fe Creek, New
Mexico, June to July, 1847, Fendler 266.
Range. Central and northern New Mexico, southern Colo-
rado, and Apache County, Arizona; also from the Chisos Moun-
tains of Texas.
This, the typical form of the species, characterized chiefly by
the large flowers and sparse pubescence of the calyx, is the only
phase of the species found in northern New Mexico. It inter-
50 MADRONO [Voleia
grades with subsp. occidentalis on the north and west, and with
subsp. argyrocalyx on the south. Although most of the flowers of
the type collection have about thirty-two stamens, the number
of stamens present in the bulk of the material seen is between
forty and fifty-two.
Material seen. New Mexico. Rito de las Frijoles, Cockerell
in 1912 (US). Socorro County: Beartrap Canyon, San Mateo
Mts., Eggleston 18654 (NY, US); Hop Canyon, Magdalena Mts.,
Diehl 463 (P) and Herrick 608 (US), Copper Canyon, Magdalena
Mts., Goldman 1669, intermediate with subsp. argyrocalyx (US).
Bernalillo County: Albuquerque, Jones in 1884 (P). Valencia
County: Grant’s Station, Wooton in 1892, approaching subsp.
argyrocalyx (NY). Sandoval County: Balsam Park, Sandia Mts.,
Ellis 107 (NY, US) ;.Sandia Mts., Herrick in 1898 (US); Placitas,
Sandia Mts., Wooton in 1910 (US); Ellis Ranch, Sandia Mts.,
Wooton in 1910 (US); Guadalupe Canyon, San Mateo Mts.,
Eggleston 18736 (US). Santa Fe County: Santa Fe Creek,
Fendler 266, type collection (G, NY, US); Santa Fe Canyon,
Heller § Heller 3792 (G, NY, P,S, US) and Wooton in 1910 (US) ;
near Santa Fe, Arséne §& Benedict 15742 (US). San Miguel
County: mouth of Indian Creek, Standley 4547 (G, NY, US).
Taos County: Taos River Canyon, Nelson 11471 (C, G); Red
Canyon, near Questa, Hitchcock et al 4148 (C, P, Wash, WSC).
Cotorapo. Las Animas County: Brantley Canyon, Osterhout
2077 (C, NY, P); Mesa de Maya, 60 miles east of Trinidad,
Rollins 1835 (NY, WSC). Montrose County: Cimarron, Nelson
& Nelson 425 (C), with some doubt as no petals present. Fremont
County: west of Parksdale, Jones in 1913 (S); Royal Gorge,
Bacigalupi 1009 (NY, P, S), without petals, hence some doubt;
Canyon City, Brandegee 84 (C). Utan. Grand River near Moab,
Jones in 1915, intermediate with subsp. occidentalis (NY). Art-
zona. Canyon de Chelly, Nelson 36, almost surely belongs here
but no flowers on plant so identity cannot be certain (US).
Several collections from the Chisos Mountains, Texas, must
be placed here, even though they have smaller flowers than those
common to the material from New Mexico. At first glance these
Texan plants appear to be sufficiently different to be maintained
as a separate entity, but it has been found to be impossible to
construct a key that will separate them from subsp. typicus.
Their leaves are quite green when dried, but this condition may
be due to the care with which they were preserved; the stamens
vary in number from twenty-nine to forty per flower; the flowers
are about the size of those of subsp. occidentalis, but the leaves
and calyces are more nearly glabrous than they are in that plant.
It is entirely possible that these plants will prove to be genetically
distinct from either subsp. typicus or subsp. occidentalis, but field
study will be necessary in order to detect distinguishing charac-
ters, if such there be. Trxas. Brewster County, Chisos Moun-
1943] HITCHCOCK: PHILADELPHUS 51
tains: Mt. Emory, Warnock 866 (US), Upper Boot, Cory 7076 (P)
and Cory 7077 (G); 2 miles southwest of Boot Spring, Moore &
Steyermark 3160 (C, G, NY, S); Lost Mine Peak, Ferris § Duncan
2856 (NY, S); with no further locality, Mueller 8013 (NY) and
Havard 40 (G).
li. PuHILaDELPHUS MICROPHYLLUS Gray subsp. occidentalis
(Nelson) comb. nov. P. occidentalis Nelson, Bull. Torrey Bot.
Club 25: 874. 1898. P. minutus Rydb., N. Am. Fl. 22: 178.
1905; type, Black Canyon of the Gunnison, Colorado, Baker 266.
P. nitidus Nelson, Bot. Gaz. 42: 54. 1906; type, Sapinero, Colo-
rado, Wheeler 425. Second cited specimen was Baker 266, hence
P. nitidus is surely the same as P. minutus.
Pubescence much the same as in subsp. typicus, the leaves
often glabrate above, mostly 10-16 (to 25) mm. long, elliptic-
lanceolate to lanceolate; petals 9-11 mm. long, not pubescent on
outer surfaces; stamens usually about 40 (29-45), united at base
only, if at all; styles wholly united or free one-third to one-half
their length.
Type. Near Rock Springs, Sweetwater Canyon, Wyoming,
July 25, 1897, Smith 3595.
Range. Central and northern Colorado, southwestern Wy-
oming, and eastern Utah.
The subsp. occidentalis merges with the subsp. typicus and is
distinguished from that phase with some difficulty. In general,
however, it has smaller flowers and leaves. In the western part
of its range it also intergrades with subsp. stramineus but can
usually be distinguished from that plant because of its greater
number of stamens (forty or more as compared with thirty-two)
and less pubescent calyces.
There can be little doubt that P. nitidus and P. minutus are
the same entity since Nelson cited three collections of P. nitidus,
one of which was collected by Baker (no. 266), in Black Canyon,
Colorado. This is the same collection which Rydberg, a year
previously, had selected as the type of his P. minutus. Nelson
specified that P. nitidus had “styles distinct down to the ovary.”
Rydberg claimed P. minutus had “styles united one-half to two-
thirds their length.” Although I have seen no flower which I
would describe as having styles distinct to the ovary, I am
convinced that the degree of union of the styles is not a particu-
larly dependable character. One of the other two collections
which Nelson cited under P. nitidus (Jones 6803), from Belknap,
Utah, has the styles united to the stigmas in the flowers I have
examined. Thus it seems evident that free or partially free
styles are not a constant characteristic of subsp. occidentalis.
Neither are the styles of subsp. typicus always united, for ex-
ample, in a collection from the Sandia Mountains, New Mexico
(Ellis 107), they are but incompletely united.
Philadelphus minutus was separated from P. occidentalis in
52 MADRONO [Vol. 7
Rydberg’s key on the basis that the styles were partially free in
the former; thus it can be seen that Rydberg and Nelson both
believed their species to be unique and separable from P. occi-
dentalis for the same reason. I hesitate, therefore, to reduce P.
minutus to synonymy but can find no true criterion by which it
can be maintained as distinct in any way from P. occidentalis. The
calyx-tube, according to Rydberg, is 2 mm. long both in P.
occidentalis and P. minutus, whereas it is “4-5 mm. long” in P.
microphyllus. I do not believe that there is any such difference.
Surely Rydberg could not have taken his measurements for the
two species from flowers at the same stage of development as it
is possible to find flowers of nearly equal size on plants which
Rydberg would have called P. microphyllus, P. minutus, and P.
occidentalis, respectively.
According to Nelson, P. occidentalis has styles “‘free for from
one-third to one-half their length.” Rydberg said of it “styles
usually wholly and the oblong or clavate stigmas partly united.”
Both conditions prevail and can be seen in the material here cited.
Material seen. Cotorapo, S. Colorado, Palmer (US) and
Popenoe in 1876 (US). Garfield County: Glenwood Springs,
Palmer 38115 (NY, US). Montrose County: Cimarron, Jones in
1890 (P) and in 1925 (P); Black Canyon of the Gunnison, Baker
266, type collection of P. minutus (C, G, NY, P, US); Newcastle,
Cary 153 (US). Fremont County: Canyon City, Brandegee in
1877 (C) and Osterhout 2092, very close to subsp. typicus and
possibly really that form (P); Cotopaxi, Johnston & Hedgecock
740 (NY); Royal Gorge, Clokey 3791 (C, G, NY, P, S, US, WSC)
close to subsp. typicus but calyx somewhat more pubescent; Oak
Creek Canyon, Rollins 1240 (G, NY). Wyomine. Sweetwater
County: near Rock Springs, Smith 6919 and Nelson 3595 (G, NY,
US). This last collection is from the exact type locality and is
the same number as the type but that collection was asserted to
have been made by R. A. Smith rather than by Nelson. UrTan.
Daggett County: 12 miles south of Manila, Hitchcock et al 3912a
(C, P,S, WSC). Carbon County: Cottonwood Canyon, Graham
9520 (G). Beaver County: Beaver City, Palmer 151 (G, NY);
S. Utah, Palmer 151 (US). Grand County: Thompson's Springs,
Jones in 1913 (G, P, US, WSC); Grand Canyon, Graham 9916
(G).
The following collections are also being referred to subsp.
occidentalis, although they intergrade, in greater or lesser degree,
with subsp. typicus. Utau. Grand County: Moab, Jones in 1913
(C, G, NY, S, US) and in 1891 (P), these plants are peculiar in
that they have large thick leaves that are semi-glabrous on the
upper surfaces; near Wilson Mesa, Rydberg & Garrett 5380 (NY).
Summit County: Brush Creek Canyon, Uintah Mts., Goodding
1274 (G, NY, US); Brush Creek Gorge, Graham 10015 (US).
Uintah County: Ashley Creek, near mouth of Dry Fork, Graham
1943] HITCHCOCK: PHILADELPHUS 53
6270 (US) ; 10 miles northwest of Vernal, Graham 7454 (G). San
Juan County: Abajo Mts., Rydberg & Garrett 9608 and 9609 (NY,
US); La Sal Mts., Purpus 6611 (C, P, US), near Clarke Lake,
Maguire et al 5803 (C), road to Warner Ranger Station, Maguire
et al 5802 -(C, US); locality uncertain, ‘“‘marvine laccolite”’ only
data on label, hence probably also from the La Sal Mts., Jones
56680 (C, NY, P, US). Arizona. North end Carrizo Mts.,
Standley 7323 (US, with some doubt, as flowers lacking). Apache
County: Luka-Chukai Mts., Goodman & Payson 2543 (NY, Wash).
In the southwestern part of Utah and in northwestern Arizona
subsp. occidentalis merges with subsp. stramineus. The following
collections are intermediate between those two phases of the
species. Most of them have the pubescence characteristic of
subsp. stramineus but the stamen complement is more suggestive
of subsp. occidentalis, in some few cases there are as few as
thirty-two stamens but most flowers have about forty, fifty-two
being the largest number seen.
Urau. Sevier County: Belknap, Stokes in 1900 (NY, S) and
Jones 63808 (NY, P,S. US); Burrville Canyon, Jones 5633 (C, NY,
P, US). Piute County: Marysvale, Jones 5875p and 740547 (P,
US); Bullion Creek, near Marysvale, Jones 5904d (NY, P).
Arizona. Coconino County: Cape Royal, north rim Grand Can-
yon, Peirson in 1927 (P); Grand View Trail to bottom of Grand
Canyon, Ferris §& Duncan 2257 (S); Grand Canyon, Toumey (S),
Toumey 1383 (US), Hitchcock 91 (US), and Knowlton 258 (US) ;
Navaho Reservation, Vorhies 109 (C, G, NY).
2. PuirapELPHus Mearnsi Evans ex Rydb. N. Am. Fl. 22: 174.
1905.
A low rounded shrub probably not much over 1 m. tall, the
branches rigid, more or less spinescent, bark of young stems
brownish, strigose, quickly exfoliating, the old branches dark
gray; leaves elliptic or lanceolate, 6-25 mm. long, pubescence of
very coarse hairs almost equally grayish-strigose on both sur-
faces, petioles 1-3 mm. long; flowers mostly single at the ends of
very short leafy stems, the hypanthium grayish-strigose, 1.5—2.5
mm. long in flower, 3 mm. long in fruit, sepals ca. 3 mm. long, acu-
minate, strigose on outer surface, lanate within; petals whitish or
ochroleucous, oblong-lanceolate, 7-11 mm. long, ovate-lanceolate
to lanceolate, acute or sharply 2-toothed at apex; stamens 16-20
(24), the filaments short, free; styles less than 1 mm. long, united
to top, stigmas 1.5—2 mm. long, almost completely united.
Key To Supspecies oF P. MEARNSII
Petals 8-10 mm. long; leaves mostly less than 15 mm.
long, strigose on both surfaces with appressed pubes-
cence; petioles 1-2 mm. long’ ..... 0.5... ..5.54045 _ 2a. P. Mearnsit
subsp. typicus
Petals 10-11 mm. long; leaves mostly over 15 mm. long,
the upper surfaces with erect hairs; petioles 2-3 mm.
NAYES cy & By ORS Ea RGR a 2b. P. Mearnsii
subsp. bifidus
54 MADRONO [Vol. 7
2a. PuHinrapELPHUs Merarnsi subsp. typicus nom. nov. P.
Mearnsu Evans ex Rydb., l.c.
Characters as in key, the leaves rather thick, the pubescence
of very coarse short hairs.
Type collection. Upper Corner Monument, Grant County,
New Mexico, April 28, 1892, Mearns 86 (G, NY,S, US).
Other material seen. New Mexico. Eddy County: Guada-
lupe Mts., near Three Forks of Rocky Arroyo, May 6, 1982,
Wilkens 1833 (US) ; Carlsbad Cave, Bailey in 1924 (US). Texas.
Culberson County: Guadalupe Mts., above McKittrick Canyon,
July 17, 19381, Moore & Steyermark 3477. (C, G).
Because of the uniform strigose pubescence, short styles and
few stamens, this entity is readily distinguished from the other
species and even though it is known only through three collec-
tions, there can be no doubt of its validity. The Moore and
Steyermark collection is in fruit and therefore lacks flowers, but
the coarse hairs and short styles make it seem certain that it be-
longs here.
Although the species was described as having “‘about 15” sta-
mens, those flowers of the type collection which I have examined
have sixteen. The Wilkens collection also has sixteen stamens,,
a fact which helps to convince one that it is conspecific with the
Mearns collection. However, Bailey’s collection from Carlsbad
has flowers with twenty-four stamens. A comparison of the
Bailey and Wilkens collections cannot but convince one that they
are not only conspecific but that they are too similar to be sepa-
rated nomenclaturally, hence the species must be considered to
have as few as sixteen (occasionally fifteen?) and as many as
twenty-four stamens.
2b. PHirapeLPHUs Merarnsii subsp. bifidus subsp. nov.
Planta P. Mearnsi subsp. typicus similis, hoc modo differens:
foliis longioribus gracilioribusque, laminis ad 80 mm. longis,
supra hirsuto-hispidis, pilis erectis; petiolis 2-3 mm. longis,
petalis 10-11 mm. longis, apices 2-dentibus; staminibus 24, fila-
mentis non connatis.
Leaves longer and more slender than in subsp. typicus, the
blades as much as 30 mm. long, pubescence of their upper sur-
faces hirsute-hispid, the hairs slender, scarcely at all appressed,
petioles 2-8 mm. long; petals 10-11 mm. long, the apices dis-
tinctly sharply 2-toothed; stamens 24 (probably some variation),
the filaments free.
Type. Sierra Madre, near Monterrey, Nuevo Leon, Mexico,
May 2, 1906, Pringle 13879 (G).
Known only from the type collections at Gray Herbarium and
United States National Museum, one branch of the latter speci-
men with partially double flowers. Superficially this collection
seems to bear little resemblance to the type of P. Mearnsu. How-
ever, it is evident that that collection was an extremely small-
1943 | HITCHCOCK: PHILADELPHUS 55
flowered plant and that probably the other collections cited
(Wilkens 1833) are more truly representative of the entity. Since
the plant of the Wilkens collection is larger-leaved and larger-
flowered, it is more suggestive of Pringle’s plant from near
Monterrey. Because there is so much similarity between the
flowers of the two entities, the plant from Mexico is being ac-
corded subspecific rank. Toothed, oblong petals, few stamens,
short styles, and united stigmas comprise an unusual combination
of characters and their presence in both these entities makes it
seem logical to assume that there is close relationship between
the two, a relationship that is too close to be brought out by ac-
cording both of them specific status,
3. PHILADELPHUS SERPYLLIFoLIUS Gray, Pl. Wright. 1: 77.
1850.
A low rounded, rather rigid shrub 1—2 m. tall, young branches
brownish, densely strigose-pubescent, bark‘of older stems exfoli-
ating, the old stems grayish; leaves entire, ovate-lanceolate, 6—30
(averaging about 15) mm. long, 3-12 mm. broad, 3-nerved from
base, the upper surfaces greenish and often shining, from
sparsely strigose to rather densely strigose and short hirsute,
lower surfaces grayish, the pubescence rather dense, of long
straight appressed hairs and close covering of tangled tomentum,
distinctly petiolate, the petioles 1-3 mm. long; flowers commonly
1 or 8 (or 2). at the ends of short leafy shoots, the pedicels 1 or 2
mm. long; calyx usually silvery with long appressed and shorter
more or less curled hairs, but sometimes the indumentum more
sparse and calyx greenish, the tube 2-3 mm. long in flower, 3—4
mm. long in fruit, the lobes lanceolate, acuminate, 4-5 mm. long,
densely lanate on inner surfaces; petals apparently cream colored,
6-8 mm. long, more or less oval and usually somewhat emarginate
or erose; stamens usually 28 (few more or less), the filaments
free; styles less than 1 mm. in length, united, the stigmas 1.5—2
mm. long, almost completely united.
Type. “Between western Texas and El Paso, New Mexico,”
in 1851, C. Wright. Very closely matched by a specimen col-
lected near Fort Davis, Texas, Palmer 34474.
Material seen. New Mexico. Without definite locality, 1851-—
52, C. Wright 1100 (G, NY); Mexican Boundary Survey (NY),
probably the same as the first.. Texas. Head of Seco, Reverchon
54 (G), one branch of this collection has leaves only 12 mm. long
whereas a younger more vigorous shoot has leaves that are 30
mm. long. Jeff Davis County: deep canyon off Limpia Canyon,
near Fort Davis. Palmer 34474 (NY, US). Edwards County:
limestone bluffs, Upper Cedar Creek, Palmer 12333 (C). Kendall
County: Spring Creek, near Boerne, Palmer 11492 (C) and Palmer
11595, 12900 (C, P) ; Cibola Canyon, Boerne, Pennell 10418 (NY),
pubescence of leaves strigose above; Musquez Canyon, W. Texas,
Havard 41 (G). Culberson County: McKittrick Canyon, Guada-
lupe Mts., Moore & Steyermark 8479 (C, G, NY,S).
56 MADRONO [Vol. 7
Philadelphus serpyllifolius is, of course, closely related to P.
microphyllus but differs strikingly because of the tomentum of
the leaves and the extremely short styles. The collections from
Kendall County are much less pubescent than is the rest of the
material.
4. PuHILaADELPHUs PurRPusu Brandegee, Univ. Calif. Publ. Bot.
A 200. ~ VONZ.
A low spreading shrub 1-2 m. tall; young branches brownish,
densely strigose-pubescent, older bark grayish, winter buds quite
obviously not enclosed in the leaf bases, petioles 1-4 mm. long,
blades entire, ovate-lanceolate to lanceolate, usually slightly mu-
cronate, 20-35 mm. long, 3-nerved, pubescence of both surfaces
grayish-green, almost equally strigose with short thick hairs;
flowers single at the ends of short branches, the pedicels 3-6 mm.
long; calyces grayish-strigose, the tube 8—4 mm. long, the lobes
4-6 mm. long, acute; petals obovate to oval, 10-15 mm. long,
rounded, scarcely emarginate; stamens 40—50, filaments distinct ;
styles 2-3 mm. long, united about one-half to four-fifths their
length, basal portion and surrounding ovary grayish pilose;
stigmas free, 2-3 mm. long.
Type. Minas de San Rafael, San Luis Potosi, Mexico, Purpus
5368.
Known only from two collections from the type locality, Pur-
pus 5868 (C, G, NY) and Purpus 4910 (G, US).
Philadelphus Purpusii is most easily distinguished because of
the pilosity of the styles and upper ovary, but the grayish-green
strigose leaves, large petals, and exposed buds are all features
that help to make it the distinctive species that it is.
University of Washington, Seattle,
March, 1942.
A NEW SPECIES OF PHACELIA FROM SALINE VALLEY,
CALIFORNIA
LINCOLN CONSTANCE
Phacelia amabilis sp. nov. Herba annua vel biennis, omnino
glanduloso-puberula et hispida praecipue in calycibus inflores-
centibusque, circa 1 m. alta; caulis crassus ramosus; folia peti-
olata oblonga oblongo-ovatave, 8-15 cm. longa, 3-5 cm. lata,
pinnatifida, lobis oblongis dentatis, summa reducta minus alte
divisa; inflorescentia corymbosa, cymis 5-12 cm. longis in fructu
erectis; pedicelli in fructu 2-3 mm. longi; calycis lobae anguste
lanceolatae, 3-5 mm. longae, 1-2 mm. latae, corolla plus quam
dimidio breviores, capsulam leviter excedentes; corolla late cam-
panulata, 7-8 mm. longa, 8-12 mm. lata, alba, lobis integris;
appendiculae supra tubae basin minus quam 1 mm. insertae, parte
libera lata; stamina exserta; stylus exsertus, pallido-lilacinus ;
1943] KEARNEY AND PEEBLES: GILIA MULTIFLORA 57
capsula ovoidea, 3-4 mm. lata; semina plerumque 2 vel 4, 3—4
mm. longa, tenuissima pallidissima, non corrugata, superfacie
ventrali jugo saliente utrinque excavata.
Annual or biennial, about 1 m. high; stem stout, branching ;
herbage glandular-puberulent throughout and hispid, especially
on the calyces and in the inflorescence; leaves petiolate, oblong
to oblong-ovate, 8-15 cm. long, 8-5 cm. broad, pinnatifid, the
lobes oblong, dentate, the uppermost leaves reduced and less
deeply divided; inflorescence corymbose, the cymes 5-12 cm.
long and erect in fruit; pedicels 2-3 mm. long in fruit; calyx
lobes narrow lanceolate, 3-5 mm. long, 1-2 mm. broad, less than
one-half as long as the corolla, slightly exceeding the capsule;
corolla broadly campanulate, 7-8 mm. long, 8-12 mm. broad,
white, the lobes entire; appendages attached a little less than 1
mm. above the base of the tube, the free portion broad; stamens
and style exserted 5 mm. or more, the latter pale lavender; cap-
sule ovoid, 3-4 mm. long, 2-3 mm. broad; seeds usually 2 or 4,
3-4 mm. long, very thin and pale, not corrugated, the ventral sur-
face excavated on each side of a salient ridge.
Type. “In full bloom along creek, Hunter Creek, altitude
1800 feet, Saline Valley, Inyo County, California,’ April 21,
1942, Annie M. Alexander and Louise Kellogg 2681 (Herbarium of
the University of California no. 671871). Phacelia amabilis was
first thought to be an albino form of P. crenulata with which it was
growing.
In his excellent, “Revision of the Phacelia crenulata group for
North America” (Bull. Torrey Bot. Club 64: 81-96, 1383-144.
1937), Voss has provided the most complete modern treatment
yet available for any portion of this interesting and “difficult”
genus. Although interspecific differences are small in this group,
most of the entities accorded specific rank by him appear to be
rather sharply distinguished from one another. The present spe-
cies, by possession of a broadly campanulate corolla, exserted
stamens and ventrally excavated but uncorrugated seeds would
appear to be most closely related to P. congesta Hook., of Texas,
New Mexico, southeastern Arizona and northern Mexico. Be-
sides the wide discontinuity in geographical range, P. amabilis
differs from P. congesta—as indicated on the accompanying plate
—in its shorter calyx, larger and differently proportioned corolla,
more exserted stamens and style, and its larger, paler, thinner
and more broadly margined seeds.
For the past several years, the Misses Alexander and
Kellogg have collected extensively in the relatively inaccessible
desert mountains of the southwestern United States. The Her-
barium of the University of California has been enriched by
receiving their nearly 3500 numbers, as well as many valuable
duplicates, which have been distributed to other leading herbaria.
58 MADRONO-~. [Vol. 7
Priate 3. Puacetia. Figs. a—-e, P. amabilis: a, habit; b, flower; c, fruiting
calyx and capsule; d, expanded corolla; e, seeds. Figs. f—j, P. congesta var.
rupestris: f, habit; g, flower; h, fruiting calyx and capsule; i, expanded corolla;
j, seeds. (Habit drawings X 1%; flowers and capsules, x3; seeds, X 7.)
1943] KEARNEY AND PEEBLES: GILIA MULTIFLORA 59
Phacelia amabilis is but one of the rare or seldom-collected species
which has been secured as a result of their indefatigable efforts.
The writer would ordinarily have preferred to delay publi-
cation of this species until more material is available, but in view
of the curtailment of field work in the foreseeable future, it
seemed advisable to describe it at this time.
Department of Botany,
University of California, Berkeley,
November, 1942.
GILIA MULTIFLORA NUTT. AND ITS
NEAREST RELATIVES
Tuomas H. Kearney AND Rosert H. PEEBLES
Gilia multiflora Nutt., a widely distributed species of New
Mexico and Arizona, is extremely variable, but is distinguished
from its nearest relatives by having the corolla tube nearly always
two to three times the length of the lobes, and rarely less than one
and a half times the length of the calyx. Very similar to G. multi-
flora in habit, foliage, and pubescence is G. polyantha Rydb. which
differs, however, in having a shorter corolla tube, this approxi-
mately equal in length to the lobes and the calyx.
Typical G. polyantha is known apparently only from south-
western Colorado. Gilia brachysiphon Woot. and Standl., of south-
western New Mexico, would seem to be specifically distinct from
G. polyantha were it not for the occurrence in north-central Ari-
zona of a form that is intermediate in several characters. For
this reason, it seems best to treat G. brachysiphon and the hitherto
undescribed Arizona plant as varieties of G. polyantha. So far as
present information goes, the three forms of this species are
rather widely separated geographically.
Another more distantly related member of this small group of
perennial plants with filaments normally conspicuously exserted
and declined is G. Havardi A. Gray, an apparently rare species of
southwestern Texas. The corolla is more pronouncedly zygo-
morphic, especially in respect to the closely grouped and parallel-
declined stamens, than in G. multiflora and G. polyantha, and for
this reason Brand (in Engler, Pflanzenr. 47°°: 172. 1907) re-
stored this species to the genus Loeselia, where it was placed
originally by Asa Gray. Gilia multiflora and G. polyantha, how-
ever, also show a tendency to zygomorphy and the writers concur
in Gray’s final conclusion that G. Harvardi and G. multiflora are
congeners.
Gilia Macombii Torr., although evidently related to G. multi-
flora, is not considered here because the stamens are not exserted
from the corolla tube, or project only about the length of the
anthers.
60 MADRONO [Vol. 7
KEY TO THE SPECIES AND VARIETIES.
Inflorescences short, not thyrsoid, the zygomorphic
flowers in loose, few-flowered terminal clusters;
herbage eglandular, villous, even in the _ in-
florescence, with long soft white hairs; stems nu-
merous from a branched caudex, short; corolla
tube 6 to 9 mm. long, about twice as long as the
calyx, not or only slightly longer than the lobes.. 1. G. Havardi
Inflorescences elongate, thyrsoid-glomerate, the flow-
ers nearly regular; herbage usually more or less
glandular in the inflorescence, the long white hairs
mostly confined to the lower part of the plant.
Corolla tube 10 to 15 (rarely only 7) mm. long, 1.6
to 2.7 (rarely only 1.2) times as long as the
calyx and 1.5 to 3.2 times as long as the corolla
lobes, the lobes prevailingly oblong or nar-
rowly elliptic but occasionally slightly obovate;
inflorescences normally contracted and_ short-
branched; pubescence various ................ 2. G. multiflora
Corolla tube 4.5 to 6.5 mm. long, shorter than to 1.2
(exceptionally 1.7) times as long as the calyx
and shorter than to 1.5 times as long as the
corolla lobes.
Lobes of the whitish, sometimes purple-dotted,
corolla 3 mm. wide, oval or broadly elliptic;
inflorescences relatively open and_long-
branched, copiously puberulent with glandular
hairs, these intermixed with short white hairs. 3. G. polyantha
(typical)
Lobes of the pale violet corolla 1.5 to 2.7 (excep-
tionally 3.5) mm. wide.
Inflorescences relatively contracted and short-
branched, the glomerules dense; upper part
of the plant canescent with short white
hairs, the glandular hairs relatively few and
mostly hidden by the others; corolla lobes
prevailingly elliptic or oblong ............ 3a. G. polyantha
var. brachysiphon
Inflorescences relatively expanded and long-
branched, the glomerules relatively loose;
upper part of the plant commonly with
abundant glandular hairs, these intermixed
with but usually not concealed by the short
white hairs; corolla lobes _ prevailingly
oblanceolate or obovate <=) se oe 3b. G. polyantha
var. Whitingi
1. Grrr Havarpt A. Gray, Syn. Fl. N. Amered. 2, 2°. 4
1886. Loeselia Havardi A. Gray, Proc. Amer. Acad. Arts and
Sci. 19: 87. 1883.
Presidio and Brewster counties, southwestern Texas, near
Presidio (Havard in 1881, the type collection), vicinity of the
Chinati Mountains (Nealley in 1889), vicinity of the Chisos Moun-
tains (Sperry 623). The stems are less than 20 cm. long. The
species is very distinct in appearance from other members of this
group. The herbage is uniformly villous with long, soft, white,
segmented hairs, these extending even to the calyx, whereas in
1943 | KEARNEY AND PEEBLES: GILIA MULTIFLORA 61
G. multiflora and G. polyantha such hairs are confined largely to the
lower leaves. The leaves are mostly pinnatifid with 3 to 5 lobes.
The ovules are “several in each cell” (Gray, ibid.).
2. GILIA MULTIFLORA Nutt., Jour. Acad. Nat. Sci. Phila. ser. 2,
i 154. 1848.
This species seems to be known only from the mountains of
New Mexico and Arizona, where it is common and widely dis-
tributed, but it probably occurs also in northern Mexico. The
type, collected by Gambel on “‘sandy hills along the border of the
Rio del Norte, New Mexico,” has not been seen by the writers.
There is no type material in the herbarium of the Philadelphia
Academy of Sciences or in the Gray Herbarium. This is a highly
variable species, but departures from the norm in floral charac-
ters do not appear, as a rule, to be correlated among themselves
or with peculiar vegetative characters. The plants are poten-
tially long-lived and some of the specimens from southern Arizona
are distinctly woody toward base. The pubescence consists of
two main types of hairs: (1) soft, white, flattened, segmented
hairs, these often elongate near the base of the plant, especially
on the leaves, but normally becoming very short in the inflores-
cence, where they are usually intermixed with (2) clavate or
_ stipitate glandular hairs. Occasional specimens show granular
puberulence, apparently of a resinous nature. The leaves vary
from entire to pinnatifid with 3 to 7 narrow lobes.
The inflorescences are normally contracted and_ short-
branched, but occasional specimens resemble G. polyantha Rydb.
in their relatively open and long-branched inflorescences. The
insertion of the filaments is normally very nearly at the same
level, usually just below the rim of the throat, but sometimes deep
in the tube. The anthers are at nearly the same level, or some
of them are 1 to 2 mm. below the others. Counts in eight ovaries,
each from a different individual plant, showed a range from 5 to
15 ovules, indicating a variation of from 2 to 5 per cell.
Two outstanding variants, both found in Arizona, may prove
worthy of recognition as varieties when more material is avail-
able. The first, collected near Flagstaff (Rusby 729), and on
“Ivy Mesa, Mogollon Mountains’! (MacDougal 644), is character-
ized by numerous, slender, almost flexuous stems; inflorescences
with subfarinose puberulence, obscurely if at all glandular or
granular; leaves nearly all entire except near the base of the
plant, very sparsely villous; corolla tube only 7 to 8.5 mm. long
and only 1.6 to 2.0 times as long as the lobes, whereas in G. multi-
flora the tube is usually at least 10 mm. long and more than twice
as long as the lobes.
The second remarkable variant was collected in Oak Creek
Canyon, Coconino County (Whiting 1053/5300). It is distin-
1'The “Mogollon Mountains” of Arizona are the precipitous southern edge
of the Mogollon Mesa, also often referred to as the “Tonto Rim” or “Mogollon
Rim.”
62 | MADRONO [Vol. 7
guished from all forms of G. polyantha by having a corolla with
tube 2.5 times as long as the lobes, but differs markedly from
typical G. multiflora in the following characters: calyx 8.5 mm.
long and nearly equalling the tube of the corolla, whereas in G.
multiflora the calyx is normally only 5 to 7 mm. long and seldom
more than three-fifths as long as the tube; corolla lobes broadly
obovate, three-fourths as wide as long, whereas in G. multiflora
they are commonly oblong or only slightly obovate and not more
than half as wide as long; corolla throat exceptionally ample, 3.5
mm. wide at the orifice in the pressed specimen, as compared with
a usual width of 2 to 8 mm. in G. multiflora. A collection in the
Baboquivari Mountains, Arizona, (Gilman B124) has an equally
wide corolla throat but is otherwise normal in its flower charac-
ters, and the plant is decidedly woody toward base, which is not
the case in Whiting’s plant.
3. Git1a PoLYANTHA Rydb., Bull. Torr. Bot. Club 31: 6384.
1904. Gilia exrserta A, Nelson, Bot. Gaz. 40: 65. 1905. Giulia
multiflora var. polyantha Brand, in Engler, Pflanzenr. 4°°°: 1138.
LOO7.
The type of G. polyantha, which is also the type of G. easerta,
was collected at Pagosa Springs, Archuleta County, southern
Colorado, altitude 7,000 feet (C. F. Baker 538) and the typical
form of the species seems to be known only from this locality,
where it has been collected also by Bethel, Willey, and Clokey
(4251). Itis taller and has more elongate, longer-branched inflo-
rescences than most specimens of G. multiflora and is readily dis-
tinguished from that species by the flower characters stated in
the key. On the other hand, it resembles G. multiflora in habit,
foliage, and pubescence. The leaves sometimes have as many as
9 lobes. Counts of the number of ovules per ovary gave 13 for
the type specimen and 10 for the other. A. Nelson (ibid.) re-
ported the ovules as only “‘about 2 in each cell,” in the type col-
lection. The plant was stated by Nelson to be biennial, but it is
almost certainly perennial, as Rydberg described it.
3a. Gitia PotyantHa Rydb. var. brachysiphon (Woot. and
Standl.) comb. nov. Gilia brachysiphon Woot. and Standl., Con-
trib. U.S. Nat. Herb. 16: 160. 1913.
The type of Gilia brachysiphon was collected in the Organ
Mountains, Dona Ana County, New Mexico (Wooton in 1894) and
this variety is known also from several other localities, all in
southwestern New Mexico (Sierra, Grant, and Catron counties).
It differs from typical G. polyantha, and from most specimens of
the following variety, in the rarity or entire absence of glandular
hairs. Counts showed 18 ovules per ovary in the type and 14 in
another specimen.
3b. Giza PotyantTHa Rydb. var. Whitingi var. nov.
A forma typica G. polyanthae corolla violacea lobis plerumque
obovatis differt; a G. polyantha var. brachysiphon inflorescentia
laxiore et plerumque copiose glandulifera differt; a G. multiflora
1943] REVIEW 63
corollae tubo quam calyce et corollae lobis vix longiore distin-
guitur.
The type was collected at the Grand Canyon, Coconino
County, Arizona (Whiting 1072/5200, U. S. Nat. Herb. 1814983).
Other collections, all in Coconino County, at elevations of 6,800
to 7,200 feet are: Grand Canyon (Kggleston 15669, 15682), south
rim of the Grand Canyon (Collom 1073), Grand Canyon Road
(Whiting 1047/4311), near the Grand Canyon (Lemmon in 1884),?
Williams (Rusby in 18838), base of the San Francisco Peaks
(Wooton 489), Walnut Canyon National Monument (Beaubien
1054/5320). There is also, in the United States National Her-
barium, an imperfect specimen, labeled in G. R. Vasey’s hand-
writing, ““Nevada, Lt. Wheeler, 1872,” that apparently belongs
here. This specimen may have been the basis for the inclusion
of Nevada in the range of G. multiflora, in Tidestrom’s Flora of
Utah and Nevada (Contrib. U: S. Nat. Herb. 25: 485. 1925).
If it is correctly labeled as to locality, var. Whitingi has a wider
range than the other forms of G. polyantha.
There is a much stronger tendency than in G. multiflora, and
in the other forms of G. polyantha, to have the corolla lobes widest
near the apex, but this is not invariably the case. The stamens,
as is usually also the case in G. multiflora, are inserted very nearly
at the same level. Counts made on six individual plants, includ-
ing the type, showed that the number of ovules per ovary varies
from 6 to 18. Bureau of Plant Industry,
United States Department of Agriculture,
Washington, D. C.
United States Field Station
Sacaton, Arizona.
June 27, 1942
REVIEW
Wild Violets of North America. By Vioua Bratnerp Bairp. Pp.
xv + 225, with 17 illustrations in the text and 80 plates in full
color. University of California Press, Berkeley. 1942. $10.00.
This attractive volume, written in non-technical language, will
please both professional botanists and amateurs. The prepara-
tion of a complete account of North American violets is a task for
which the author is especially well equipped. Familiar with the
group since childhood, Mrs. Baird assisted her father, Ezra
Brainerd, in his well-known studies on the genus and during later
years she has continuously added to her knowledge by field and
garden studies of almost every species.
The brief preamble contains an informative résumé of the
distribution and probable sources of the North American species
of Viola, a discussion of the variation in such characters as leaf
2Cited by Brand (Pflanzenr.) under G. multiflora var. polyantha. The
collection at Cosnino, Ariz. (Jones 4043) also cited by Brand under var.
polyantha, is G. multiflora.
64 MADRONO [Vol. 7
shape and flower color, and a simple explanation of the flower
structure and function. There follows a key to the species groups
which is both technically accurate and easily understood and
utilized by the layman. Especially valuable as a scientific contri-
bution are the two pages of line drawings illustrating the vari-
ation in the shape of the pistil, the remarkable and almost ludi-
crous development of which will interest and perhaps even amuse
every reader. Since all were drawn from fresh material, persis-
tent misconceptions as to pistil shape in certain species are here
for the first time corrected.
The classification divides the group into three sections, each
of which is introduced by a brief discussion of the characters
common to its members. The account of each species is very
readable and contains information as to the outstanding charac-
ters, distribution, habitat and flowering period of each, supple-
mented by notes on any features of unusual interest such as the
derivation of the name, by whom and where the species was first
collected, and mention of the characters by which it may be dis-
tinguished from its nearest relatives. As is fitting in a treatment
which is both popular and technical, Mrs. Baird has not hesitated
to group under their respective linneons minor entities which have
been named as species. Thus the specific units as she conceives
them are readily recognizable even by those without previous
experience with taxonomic concepts.
The plates, reproduced from water colors by the late F.
Schuyler Matthews, are all natural size and indicate with an un-
usual degree of accuracy the different shades in the flower color
throughout the North American representatives of the genus.
The color of practically every species was matched with great
care with living material from native colonies or garden trans-
plants and has been faithfully reproduced.
Following the general text, in a “Key to Colors in Wild
Violets,’ the species are grouped according to their principal
color tones and these groups are still further subdivided as to the
precise shade of flower color. Next, a list of the species is in-
cluded with citations to the original descriptions and data as to
the sources of the material used as a basis for the color plates.
The volume closes with indices to the scientific and popular names.
The book is handsomely bound in violet blue cloth with gold
lettering and expertly printed in twelve point Baskerville on a
highly lustrous coated paper. The edition is limited to one thou-
sand signed copies.—ETHEL Crum.
Nee
MADRONO
A West American Journal of Botany
A quarterly journal devoted to important
and stimulating articles dealing with
plant morphology, physiology, taxonomy,
and botanical history. These volumes
should be a part of every botanist’s li-
brary and should be made accessible to
students of all universities and colleges.
Volume I, 1916-1929. . . $5.00
Volume II, 1930-1934 .. 5.00
Volume III, 1935-1936 . 5.00
Volume IV, 1937-1938 . 5.00
Volume V, 1939-1940 . . 5.00
Volume VI, 1941-1942 . 5.00
Single numbers....... 0.75
The subscription price of MADRONO
is $2.50 per year. We solicit your pat-
ronage,
Address all orders to:
William Hiesey Bus. Mgr.
Carnegie Institution of Washington
Stanford University, California
NUMBER 3
MADRONO
A WEST AMERICAN JOURNAL OF
BOTANY
orewortHY Puants From IpaHo. II, Arthur Cronquist
Leerrmacy or THaticrraum potycarrum Wats. Arthur Cronquist ..
ann Learrimate NAMES—AND THALICTRUM PpoLycaRPUM S. Warts.,
C. A. Weatherby Dkk io
Published at North Queen Street and McGovern Avenue
Lancaster, Pennsylvania
July, 1943
MADRONO
A WEST AMERICAN JOURNAL OF BOTANY
Board of Editors
Herzert L. Mason, University of California, Berkeley, Chairman.
LeRoy Asrams, Stanford University, California.
Epecar Anperson, Missouri Botanical Garden, St. Louis.
Lyman Benson, University of Arizona, Tucson.
Hersert F. Copetanp, Sacramento Junior College, Sacramento, California.
Ivan M. Jonnston, Arnold Arboretum, Jamaica Plain, Massachusetts.
Miunprep E. Maruias, University of California, Berkeley.
Bassetr Macume, Utah State Agricultural College, Logan.
Marion Ownsey, State College of Washington, Pullman.
Secretary, Editorial Board—ANNETTA CARTER
Department of Botany, University of California, Berkeley
Business Manager—Wittiam Hussey
North Queen Street and McGovern Avenue, Lancaster, Pennsylvania
or
Carnegie Institution of Washington
Stanford University, California
Entered as second-class matter October 1, 1935, at the post office at
Lancaster, Pa., under the act of March 3, 1879.
Established 1916. Published quarterly. Subscription Price $2.50 per year.
Completed volumes I to V inclusive, $25.00; each volume $5.00; single numbers
$0.75.
Papers up to 15 or 20 pages are acceptable. Longer contributions may
be accepted if the excess costs of printing and illustration are borne by the
contributor. Range extensions and similar notes will be published in con-
densed form with a suitable title under the general heading “Notes and News.”
Articles may be submitted to any member of the editorial board. Manuscripts
may be included in the forthcoming issue provided that the contributor pay the
cost of the pages added to the issue to accommodate his article. Reprints of
any article are furnished at a cost of 4 pages, 50 copies $4.10; 100 copies $4.50;
additional 100’s $0.85; 8 pages, 50 copies $5.95; 100 copies $6.60; additional 100’s
$1.30; 16 pages, 50 copies $8.35; 100 copies $9.35; additional 100’s $2.00. Covers,
50 for $2.75; additional covers at $1.65 per hundred. Reprints should be
ordered when proofs are returned.
Published at North Queen Street and McGovern Avenue, Lancaster,
Pennsylvania, for the
CALIFORNIA BOTANICAL SOCIETY, INC.
President: H. S. Reed, University of California, Berkeley. First Vice-
President: N. T. Mirov, California Forest and Range Experiment Station, Uni-
versity of California, Berkeley. Second Vice-President: Jack Whitehead, Uni-
versity of California Botanic Garden, Berkeley. Secretary: Clarence R. Quick,
United States Department of Agriculture, 26 Giannini Hall, University of Cali-
fornia, Berkeley. Treasurer: William Hiesey, Carnegie Institution of Wash-
ington, Stanford University, California.
Annual membership dues of the California Botanical Society are $2.50,
$2.00 of which is for a year’s subscription to Madrofio. Dues should be
remitted to the Treasurer. General correspondence and applications for
membership should be addressed to the Secretary.
1943] MATHIAS AND CONSTANCE: TAUSCHIA 65
A NEW SPECIES OF TAUSCHIA FROM THE STATE OF
WASHINGTON
Mivprep EF. Maruias anp LINcoLN CONSTANCE
Tauschia Hooveri sp. nov. Herba acaulescens, pseudoscapo
e tubere globose, 10-14 cm. alta, omnino glabra glauca; folia in
ambitu ovato-triangularia, petiolo excluso, 3-5 cm. longa, 1—6 cm.
lata, pinnata vel partim bipinnata; foliola linearia, acuminata,
15-35 mm. longa, 1-2 mm. lata, integra, calloso-apiculata, mar-
ginibus reflexis; petioli 2-3 cm. longi; pedunculi graciles, 2—4
cm. longi; calycis dentes obsoleti; corollae albae; antherae pur-
pureae; styli breves, filiformes, recurvati; carpophorum ad
medium bifidum, ramis ad apicem approximatis; fructus lineari-
oblongus ad apicem versus leviter attenuatus, 5-7 mm. longus,
circa 2 mm. latus, glaucus, costis evidentibus filiformibus; vittae
parvae solitariae in valleculis, 2 in commissura; semini facies fere
plana.
Acaulescent with a pseudoscape 10-14 cm. high arising from
a globose tuber; glabrous and glaucous throughout; leaves ovate-
deltoid in general outline, excluding the petiole 3-5 cm. long,
4-6 cm. broad, pinnate to partially bipinnate, the leaflets linear,
acuminate, 1.5—3.5 cm. long, 0.1-0.2 cm. broad, entire, callous-
tipped, the margins reflexed; petioles 2-3 cm. long; peduncles
slender, 2-4 cm. long; involucre and involucel wanting; fertile
rays 3—7, stout, unequal, 2-10 mm. long; pedicels 1-3 mm. long;
calyx teeth obsolete; flowers white, the anthers purple; styles
short, filiform, recurved; carpophore 2-cleft about one-half way
to the base, the halves approximate to the tip; fruit linear-
oblong, tapering slightly at apex, 5-7 mm. long, about 2 mm.
broad, glaucous, the ribs filiform but evident; oil tubes small,
usually solitary in the intervals, 2 on the commissure; seed face
nearly plane.
Type. “Near Cowiche, Yakima County, Washington,” April
20, 1942, Robert F. Hoover 5689 (University of California Her-
barium no. 671873). There is one other collection, “Hills south
of White Swan, Yakima County,’ March 24, 1942, Hoover 5616
(Univ. Calif. Herb.). In a letter from the collector dated
November 5, 1942, the following additional information is given:
“In answer to your question about the habitat of T’auschia Hooveri,
it grows in ‘scablands, in rather barren rocky clay, with
Artemisia rigida, Sisyrinchium Douglasii, Eriogonum thymoides, Viola
trinervata, etc.”’
To the best of our knowledge, this species has never been
collected before, although the Yakima area has been rather in-
tensively botanized previously by various collectors. It un-
doubtedly escaped notice because the plants were in ripe fruit
Mapronxo, Vol. 7, pp. 65-96. July 29, 1943.
IAG
AS
AUG +
66 MADRONO [Vol. 7
d
Fic. 1. TYauschia Hooveri: a, in anthesis, x1; 6, in fruit, X1; c, side view
of fruit, x 5; d, cross section of fruit, X< 10.
before the end of April, and because in anthesis they so greatly
simulate certain of the “bulbous” species of both Lomatium and
the two species of Orogenia.
In a recent series of nomenclatorial transfers, the authors
(Bull. Torrey Bot. Club 68: 121-124. 1941) proposed to reduce
the monotypic genus Hesperogenia Coulter and Rose under
Tauschia Schlecht. This step was not taken without some mis-
giving inasmuch as Hesperogenia is known only from Mount
Rainier, Washington, and the geographically nearest undoubted
Tauschia is not known to occur north of southwestern Oregon.
In addition, the suborbicular fruit of Hesperogenia is unique in
1943] CLOKEY: CHARLESTON MOUNTAINS CACTACEAE 67
Tauschia, although its other characters fit the latter genus very
well. The discovery of YTauschia Hooveri, however, ends the
supposed geographical isolation of Hesperogenia and appears to
confirm the desirability of including it under T'auschia.
We are glad to have the opportunity to name this remarkable
species for Dr. Hoover, whose critical collections and keen ob-
servations have been invaluable to us in our work with the
Umbelliferae.
Department of Botany,
University of California, Berkeley,
November, 1942.
NOTES ON THE FLORA OF THE CHARLESTON
MOUNTAINS, CLARK COUNTY, NEVADA
V. CACTACEAE
Ira W. CLOKEY
This treatment of the Cactaceae continues a series devoted
to a study of the flora of the Charleston Mountains in southern
Nevada and published as follows: Madrofio 4: 128-130. 1937;
Bull. So. Calif. Acad. Sei. 87: 1-11. 1988; l.c. 38: 1-7. 1939;
and Madrofio 6: 211-222. 1942. I wish to express my thanks
to Dr. L. Benson, Dr. E. U. Clover, Mr. Fred Gibson and Dr. Ira
L. Wiggins for assistance in the study of the Cactaceae of the
Charleston Mountains and for specimens of and information re-
garding these and related species. Specimens of all of the spe-
cies treated below except those of Mammillaria tetrancistra Engelm.
have been widely distributed to the herbaria of the world. Types
of the new species are all in the Clokey Herbarium now at the
Bull. So. Calif. Acad. Sci. 87: 1-11. 1988; lc. 838: 1-7. 1939;
garding these and related species. Specimens of all the spe-
University of California, Berkeley.
Key To THE GENERA OF CACTACEAE
Stems jointed, cylindrical or flat; leaves small, subulate, de-
ciduous; areoles on tubercles or on flat surfaces, with
numerous, barbed glochids; glochids and flowers produced
from the same areoles; ovaries with areoles and glochids;
spines barbed or not barbed .......................... 1. Opuntia
Stems not jointed, cylindrical, without leaves; areoles on
ridges or tubercles, without glochids; flowers produced
above the areoles ; ovaries with or without areoles, with-
out glochids; spines not barbed.
Stems ribbed; spines borne on definite ridges.
Flowers borne above old spine- -bearing areoles, solitary,
appearing lateral, purple or crimson; tube and ovary
spiny; fruit colored, thin-skinned, spiny Re Mig e ee 2. Echinocereus
Flowers borne above young areoles, appearing sub-termi-
nal in a circle near the top of plant; ovary scaly;
Ube reens WwATHOUG SPINES: |, Gydie aa class dod a Wath be 3. Echinocactus
Stems not ribbed; spines borne on tubercles arranged in
rows or scattered.
68 MADRONO [Vol. 7
Flowers lateral; tubercles not grooved on upper side; one
or more of the spines hooked; fruit red; seeds black,
rugose, with a thick corky base in our species ...... 4. Mammillaria
Flowers terminal; tubercles grooved on upper side; none
of the spines hooked; fruit green or rose, ripening
slowly; seeds brown, pitted, without a thick corky
| Of: |< Sane RO igi ee ooo bes, OM at lad ts um Be IU Dae. 5. Coryphantha
KEY TO THE SPECIES oF OpuntTIA MILL.
Joints cylindrical and tuberculate.
Spines smooth, covered with loose, hyaline sheaths.
Stem slender and solid, with a woody axis; tuber-
cles flattened; spines solitary or absent ...... 1. O. ramosissima
Stems thick and fleshy with a reticulated, cylindri-
cal axis; tubercles raised; spines more than
one.
Tubercles two to three times as long as wide;
fruit: dry, with stoutespines 2.02 9.27. 2. O. acanthocarpa
Tubercles less than twice as long as wide.
Plants erect, with one or few main stems;
stems loosely branched; flowers yellow,
tinged with red; fruit dry, with stiff spines
on thesuppeér areoles!' sn 3. O. echinocarpa
Plants low, bushy, without a main stem; stems
compactly short branched; flowers green-
ish-yellow; fruit somewhat fleshy, yellow,
with a few delicate, deciduous spines ....
Spines rough, without sheaths; plants prostrate ....
Joints flattened, not tuberculate.
Joints spineless; flowers magenta; fruit dry, spine-
TOSS: | 25-2204 Se ee Be pee Ree Re ee ee 6. O. basilaris
Joints spiny; flowers yellow.
Areoles 2-3 cm. apart; spines white or with brown
base; fruit purple, juicy, spineless.
Fruit 3-3.5 cm. long, with green pulp; slopes
above Griffith’s mine at an elevation of 2450
. multigeniculata
. Parishii
Siler
eke)
MEET Si ia Gt ers se A a ae eee eae 7. O. charlestonensis
Fruit 4-6 cm. long, with purple pulp; at eleva-
tions of 1200-1350 meters ................. 8. O. phaeacantha
Areoles about 1 cm. apart; spines white (occasion-
ally brownish) ; fruit dry, spiny.
Spines acicular, 3-5 cm. long and bristle-like, up :
too D2cm longs ce ee ee ee: 9. O. erinacea
Spines all acicular, stiff, not over 2-3 cm. long .. 10. O. polyacantha
1. Opuntia RAMosissiMa Engelm. Am. Jour. Sci. ser. 2, 14: 339.
1852.
Mohave and Colorado deserts, from California, Nevada and
Arizona south to Sonora. Local habitat, widely scattered in
sandy or gravelly soil in the Larrea Belt. Best developed in
sandy soil east of Wilson’s ranch where the plants are about one-
half meter tall. Cottonwood Springs, altitude 1030 meters,
Clokey 8036; E-Spear ranch, altitude 1640 meters, Train 1743;
south of Indian Springs, altitude 1200 meters, Clokey 8026; east
of Wilson’s ranch, altitude 1100 meters, Clokey 8437. May to
August.
1943] CLOKEY: CHARLESTON MOUNTAINS CACTACEAE 69
2. OPUNTIA ACANTHOCARPA Engelm. & Bigel. Proc. Am. Acad.
3: 308. 1856.
Mohave and Colorado deserts from California to southern
Utah, south to Sonora. Local habitat, in sandy or gravelly soil
in the Larrea Belt. Most abundant at Wilson’s ranch. Wilson’s
ranch, altitude 1150 meters, Clokey S024, 8433. June.
3. OPUNTIA ECHINOcCARPA Engelm. & Bigel. Proc. Am. Acad. 3:
305. 1856.
Mohave and Colorado deserts of California, east to Utah and
Arizona, south to Lower California. Local habitat, local in the
upper Larrea and lower Juniper belts. Most abundant in Kyle
Canyon at an elevation of about 1600 meters. Kyle Canyon,
Clokey 7202, 8025. June.
4. Opuntia multigeniculata Clokey sp. nov. E Sect. Cylin-
dropuntia, humilis, subadscendens, habitu conferta, plus minusve
0.5 m. alta, ad 1.5 m. lata, ramorum ligno debili, reticulato; ramis
majoribus depresso-tuberculatis, ca. 1.5—2 em. crassis, geniculis
lateralibus numerosis, 3-5 cm. longis, ca. 2 cm. crassis, tuberculis
confertis 4-6 mm. longis, 2-3 mm. latis, 4-5 mm. altis, apice
areola pallide brunneo-lanosa, 4-5 mm. longa, 2—2.5 mm. lata
coronatis; aculeis ca. 12, geniculum ipsum fere occultantibus,
delicatis, vaginis albidis barbulatis; aculeis centralibus 2—4, 15-18
em. longis, lateralibus gracillimis; foliis teretibus apiculatis ca.
2 mm. longis; glochidiis ex areola summa albidis, 1.5 mm. longis;
floribus in apice geniculi aggregatis, ca. 2.5 cm. longis totidemque
latis; ovario conferte tuberculato, areolis prominentibus, ellip-
soideis, lana pallide brunnea; aculeis paucis, pergracilibus, de-
ciduis; perianthii laciniis pallide viridi-luteis, spatulatis, obtusis
apiculatisve, 15-18 mm. longis; stylo cum stigmatibus luteis, stig-
matum lobis 6-8; staminibus luteis; fructu globoso, luteo, sub-
carnoso, ca. 2 cm. longo, tuberculato, exaculeato, profunde
umbilicato; hypanthii cavo pulpa incolora farcto; seminibus
rotundatis, pallide luteis, haud nitidis, levibus, 2.5-8 mm. di-
ametientibus, commissura brevi, lata, manifesta.
A low, semi-ascending, compact Cylindropuntia, 0.5 m. or less
high, up to 1.5 m. wide, with stems having a weak, reticulated,
woody framework; main stems low tuberculate, about 1.5-2 cm.
thick, with crowded lateral joints 3—5 cm. long, about 2 cm. thick;
lateral joints with closely placed tubercles, 4-6 mm. long, 2-3
mm. wide, 4-5 mm. high, with the entire upper end of the tuber-
cle occupied by an areole 4-5 mm. long, 2—2.5 mm. wide, filled
with light tan-colored wool; spines about 12, almost concealing
the surface of the joints, delicate, white-sheathed, barbed; 2 to 4
central spines 15-18 mm. long; lateral spines very slender; leaves
terete, apiculate, about 2 mm. long; glochids white, 1.5 mm. long,
from the upper end of the areoles; flowers clustered at the tip
70 MADRONO [Vol. 7
of the joints, about 2.5 cm. long and broad; ovary closely tu-
berculate; areoles prominent, oval, filled with light tan-colored
wool; spines few, very delicate, deciduous; perianth-segments
light greenish-yellow (Ridgway’s Col. Stand. and Nom. plate 5,
25 YG-—Y, b), spatulate, obtuse or apiculate, 15-18 mm. long;
style and stigmas yellow; stigma lobes 6 to 8; stamens yellow;
fruit globose, yellow, with somewhat fleshy walls, about 2 cm. high,
tuberculate, spineless, deeply umbilicate; seed cavity filled with
colorless jelly; seeds circular, light yellow, dull, smooth, 2.5-3
mm. in diameter; commissure short, broad, distinct.
Definitely known only from the type locality on an open,
rocky ridge east of Wilson’s ranch, Charleston Mountains, Clark
County, Nevada, along road from Blue Diamond mill to the mine,
at an elevation of 1400 meters; Clokey S430 (type), S639, 8760.
Flower, May; fruit, July to September.
Mr. Fred Gibson of the Boyce Thompson Southwestern Arbo-
retum states that plants similar to this are found near Prescott,
at Congress, and along the Colorado River in Arizona. These
have not been examined.
Opuntia multigeniculata is intermediate between the series
Thurberianae and Echinocarpae. In the Echinocarpae it is
closest to O. echinocarpa Engelm. & Bigel. It differs from this
species in having weaker, less woody, shorter and more crowded
stems. The fruit of O. echinocarpa is dry. The somewhat fleshy
fruit and the seeds resemble O. Whipplet Engelm. & Bigel. The
joints, however, are thicker, shorter, more crowded and more
thickly armed than O. Whipplei. In the original description Engel-
mann and Bigelow (Proc. Am. Acad. 3: 807. 1856), state that
the color of the flowers of O. Whipple is red. Britton and Rose
(Cact. 1: 55-56. 1919) and Benson and Thornber (Cact. Ariz.
36. 1940) give the color of the flowers as yellow.
Neither O. echinocarpa nor O. Whipplet grow near the type
locality of O. multigeniculata.
5. OPUNTIA PaRisHm Orcutt, West. Am. Sci. 10: 81. 1896.
Mohave Desert of California and southern Nevada. Local
habitat, locally abundant in sand or gravelly soil in the Larrea
Belt at elevations of 1000 to 1200 meters. Cottonwood Springs,
Clokey 8028; south of Indian Springs, Clokey 8027; east of Wil-
son’s ranch, Clokey 8434. June.
Drifting sand usually partially covers the old joints. This is
a characteristic plant that could not be mistaken for any other
Cylindropuntia in this region.
EXPLANATION OF THE Ficures, PLATE 4.
Puate 4, CAcTACEAE OF THE CHARLESTON Mountains, Nevapa. Fig. A,
Opuntia multigeniculata Clokey. Fig. B, Opuntia acanthocarpa Engelm. and
Bigel. Fig. C, Opuntia charlestonensis Clokey.
Figure A
Figure B
Ficure C
Puate 4, CACTACEAE OF THE CHARLESTON Mountrarns, NEVADA.
' at ma
‘ al eet i
= t 2 my - «
ie 4y P
i ma lone :
" i i
ion h '
' ; *
i co
on
: < : ‘
ae *
: \
ah
N
'
‘
‘ (
'
'
?
=)
~
2
r , '
|
‘
.
‘
‘
. 4 2 ,
)
ees)
7 4
' .
rl
|
sy ‘
\
\
‘ ;
\ mt: ,
: vn
1943] CLOKEY: CHARLESTON MOUNTAINS CACTACEAE 71
6. Opuntia BasiLaRis Engelm. & Bigel. Proc. Am. Acad. 3:
298. 1856.
Colorado and Mohave Deserts and surrounding mountains
from California to Utah, south to Sonora. Local habitat, widely
scattered in the Larrea Belt below 2000 meters. Clark Canyon,
Clokey & Anderson 7201; southeast Indian Springs, Train 1759;
Kyle Canyon, Clokey 7205, 8032; Trout Creek Canyon, Clokey &
Anderson 7206. May.
7. Opuntia charlestonensis Clokey sp. nov. Humilis, patens
ad 0.5 m. alta, 1.5 m. lata, ramis primo adscendentibus demum
prostratis, tum geniculis 2—3 junioribus tantum suberectis; genicu-
lis apicalibus vel lateralibus, ovalibus vel obovatis, saepius apice
rotundatis, 10-18 cm. longis, 10—12 em. latis, viridi-lutescentibus,
primo vere rubentibus; areolis rotundatis vel ovatis, 5-6 mm.
latis, 2-2.5 cm. distantibus; glochidiis 3-4 mm. longis, pallide
brunneis in areolae apice; aculeis 4-6 acicularibus longitudine
ludentibus, longissimis ad 4.5 em., subcomplanatis tortisque,
undique vertentibus, albidis vel albidis basi pallide brunneis;
floribus primo laete luteis demum conferte roseis rubro-suffusis,
4—6 cm. longis, ca. 4 cm. latis; perianthii laciniis externis ovatis,
acutis, mucronatis, internis ovatis, apice rotundatis saepius mucro-
natis; staminibus luteis; stylo cum stigmatibus luteis vel rubes-
centibus; fructu ellipsoideo 3—3.5 cm. longo, 1.5—2 cm. crasso,
sordide purpurascente, cortice externo pulpaque viridibus; se-
minibus applanatis, 4-5 mm. diametientibus.
A low spreading plant up to 0.5 m,. high, 1.5 m. wide, the main
branches at first ascending, later prostrate with only the younger
two to three joints ascending; joints arising from the faces as
well as from the edges of older joints, oval or obovate, mostly
rounded at tip, 10-18 cm. long, 10-12 cm. wide, yellowish-green,
in the spring purplish; areoles circular or oval, 5-6 mm. across,
2—2.5 cm. apart; glochids in the upper part of the areoles, 3—4
mm. long, light brown;-spines four to six, acicular, varying in
length, the longest up to 4.5 cm. long, somewhat flattened and
twisted, spreading in all directions, white or white with light
brown base; flowers clear yellow at first, turning salmon tinged
with red, 4-6 cm. long, about 4 cm. wide; outer perianth-segments
ovate, acute, mucronate; inner segments ovate, rounded at tip,
usually mucronate; stamens yellow; style and stigmas yellow or
reddish tinged; fruit oval, 3—-3.5 cm. long, 1.5—2 em. thick, dull
reddish-purple, with green rind and pulp; seeds flat, 4-5 mm. in
diameter,
Known only from hillsides adjacent to Griffith’s mine, associ-
ated with Pinus monophylla Torr. & Frém. and P. scopulorum
(Engelm.) Lemmon, at an elevation of about 2450 meters, Clokey
7208, 7592, 8029 (type), 8688, 8770. Flower, July to August;
fruit, September to October.
72 MADRONO [Vol. 7
As suggested by Dr. Ira L. Wiggins, O. charlestonensis is most
closely related to O. megacarpa Griffiths, a plant of the western
edges of the Mohave and Colorado deserts in California. O.
megacarpa has joints 20-80 cm. long, fruit 7-12 cm. long and
seeds 7-8 mm. in diameter.
8. OPUNTIA PHAEACANTHA Engelm. in Gray, Mem. Am. Acad.
4:52. 1849.
Texas and Chihuahua to Arizona and the Charleston Moun-
tains, Clark County, Nevada. Local habitat, among Quercus,
Fraxinus and Amelanchier at Wilson’s ranch and in open wash on
gravelly soil at the mouth of Pine Canyon. Wilson’s ranch, alti-
tude 1180 meters, Clokey 8031, 8424, 8431, 8761; mouth of Pine
Canyon, altitude 1350 meters, Clokey S656, 8685. Flower, May
to June; fruit, July to August.
The plants of O. phaeacantha Engelm. from the Charleston
Mountains do not agree entirely with the original description but,
except for the size of the fruit, fall well within the range included
in that highly variable species by Britton and Rose (Cact. 1: 144.
1919), and Benson and Thornber (Cacti Ariz. 58. 1940). Bois-
sevain and Davidson (Colo. Cacti. 12. 1940) give the size of the
fruit as 4 to 6 centimeters in length, which agrees with the fruit
of our plants. At Wilson’s ranch, where the joints are up to 20
centimeters long, O. phaeacantha grows among Quercus, Fraxinus
and Amelanchier. This protection may well account for the un-
usual size of the joints. The red color shown in winter and
spring on the joints at Pine Canyon, but not at Wilson’s ranch,
is evidently due to cold. Cuttings from the Pine Canyon station
grown in South Pasadena remain green throughout the winter.
9. OpuNTIA ERINACEA Engelm. Proc. Am. Acad. 8: 301. 1856.
Mohave Desert of California, east to Utah and Arizona.
Local habitat, most abundant on a dry ridge east of Wilson's
ranch along road from Blue Diamond mill to the mine, at an ele-
vation of 1200 to 1400 meters. Mountain Springs, elevation 1700
meters, Clokey $423; ridge east of Wilson’s ranch, Clokey 8033,
$436. June to July.
10. OpuNTIA PoLtyAcANTHA Haworth, Suppl. Succ. 82. 1819.
North Dakota to Washington, south to Texas, Nevada and
Arizona. Local habitat, occasional in the Juniper Belt. Harris
Springs road, elevation 1800 meters, Clokey 7591; Lee Canyon,
elevation 1800 meters, Clokey 7204; Willow Springs, elevation
1200 meters, Clokey S030.
Key To THE SPECIES OF ECHINOCEREUS ENGELM.
Plants forming large clusters with many heads; spines
few, white or gray, long and flexuous; flowers scarlet,
funnelform: 2262329 ee eee 1. E. mohavensis
Plants with few heads, ascending; spines white to brown,
stout, curved; flowers purple, broad ................ 2. E. Engelmannii
1943] CLOKEY: CHARLESTON MOUNTAINS CACTACEAE 73
1. EcutNocEereus MoHAVENSIS (Engelm. & Bigel.) Rimpler,
Forst. Handb. Cact. 2: 803. 1885.
Cereus mohavensis Engelm. & Bigel. in Engelm. Am. Acad.
Arts and Sci. Proc. 3: 281. 1856.
Southeastern California to southern Utah, northern Arizona
and south to Mexico. Local habitat, occasional as single plants
in the Juniper Belt extending in places to the Pinyon and Yellow
Pine belts. Grows in gravelly soil and on rock ledges. Cathe-
dral Rock, elevation 2425 meters, Clokey 7211; Clark Canyon,
elevation 1800 meters, Clokey & Anderson 7200; Charleston Park,
elevation 2300 meters, Clokey 5034, elevation 8000 feet, Alexander
755; Kyle Canyon, elevation 2425 meters, Clokey 7210, elevation
1700 to 2400 meters, Clokey 8435; Lee Canyon, elevation 2670
meters, Clokey, Clokey & Baker 7597. May, June.
2. EcuiINocerEeus ENGELMANNU (Parry) Riimpler, Forst. Handb.
Cact. 2: 805. 1885.
Cereus Engelmannu Parry ex Engelm. Am. Jour. Sci. ser. 2,
14: 338. 1852.
Mohave and Colorado deserts of California, east to Utah and
Arizona, south to Mexico. Local habitat, widely scattered in the
Larrea and lower Juniper belts at elevations up to 1400 meters.
Kyle Canyon, Clokey 7590; Trout Creek Fan, Clokey 7207.
KEY TO THE SPECIES oF Ecuinocactrus MILL.
Spines stout, flattened, annulate; flowers yellow.
Stems globose, aggregate; axils of scales of ovary and
UNTO WO OVI ee ect pee ae a wk Fe a ds Pda dN BG a 1. H. polycephalus
Stems simple, globose to cylindrical; axils of scales on
ovary and fruit not woolly ....................... 2. H. acanthodes
Spines subulate, not annulate; scales on ovary and fruit
few, without wool in the axils; flowers amaranth
purple (Ridgway) to pink ...........2.<..-2s.0.54. 3. EH. Johnsonii
1, Ecurnocacrus potycerHatus Engelm. & Bigel. Proc. Am.
Acad. 8: 276. 1856.
Eastern Mohave Desert of California, east to Utah and Ari-
zona, south to northern Sonora. Local habitat, sparsely scat-
tered in gravelly soil or on rock ledges in the Larrea Belt. South
of Indian Springs, elevation 1300 meters, Clokey 7598, elevation
1250 meters, Clokey S432. July.
2. Ecuinocactus acantHopes Lemaire, Cact. Gen. Nov. Sp.
106. 18389.
Ferocactus acanthodes (Lemaire) Britton & Rose, Cact. 3: 129.
1922.
Southern California to southern Nevada, Arizona and Lower
California. Local habitat, on gravelly hills and rock ledges in
the Larrea Belt at elevations of 1100 to 1400 meters. Most abun-
dant on a ridge east of Wilson’s ranch along road from Blue
Th MADRONO [Vol. 7
Diamond mill to the mine. South of Indian Springs, Clokey 7212,
7593; rocky ridge east of Wilson’s ranch, Clokey 8428, 8429.
3. Ecuinocactus JoHNsonut Parry in Engelm. Bot. King’s
Geol? Expl 40th Par] 5:1 7s 21871:
Ferocactus Johnsoni (Parry), Britt. & Rose, Cact. 3: 141.
1922.
Southwestern Utah, northwestern Arizona, southern Nevada
and extreme eastern Inyo County, California. Although Echino-
cactus Johnsoni so far has not been found in the Charleston
Mountains it is to be expected since it occurs both to the east
and to the west of the range. It occurs very locally on hot, steep,
gravelly slopes in the Larrea Belt and should be looked for on the
lowest foothills. East of the range: Frenchman’s mine, 7 miles
east of Las Vegas, elevation 630 meters, Clokey 5900; ridge south
of Logandale, Clark County, elevation 550 meters, Clokey 5901.
West of the range the type locality of Hchinocactus Johnsoni
Parry var. octocentrus Coult. is in the mountains east of Resting
Springs, Inyo County, California.
The number of central spines is rather indefinite and variable
even on the same plant. The spines graduate in thickness and
length from the largest centrals to the shortest laterals so that
there is no satisfactory line of demarcation.
MammiLuaria Haw.
1. MAMMILLARIA TETRANCISTRA Engelm, Am. Jour. Sci. ser. 2,
14:°387. 1852.
Phellosperma tetrancistra (Engelm.) Britt. & Rose, Cact. 4: 60.
19238.
Southern Utah to Nevada, Arizona and southeastern Cali-
fornia. Infrequent and isolated. Local habitat, known in the
Charleston Mountains from a single specimen collected on a
gravelly hillside in the Larrea Belt south of Indian Springs at an
elevation of 1250 meters, Clokey 8037 (Clokey Herbarium).
Key To THE SPECIES OF CoRYPHANTHA (ENGELM.) LEMAIRE
Flowers 2-3 cm. long, straw-colored with pinkish midrib to tinged
with pink throughout; fruit green; seeds brown; Larrea and
Jui per WelES® . oo oe ose Ae oy tn oa Ae ee 1. C. deserti
Flowers 3.5-4.5 cm. long, eugenia red to old rose (Ridgway) ;
fruit more or less rose-colored; seeds reddish-brown; associ-
ated with Pinus monophylla Torr. & Frem., P. scopulorum
(Engelm.) Lemmon and Cercocarpus ledifolius Nutt. ...... 2. C. rosea
EXPLANATION OF THE FiGuRES, PLATE 5.
Puate 5. CAcracear oF THE CHARLESTON Mountarns, Nevapa. Figs. a, b,
Opuntia acanthocarpa Engelm. and Bigel. Figs. c, d, Opuntia echinocarpa
Engelm. and Bigel. Figs. e, f, g, Opuntia multigeniculata Clokey. Fig. h,
Opuntia echinocarpa Engelm. and Bigel. Fig. i, Coryphantha rosea Clokey.
PLATE 5. CACTACEAE OF THE CHARLESTON Mountatins, NEvADA.
1943] CLOKEY: CHARLESTON MOUNTAINS CACTACEAE 15
1. CorYPHANTHA DEsERTI (Engelm.) Britt. & Rose, Cact. 4: 46.
1923.
Mammillaria deserti Engelm. Bot. Calif. 2: 449. 1880.
Eastern Mohave Desert of California to southern Nevada and
possibly northwestern Arizona. Local habitat, widely scattered
on gravelly hills and in washes in the Larrea and lower Juniper
belts at elevations of 1100 to 1900 meters. It is most abundant in
the wash below Rocky Gap Springs. The flowers, which last
several days, are open only in the afternoons. MHarris Springs
road, Clokey 7794; Kyle Canyon at mouth, Clokey 7594; Kyle Can-
yon at junction of Harris Springs road, Clokey 8425, 8426; Rocky
Gap Springs, Clokey 8718.
Jepson (FI. Calif. 2: 550. 1936) regards C. deserti as a syno-
nym of C. chlorantha (Engelm.) Britt. & Rose and reports the
flower color as yellow or stramineous. C. chlorantha (Mammil-
laria chlorantha Engelm. in Rothr. Bot. Wheeler 127. 1878) was
originally described as having yellow or greenish-yellow flowers
and white stigma lobes. Engelmann described the petals of M.
deserti (Bot. Calif. 2: 449. 1880) as light straw color turning to
purplish at tips. On the examination of the flowers of hundreds
of plants in the Charleston Mountains and over 300 dried flowers
the writer failed to find a single flower without a distinct pink
tinge at least on the midrib of the perianth-segments. This pink
color was not confined to the tips but extended throughout the
length of the petals. The flowers are 2.5—8 cm. long; the stigma
lobes are yellow or partially tinged with pink. Several collec-
tions of C. deserti in 1942 by J. P. Hester from southern Clark
County, Nevada, and eastern San Bernardino County, California,
one of the latter from near to or possibly from the type locality,
are excellent matches for the Charleston Mountains plants. F. W.
Gould 1589 and Hester Utah 1, both from the Beaverdam Moun-
tains west of St. George, Utah, from near the type locality, agree
with the original description of C. chlorantha. The flowers are
3.5—-4 em. long, the petals are yellow or greenish-yellow without
a trace of pink, the stigma lobes are yellowish. Hester reports
that the flowers of C. chlorantha are fragrant while those of C.
deserti are not. From the study of this material it is evident that
C. deserti should not be considered even as a variety of C. chlo-
rantha.
2. Coryphantha rosea Clokey sp. nov. Globosa vel ovata,
4—15 em. alta, 4-10 cm. crassa, tota aculeis dense tecta; tuberculis
cylindricis, 10-15 mm. longis, supra totis sulcatis, sulcis areo-
lisque tenellis albido-lanosis; aculeis 28—35 acicularibus, centrali
semper angulo recto insidente, 5-10 mm. longis, saepius apice
rubentibus (raro nigricantibus), centrali in lateralibus aculeis
colore crassitudine intermediis sensim transeunte; floribus in
axillis tuberculorum ad plantae ipsius apicem rubris vel roseis,
3.5—4.5 em. longis, squamis tubi laciniisque externis e basi angus-
16 MADRONO [Vol. 7
tatis, acuminatis, longe fimbriatis; laciniis internis similibus neque
fimbriatis; filamentis luteis; stylo cum stigmatibus luteis vel
roseis, stigmatis lobis linearibus; fructu ovoideo, praesucculento,
perianthii reliquiis siccis persistentibus, tarde maturante, plus
minusve rubro-colorato, 2—2.5 cm. longo, 1.2—1.4 cm. crasso;
semine rubro-brunneo, foveolato, 2 mm. longo, 1.25 mm. crasso.
Globular to oval, simple, 4-15 em. high, 4-10 em. thick ; arma-
ment dense, hiding the surface of the plant; tubercles cylindric,
10-15 mm. long, grooved through the length on the upper side,
with tufts of white wool in the axils and areoles of the younger
tubercles; spines 28 to 35, acicular, central permanently at right
angles to the surface of the plant, 5-10 mm. long, more or less
tipped with reddish-brown (rarely black), the central spines
grading in thickness and depth of color to the laterals, on the
upper, younger tubercles the spines are nearly erect, with age the
spines lose their color and radiate almost at right angles to the
tubercles; flowers borne in the axils of the young spiny tubercles
forming a circle at the top of the plant, eugenia red to old rose
(Ridgway’s Color Stand. and Nom. plate XIII, I’ red), 3.5—4.5 em.
long; scales on the tube and outer perianth-segments tapering
from the base, acuminate, long fimbriate ; inner perianth segments
similar but not fimbriate; filaments yellow; style and stigmas yel-
low or rose; stigma-lobes linear; fruit oval, very juicy, with the
withered perianth persisting, ripening slowly, more or less colored
with eugenia red or old rose, 2—2.5 cm. long, 1.2—1.4 em. thick;
seeds reddish-brown, pitted, 2 mm. long, 1.25 mm. wide.
Widely but sparsely scattered. Cold Creek, with Juniperus
utahensis (Engelm.) Lemmon at 1850 meters, Clokey 8035; ridge
south of Deer Creek, with Cercocarpus ledifolius Nutt. at 2650
meters, Clokey 7595, 7596; ridge along Lee Canyon with Juniperus
utahensis (Engelm.) Lemmon at 1900 meters, Clokey 7209; Kyle
Canyon to Deer Creek with C. ledifolius and Pinus monophylla
Torr. & Frém. at 2400 meters, Clokey 7208, 8038 (type); Kyle
Canyon with Pinus scopulorum (Engelm.) Lemmon at 2200 meters,
fruit ripe, Clokey 8769. Flower, June; fruit, September, October.
Coryphantha rosea is quite similar to C. deserti and C. chlorantha
(Engelm.) Britt. and Rose in the appearance of the armament
and in the shape and ciliation of the perianth-segments. It differs
from C. chlorantha in the color of the flowers and fruit and from
C. deserti in the size and color of the flowers and in the color of
the fruit. It grows at higher elevations than C. chlorantha and
C. deserti and, so far as known, is confined to the middle eleva-
tions of the Charleston Mountains. No intermediate plants have
been found. C. Alversonit (Coult.) Orcutt, a plant growing at
lower elevations in extreme southeastern California, has darker,
heavier spines and is reported to have white stigma-lobes and
green fruit.
South Pasadena, California,
April, 1942.
1943] CRONQUIST: IDAHO PLANTS 17
NOTEWORTHY PLANTS FROM IDAHO. II
ARTHUR CRONQUIST
During the summer of 1941 Mrs. Cronquist and I spent two
months collecting plants in central Idaho. A complete set is de-
posited at the Missouri Botanical Garden, and nearly complete
sets at the Gray Herbarium and the University of Idaho, Southern
Branch; partial sets are deposited at the University of Minnesota
and the Utah State Agricultural College. Several collections
represent plants not previously known to occur in the state; a few
are apparently undescribed. In the following discussion, unless
otherwise indicated, the collection numbers are my own.
There are in Idaho at least two entities that have commonly
been referred to Melica bulbosa Geyer ex Port. and Coult. One,
the usual form of the species, is single-stemmed or loosely cespi-
tose, with obvious bulbs at the bases of the culms. The other is
very densely cespitose, growing in large tufts, with the bases of
the culms only slightly if at all bulbous. Although the degree
of bulbousness of the culms is known to be a variable character
in M. bulbosa, the appearance of the cespitose form is so striking
that in the field it might well be taken for a different species.
The clumps are very dense, and difficult to separate into speci-
mens suitable for herbarium sheets. It may be significant that in
1899 Aven Nelson considered the habit of specimens he collected
so remarkable that he noted on the labels sent out with them, “in
dense tufts with matted roots.” Dr. W. S. Boyle, of the Uni-
versity of California, who is making a cytogenetic study of the
genus, writes me that the cespitose plants I sent him do not differ
sufficiently from the usual form to warrant separate treatment.
Since the two forms are decidedly different in appearance, and
are readily distinguishable in the field, I prefer to regard them
as two varieties of one species.
MELIca BULBOSA Geyer var. typicanom. nov. M. bulbosa Geyer
ex Port. and Coult. Syn. Fl. Colo. 149. 1874.
MEtica BULBosA Geyer var. caespitosa var. nov. Planta caespi-
tosa culmis ad basim haud vel vix bulbosis, cetera similis var.
typicae.
Type. Open rocky granitic slope above west side of Alturas
Lake, Blaine County, Idaho, altitude 7200 feet, June 22, 1941,
2603 (Missouri Botanical Garden; isotypes, Utah State Agricul-
tural College, University of Idaho, Southern Branch). Cotypes.
Ipano. Hillside above Mill Creek, 12 miles west of Challis, Custer
County, altitude 7800 feet, July 8, 1941, 2977; granitic outcrop
above Wildhorse Creek, 25 miles southwest of Chilly, Custer
County, altitude 7900 feet, July 28, 1941, 33854. Wvyomine. Teton
Mountains, Uintah County, August 16, 1899, Nelson and Nelson
6540.
18 MADRONO [Vol. 7
Siranion HaNseni X AGROPYRON sPicaTtuM. On a rocky granitic
hillside above Bull Trout Lake, Boise County, 35 miles west of
Stanley at an altitude of 7000 feet, a colony of highly variable
hordeaceous grasses was observed. A few spikes of the extreme
of the robust plants were taken with the other plants of the same
number (3615). A mile upstream and 200 feet higher on the
same hillside, specimens of Agropyron spicatum (Pursh) Scribn.
and Smith were seen and collected (3635). On examining my
number 3615 in the herbarium, I find that the large spikes are
quite uniform and represent Sitanion Hansent (Scribn.) J. G.
Smith. Other specimens of this number are so extremely vari-
able as to be abnormal, transcending the differences between
Sitanion and Agropyron. Spikes with all the spikelets solitary and
with all the spikelets paired at the nodes occur on the same plant.
One culm branches at the summit into two variable spikes. Tests
made on a few rachises indicate that they disarticulate, but only
on considerable urging. Individual portions of some of the spikes
are strongly reminiscent of Agropyron spicatum. It is my opinion
that the specimens represent a natural hybrid between Sitanion
Hanseni and Agropyron spicatum.
Carex obovoidea sp. nov. Planta caespitosa, culmi 20—40 cm.
alti, vaginae ventraliter rubro-maculatae, ad apices rubiginosae,
inferae aphyllae; spiculae plures androgynae, coarctatae in ca-
pitulum 15-30 mm. longum, 8-20 mm. latum; squamae late
ovatae, castaneae, marginalibus perspicuae; utriculi 8-3.5 mm.
longi, 1 mm. lati, plerumque attenuatis ab basibus spongiosi-
turgidus, dorsaliter paucinervosi, ventraliter enervosi; achaenium
obovatum, haud vel vix stipitatum, 1.5 mm. longum, minus quam
1 mm. latum; stigmata 2.
Caespitose, 20-40 em. high, aphyllopodic; sheaths red-dotted
ventrally and copper tinged at the mouths; spikes small, numer-
ous in an ovoid or oblong head 15-30 mm. long, 8-20 mm. wide,
sometimes slightly interrupted below, androgynous; scales broadly
ovate, brown, with conspicuous hyaline margins; perigynium
corky-thickened at the base, yellowish brown, changing to green
upwards, or somewhat green-margined, 3—3.5 mm. long, 1 mm.
wide, tapering from base to apex, or a little more strongly nar-
rowed near the middle, prominently few-nerved dorsally, nerve-
less ventrally, margins serrulate, beak obliquely cut and dark
brown at the tip; achene lenticular, obovate, broadest above the
middle, scarcely stipitate, 1.5 mm. long, less than 1 mm. wide;
stigmas 2.
Type. Mossy rocky meadow along tiny spring above Beaver
Creek, near Marsh Creek, 25 miles northwest of Stanley, Custer
County, Idaho, altitude 6500 feet, July 3, 1941, 2872 (Missouri
Botanical Garden; isotypes at University of Idaho, Southern
Branch and Utah State Agricultural College).
1943] CRONQUIST: IDAHO PLANTS 79
This species is a member of the small section Vulpinae, as
treated by Mackenzie (1). It differs from the related C. neu-
rophora Mack. and C. nervina Bailey in having the perigynia ven-
trally nerveless, and the sheaths ventrally red-dotted and copper-
colored at the mouths. It differs from C. Cusicki Mack., of the
related section Paniculatae, in its shorter stature, more contracted
inflorescence, spongy-thickened perigynium-base, and tapering
rather than abruptly contracted perigynium. It differs from all
of these in its obovate achenes.
Astragalus reclinatus sp. nov. Planta perennis prostrata,
caulibus 1 mm. crassis ad 50 cm. longis plus minusve strigosis ;
folia similiter strigosa foliolis 1-5 (plerumque 3) foliolo extremo
20-30 mm. longo, 8-5 mm. lato, eis lateralibus reductis vel de-
fectis, stipulae virides parvulae 2—3 mm. longae, flores albidi-pur-
purei 12 mm. longi alis rostrum superantibus calyce 6 mm. longo,
fructus sessilis 15-17 mm. longus, 3—4 mm. latus albido-strigosus
suturis prominentibus haud intrusis.
Stems several from a perennial tap root, prostrate, spreading,
up to 50 cm. long or more, about 1 mm. thick, green, striate, more
or less pubescent with thick, short, nearly or quite basifixed hairs;
leaves similarly pubescent, scattered, short-petiolate, with 1-5
(commonly 3) linear to narrowly lanceolate leaflets, the terminal
one enlarged, mostly 20-30 mm. long and 38-5 mm. wide, tapering
and acute at both ends, the lateral ones commonly somewhat re-
duced, mostly 10-20 mm. long and 1-3 mm. wide, sometimes
absent; stipules tiny, green, triangular or ovate, free, 2-3 mm.
long; flowers whitish, with a trace of purple, mostly 2—10 in short
racemes; banner 12 mm. long, abruptly bent nearly at right
angles 5 mm. from the base, the expanded portion 9 mm. long,
8 mm. wide and suborbicular when spread out; wings about 4
mm. wide and 11 mm. long, including the 5 mm. claw; keel 9 mm.
long, 4 mm. high at the end, blunt; calyx with short black and
white hairs intermingled in varying proportions, the tube 3.5—4
mm. long, the teeth triangular and about 2 mm. long; pod sessile,
I-celled, mostly 15-17 mm. long and 3-4 mm. wide, little-com-
pressed, closely strigose with short white hairs, sutures prominent
and not at all intruded.
Type. Moist alkaline bottoms along roadside two miles
south of Dickey, Custer County, Idaho, altitude 6300 feet, July 14,
1941, 3086 (University of Minnesota; isotypes at Missouri Bo-
tanical Garden, Utah State Agricultural College and University
of Idaho, Southern Branch).
This species of the section Homalobus is related to A. flexuosus
Dougl. ex G. Don and the polymorphic A. decumbens (Nutt. ex
T. & G.) Gray. The latter is common in the general area from
which the type of A. reclinatus was taken, but was not seen in the
immediate vicinity, nor have I ever seen it in the same type of
80 MADRONO [Vol. 7
habitat. 4d. fleruosus, predominantly a species of the great plains,
is not known to occur in Idaho.
The most conspicuous difference between A. reclinatus and its
close relatives is its prostrate habit. Probably more important
is the type of leaf. In A. flexuosus the lateral leaflets are several
to numerous and well developed. The leaves of A. decumbens are
highly variable, the leaflets often being in part reduced to phyl-
lodia, but the terminal leaflet is affected as much as the others,
and variation in that direction connects the species with the
closely related A. diversifolius Gray, a rush-like plant in which the
leaves are reduced to naked or nearly naked narrow rachises. In
A. reclinatus the tendency is for the terminal leaflet to be enlarged
at the expense of the others. None of the leaves on my speci-
mens have more than five leaflets, and some are quite simple.
The flower, although superficially resembling that of the
larger-flowered types of A. decumbens (“A. hylophilus’ for
example), shows its own distinctive differences. In A. decumbens
the wings scarcely surpass the keel; in d. reclinatus the wings
obviously surpass the keel, though not so prominently as in A.
flexuosus. The upward widening of the pods commonly found in
A. decumbens is inconspicuous or absent in A. reclinatus. ‘The
cross section of the pod is less compressed than in A. decumbens,
and a little more so than in A. fleruosus. The stipules in A. recli-
natus are tiny and green; in dA. fleruosus and A. decumbens they
are larger, mostly 5 mm. long or more, and somewhat scarious.
Although A. reclinatus is represented by only one collection,
its distinguishing features, extending from technical characters of
the flower to leaf type and habit, are so pronounced that I see
little room for doubt as to the validity of the species.
GENTIANA cCaLycosa Griseb. subsp. asepaLaA Maguire. Madrojfio
6: 151. 1942. G. idahoensis Gandoger, Bull. Bot. Soc. Fr. 65:
60. 1918.
This recently described entity is represented in my collections
by the following numbers: 3640, 3681, 3756, and 3757. Prior to
the publication of Maguire’s paper (2), this material had been
identified as G. idahoensis Gandoger, the type of which was col-
lected by Evermann at Pettit Lake, Blaine County, Idaho. At
that time I doubted that it was more than subspecifically distinct
from G. calycosa. It is now evident that G. idahoensis should be
added to the synonymy of G. calycosa subsp. asepala.
Agastache Cusickii (Greenm.) Heller was described from the
Steens Mountains, Oregon, and is known to Peck (3) only from
the Steens Mountains. An isotype is in the University of Minne-
sota herbarium. oO Doe
1943 | CRONQUIST: THALICTRUM POLYCARPUM 83
THE LEGITIMACY OF THALICTRUM POLY-
CARPUM WATS.
ARTHUR CRONQUIST
In a recent article in Madrofio (7: 1-4. 1943), Croizat main-
tains that the name Thalictrum polycarpum Loret (1859), although
illegitimate and not validly published, renders 7. polycarpum
Wats. (1879) illegitimate, as a later homonym. He cites article
61 of the Rules, in part, as follows: “Even if the earlier homonym
is illegitimate, or is generally regarded as a synonym on taxo-
nomic grounds, the later homonym must be rejected.”
Croizat admits that T. polycarpum Loret was not validly pub-
lished. The part of article 61 which he does not quote, states:
“A name of a taxonomic group is illegitimate and must be re-
jected if it is a later homonym, that is, if it duplicates a name
previously and validly published for a group of the same rank
based on a different type.’ (Italics mine.) The Rules are clear
enough; a name must be validly published if it is to prevent the
later use of the same name for a different plant. To further
clinch the case, article 19 states: “A name of a taxonomic group
has no status under the Rules, and no claim to recognition by
botanists, unless it is validly published.”’
It seems clear that the name T’.. polycarpum Loret, since it was
not validly published, cannot illegitimize T. polycarpum Wats.
University of Minnesota,
Minneapolis, Minnesota,
February, 1943.
VALID AND LEGITIMATE NAMES—AND THALICTRUM
POLYCARPUM S. WATS.
C. A. WEATHERBY
To devise a set of rules which shall deal adequately with the
almost endlessly various nomenclatural situations which exist or
may arise is an exceedingly difficult task. It is therefore not sur-
prising that, in the eight years since the publication of the current
Rules of Botanical Nomenclature, a number of cases have been
brought to light which are not directly covered by the rules or in
which the application of the rules is doubtful. It is not an un-
reasonable hope that, through the exercise of that faculty mis-
called common sense, through carefully considered discussion of
questions actually arising in the course of taxonomic investiga-
tion, through decisions by a competent tribunal and cautious
amendment of the rules where really necessary, these doubtful
points gradually can be made clear. Their solution has not been
helped by much of the theoretical discussion of nomenclature for
84 MADRONO [Vol. 7
its own sake published since 1935, especially when it raises
illusory difficulties for the bewilderment of the unwary.
Unhappily, Dr. Croizat’s discussion of the name of Thalictrum
polycarpum S. Wats. (Madrofio 7: 1-4. 1943), must be classed
with those which obscure rather than elucidate their subject.
There is no real ambiguity in the use of the terms valid and legiti-
mate and their opposites in the rules. A valid name is one pub-
lished according to the prescriptions of articles 86 to 45 and
otherwise in accordance with the rules (art. 16, note). A valid
name is ipso facto legitimate; that term, however, is reserved for
requirements other than publication. An illegitimate name is
one which, though validly published, fails to meet other require-
ments of the rules (arts. 60 to 69). A fully invalid name is like
the last state of the wicked in Buddhist theology—it simply does
not exist. Being, for nomenclatural purposes, non-existent, the
rules pay no attention to it.
Any confusion which Dr. Croizat detects arises, not from the
rules, but from his own unnecessary attempt to redefine the above
terms. The first sentence of article 61, the most important for
the present question, has entirely eluded him. It reads: ““A name
of a taxonomic group is illegitimate and must be rejected if it is
a later homonym, that is if it duplicates a name previously and
validly published [italics here are mine] for a group of the same
rank based on a different type.” That is, the earlier homonym
must have been validly published in order to bring about rejection
of the later one. According to the rule voted at Amsterdam, a
provisional name is not validly published; therefore it cannot be
a bar to the later independent use of the same name.
The rule as to provisional names was formulated in the course
of debate at the Amsterdam Congress, as a substitute for a pro-
posal to outlaw them originally made at the preceding Cambridge
Congress and there referred to the Executive Committee (which
took no action). As published in the Journal of Botany (74: 75.
1986), it carries no definition of nomina provisoria (invalid) and
nomina alternativa (valid) and no adequate examples.’ Until it is
perfected there may be doubt as to whether a given name is pro-
visional or not. Had Dr. Croizat argued that such doubt existed
in regard to Loret’s names and that it was therefore safer to
employ Greene’s name for the Californian Thalictrum rather than
Watson’s, he would have been on firmer, though still highly un-
certain, ground and his argument would have been more to the
point. For the status of Loret’s names is, in the light of article
61, the only consideration of any importance in determining the
correct name for the species. If Thalictrum polycarpum Loret is
provisional, then T. polycarpum S. Wats. is the legitimate name
1 The reference to the original proposal does not help; there the definition
was drawn and the examples presumably selected to cover both kinds of names.
1943] ISENBERG: REDWOOD BARK 85
and must be used. If JT. polycarpum Loret is validly published,
then T. polycarpum Wats. becomes a later homonym and T.
ametrum Greene (or some other synonym; I do not know the tax-
onomy) must be taken up.
In view of the imperfect elaboration of the rule as to pro-
visional names, the only safe course seems to be to admit as
validly published under it only names which conform closely to
the example from de Wildeman given. He had a plant which he
unreservedly accepted as a new species, but he was in doubt as
to the genus to which he should refer it, and called it both Cymbo-
pogon Bequaerti and Andropogon Bequaerti. According to the rule,
both these names are validly published. But if de Wildeman had
written: “I do not feel justified in describing this as a new spe-
cies; but if I were to do so, I should call it either Cymbopogon
Bequaerti or Andropogon Bequaertiv’—that is, if he had not defi-
nitely accepted the species*—then, as I see it, both names would
be provisional and not validly published.
The hypothetical quotation is very near what Loret actually
did. On the above basis, then, 7’. polycarpum Loret is a pro-
visional name, not validly published and incapable of functioning
as an earlier homonym, and T. polycarpum S. Wats. is the correct
name for the species.
Gray Herbarium, Harvard University,
February, 1943.
THE ANATOMY OF REDWOOD BARK
Irvinc H. IsenBeRG
This paper is one of a series originating from the laboratories
of The Institute of Paper Chemistry, Appleton, Wisconsin, and
covering a fundamental study of the botanical, chemical and
other characteristics of the California redwood; this work has
been sponsored by The Pacific Lumber Company, Scotia, Cali-
fornia. The present paper is a portion of a botanical study made
by the author at the Institute in 1939. Although considerable
data have appeared in the literature on the anatomy of the wood
of Sequoia sempervirens (see J. N. Mitchell, Jour. Forestry 34:
988-93. 19386 for discussion and list of references) the only
published work on the anatomy of redwood bark seems to be that
included in the article by Abbe and Crafts, who examined the
phloem of white pine and other coniferous species (L. B. Abbe
and A. S. Crafts, Bot. Gaz. 100: 695-722. 1939). Their studies
2 On this point—the only definition of a provisional name we have—see Int.
Bot. Congress. 1930. Nomenclature. Prop. Brit. Botanists 16 (art. 44); Rec.
Synopt. 41 (art. 37 ter).
86 MADRONO [Vol. 7
were apparently confined to the inner bark and stressed matura-
tion of the sieve cells.
The bark of a merchantable redwood tree, Sequoia semper-
virens (Lambert) Endlicher, is composed of secondary phloem,
living and dead, and of periderm. The redwood is a thick-barked
tree; the bark is sometimes as much as a foot thick but is usually
much less. In external appearance it is reddish, deeply fur-
rowed, and scaly. In transverse section the bark appears as two
strikingly different colored rings—the very thin, whitish, inner
one and the thicker, reddish-brown, outer one. The light-colored
layer, which may have a pinkish tinge, rarely exceeds one quarter
of an inch in thickness.
Primary Bopy
Although the primary bark tissues disappear early in the life
of a redwood tree, a brief review of their structure seems to be in
order before proceeding to a more detailed study of the secondary
phloem. The primary body is self-sufficient and contains all the
fundamental tissues and body parts. A cross section of a red-
wood stem at the end of the first growing season shows (pl. 6)
the pith, primary xylem, cambium, primary phloem, cortex, and
epidermis, as well as the secondary xylem and phloem. An
accessory protective layer, the hypodermis, is located just be-
neath the epidermal layer. A few longitudinal resin canals are
also evident in the cortex in plate 6.
The primary tissues lying outside the cambium are pushed
outward by the development of secondary tissues. The increase
in circumference to which these tissues must accommodate them-
selves quickly surpasses their ability to respond, with the result
that, sooner or later, they are crushed or ruptured and killed by
exposure, and especially by the stoppage of food and water sup-
plies by the cork layers which develop within them. Obviously,
the secondary phloem is very important functionally to the tree,
because it soon replaces the primary phloem.
SrconpaRyY PHLOEM
The secondary phloem of redwood contains four types of
cells—sieve cell, longitudinal parenchyma, fiber, and ray paren- ~
chyma (pl. 8).
The sieve cell is the cell characteristic of secondary phloem
from the viewpoint both of structure and function. In some
specimens of redwood bark examined, the sieve cells comprised
the majority of the cells, both in the inner living and in the outer
dead bark; in others, there occurred greater amounts of phloem
parenchyma. Similar to other gymnosperms, the sieve cell ele-
ments are not arranged in series, end to end, forming definite con-
ducting lines, but are separate and distinct. The sieve plates are
scattered irregularly on the radial walls of the sieve cell ele-
Puate 6. SeQUOIA SEMPERVIRENS. Cross section of one year old stem.
x 73.
Priate 7. SEQUOIA SEMPERVIRENS. Cross section of inner and small part
of outer bark. X20. . CaM aM MMSE aa N a ei und SSE OAD Ee a weg BI Bie 8 lb. S. spathulifolium
subsp. pruinosum
AA. Leaves of rosettes green or glaucous, not de-
cidedly pruinose, 1-2 mm. thick; rosettes 1-5
cm. in diam.; stems of offsets 1-2 mm. thick
with the leaves usually restricted to the
rosettes which are terminal on the shoots .. B
B. Leaves glaucous, crenulate; stems of offsets
1.5-2 mm. in diam.; flowers 8-17 mm.
Ins GRAM Ok Ne een la. S. spathulifolium
subsp. typicum
BB. Leaves green, usually neither glaucous nor
pruinose, minutely crenulate; stems of
offsets 1-1.5 mm. in diam.; flowers 7-11
Mm. in diame te eee ee Ic. S. spathulifolium
subsp. anomalum
la. Sepum spatHutiroLt1um Hooker subsp. typicum. Sedum
spathulifolium Hooker, Flor. Bor. Am. 1: 227. 1834. 8S. californi-
cum Britton, Bull. N. Y. Bot. Gard. 3: 44. 1903. S. Woodu
Britton, No. Am. Flora 22: 73. 1905.
The distinctive features of subsp. typicum are the: glaucous
leaves, the more slender habit compared with subsp. pruinosum,
and the slightly larger size compared with subsp. anomalum.
These vegetative differences are well demonstrated by figure 3
(pl. 22). In floral characters, the subsp. typicum is likewise inter-
mediate between the other two subspecies. The three-parted
cymes usually contain from 20—40 flowers which are 8-17 mm. in
diameter.
EXPLANATION OF THE Figures. PLaTeE 22.
Prate 22. Hasir Stupies or Species oF SEDUM IN THE SUBGENUS GORMANIA.
Fic. 1. Sedum spathulifolium subsp. typicum, cultivated in greenhouse at
Ithaca; collected along Grave Creek, Josephine County, Oregon. Photo by
W.R. Fisher. Fic. 2. Sedum spathulifolium subsp. pruinosum, on rocks along
Pacific Ocean, south of Crescent City, California. Photo by Harold Trapido.
Fic. 3. Rosettes cultivated in greenhouse at Ithaca: a, Sedum Purdyi, from
canyon of Canyon Creek, Siskiyou County, California; b, Sedum spathulifolium
subsp. pruinosum, from rocks along Pacific Ocean south of Crescent City, Cali-
fornia; c, Sedum spathulifolium subsp. anomalum, from Yosemite National
Park, California; d, Sedum spathulifolium subsp. typicum, from Josephine
County, Oregon. Photo by W. R. Fisher. Fic. 4. Type collection of Sedwm
spathulifolium subsp. pruinosum var. “Cape Blanco,” cultivated at Ithaca.
Photo by W. R. Fisher.
169
SEDUM
CLAUSEN AND UHL
“VINVWHOY) SONADHAS AHL NI W
aqag
40 SaINGadG AO
SaIdaALS
LIGV fF]
0Z ALVIg
170 MADRONO [Vol. 7
The subsp. typicum is widely distributed in the northern Coast
Ranges, the Cascade Mountains and the northern part of the
Sierra Nevada. It occurs from the lower Bella Coola and Fraser
River valleys in British Columbia, to
the Santa Cruz Peninsula in California.
The altitudinal range is from 92 to
1920 meters. Since most of the speci-
mens examined are without informa-
tion about altitude, it is likely that this
range will be increased as more data
become available. The subsp. typicum
usually grows in rocky places in ravines
and canyons and on steep slopes.
Specimens seen. Highest altitude:
1920 m., 3 km. east of White Mt., Sis-
kiyou Mts., Siskiyou Co., Calif., Wheeler
3171 (NY); lowest altitude: 92 m., 3
km. south of Oregon City, Clackamas
Co., Ore., Clausen and Trapido 4925
(CU); northernmost: 52°30’ N., Burnt
Ridge Creek, 50 km. east of Bella Coola,
Skeena Co., B. C., McCabe 1510 (UC)
(Since the dried specimens are poor,
their identity is questionable. This oc-
currence should be checked by further
collections); easternmost: 120° W.,
Emigrant Gap, Calif., Jones 3345
(CLOK, NY); westernmost: 124° W.,
__ Fic. 2, Map showing Snow Camp Lookout, Curry Co., Ore.,
distribution of the species Leach 2292 (ORE): th t:
and subspecies of the section hide ( ); southernmost:
Rosulata of Sedum: @=S. 87°30’ N., Los Gatos Cafion near Los
spathulifolium subsp. typi- Gatos, Santa Clara Co., Calif., Heller
cum, x8. §. subsp. prt 7/22 (KM, UGC): oldest. Sepia omlismie
tae Sipe anom~ summit of Mt. Diablo, Calif., Brewer 842
‘ (UC); type: not seen (collected by
Douglas on dry rocks of the Columbia and Salmon rivers) ; types
of names assigned to synonymy: north side of Mount Shasta, Sis-
kiyou Co., California, alt. 5000-9000 ft., June 11-16, 1897, Brown
336 (type of S. californicum Britton, NY); Oregon City, Oregon,
1866, Wood (type of S. Woodii Britton, NY). Number of collec-
tions seen: 63. Specimens of this subspecies examined in con-
nection with the present study have been annotated as Sedum
spathulifolium Hooker, subspecies A.
Hooker described Sedum spathulifolium as common on dry
rocks of the Columbia and Salmon rivers. His description is not
sufficiently complete to make positively certain which of the sub-
species he was describing. Our basis for interpretation of the
name rests on the plants which are the prevalent kind in the
valley of the Columbia River. We have seen no specimens from
ee
1944] CLAUSEN AND UHL: SEDUM 171
as far inland as the Salmon River. The type specimen of S. cali-
fornicum looks like typical S. spathulifolium, but the leaves are
pruinose in the dried condition and Britton described the petals
as white. Possibly the color of the flowers may have faded, as
sometimes happens. Cooke, in a letter of January 13, 1941, and
in a publication (6), questioned whether the specimens of 8S. cali-
fornicum and other of Brown’s collections really originated on Mt.
Shasta. He favored the idea that they came from the eastern
ridges or ranges of the Klamath Province. According to Cooke,
A. A. Heller who knew Brown personally, wrote that no doubt
all of Brown’s Sedum collections were made somewhere on Mt.
Eddy. As far as we understand the present distribution of these
plants, this theory impresses us as reasonable, particularly since
Cooke has been unable to find the plant on the north side of Mt.
Shasta, and Trapido and the senior writer were likewise unsuc-
cessful in their search there in 1940. At Oregon City, the type
locality of S. Woodiu, Trapido and the senior writer found S.
spathulifolium frequent on rocks, but plants from there do not
seem distinctive.
The time of flowering and fruiting depends somewhat on the
latitude and on the altitude. Extreme dates of flowering for three
districts, as determined from data with herbarium specimens, are:
Santa Cruz Peninsula, Calif. ... April 28-May 28
Western Orevon 2.....:...-. May 9-June 24
Fraser River Valley, B.C. ..... June 15-July 15
A dwarf variety, collected originally at 1524 meters, at Snow
Camp Lookout in Curry Co., Ore., has been designated by Hender-
son (in M. E. Peck, Man. High. Plants Oregon. p. 360. 1941) as
var. minus. This may interest horticulturists and should be listed
as a variety under subsp. typicum. The type is in the herbarium
of the University of Oregon. Praeger’s var. purpureum should be
referred to subsp. pruinosum and his var. majus to subsp. anomalum,
though both these varieties have in the past been placed under
typical S. spathulifolium.
Plants of six collections have been studied cytologically. All
are n=15. Previously, Hollingshead (9) had reported a diploid
number of 30 for three of the collections of S. spathulifolium.
Collections for which the euploid number is now reported are:
Clausen and Trapido 4925, South Fork of Eel River, 8 km. west of
Hullville, Lake Co., Calif.; Clausen and Trapido 5010, north side
of Grave Creek, 1.4 km. east of junction with Rogue River,
Josephine Co., Ore.; Clausen and Trapido 5022, bluffs along east
side of Willamette River, 3 km. south of Oregon City, Clackamas
Co., Ore.; and Clausen 77, 141 and 324, all from cultivated sources.
The chromosome complements of these several collections are
sufficiently similar that one collection cannot be separated from
another on a basis of the chromosomes. There is some variation
172 MADRONO [Vol.7
in the size of the chromosomes in a complement, but none of the
chromosomes are sufficiently distinctive to be identified.
Sedum spathulifolium subsp. typicum is frequent in the horti-
cultural trade. It is grown in rock gardens and is usually cor-
rectly named as to species. Rarely it is listed as S. Woodii. The
var. minus also is offered. Other varieties are listed, but these
are either referable to other subspecies or are nomina nuda.
The first illustration of typical S. spathulifolium seems to have
been published in 1883 in the Garden (vol. 24, page 462). An
earlier plate which appeared in Gartenflora in 1872 is so poor that
there is doubt whether the plant figured really is S. spathulifolium.
1b. Sepum spaTHuLiFroLtium Hooker subsp. pruinosum (Brit-
ton) Clausen et Uhl, comb. nov. Sedum pruinosum Britton, No.
Am. Flora 22: 72. 1905.
The distinctive characteristics of subsp. pruinosum are the very
pruinose condition of the leaves, stems, floral bracts and calyx;
the large size, with the leaves of the rosettes 2-3 mm. thick; and
the stout sterile stems, 1.5-3 mm. thick, which are frequently
_ leafy throughout, whereas in the other two subspecies the leaves
usually occur only in dense rosettes at the ends of the shoots. In
cultivation at Ithaca, subsp. pruinosum maintains its distinctive
characteristics and thrives more successfully than do either of the
other two subspecies. Also it seems to prefer more moisture.
Whereas subsp. anomalum scarcely persists under the same con-
ditions of culture, subsp. pruinosum becomes robust, flowers and
spreads.
The subsp. pruinosum is restricted almost entirely to rocks and
bluffs along the Pacific Ocean from near Little River, Humboldt
County, California, to Nanaimo on Vancouver Island. In the “Pre-
liminary Catalogue of the Flora of Vancouver and Queen Char-
lotte Islands” (10), the report of Sedum spathulifolium from the
Queen Charlotte Islands may refer to subsp. pruinosum, but we
have not seen specimens from there. The altitudinal range is not
great, essentially from sea-level to about 30 meters or perhaps a
little higher. The plants trail, creep or even hang over the
boulders along the coast, often forming great patches or mats on
the rocks. They must frequently be bathed in the spray of the
surf. Two collections from along the lower course of the Colum-
bia River and several from the islands of Puget Sound suggest
that this population is connected with the population of subsp.
typicum of inland districts by specimens somewhat intermediate
in character and should not be segregated as a separate species.
This opinion must yet be checked by genetical experiments.
Specimens seen. Highest altitude: 30 m., coast bluffs near
Trinidad, Humboldt Co., Calif., Tracy 2582 (UC); lowest alti-
tude: 3 m., 8 km. south of Crescent City, Del Norte Co., Calif.,
Clausen and Trapido 4987 (BH, CU, NY, US); northernmost:
49°10’ N., Nanaimo, Rosendahl 1886 (CLOK); easternmost:
1944] CLAUSEN AND UHL: SEDUM 173
122°30’ W., Lummi Island, Whatcom Co., Wash., Muenscher 7930
(CU); westernmost: 124°10’ W., Gold Beach, Curry Co., Ore.,
Hoyt 64 (DS); southernmost: 41°2’ N., south of Little River,
Humboldt Co., Calif., Wiggins 5878 (FM, UC); oldest: 1887,
Cedar Hill, Vancouver Island, B. C., Macoun (NY); type: Cres-
cent City, California, 1908, Eastwood (NY). Number of collec-
tions seen: 82. Specimens of this subspecies, examined in con-
nection with the present study, have been annotated as Sedum
spathulifolium Hooker, subspecies B.
Miss Eastwood sent Dr. Britton a fresh piece of her original
collection. This was cultivated in New York and flowered there
on May 8, 1904 (N. Y. Bot. Gard. no. 18748, Rose 540). The type
appears pruinose and consists of two floral stems, plus a few
leaves. There are really no good rosettes on the specimen.
Living plants and herbarium specimens collected by Trapido and
the senior writer on cliffs and ledges along the shore of the Pacific
Ocean 8 km. south of Crescent City, on July 24, 1940 (Clausen
and T'rapido 4937) are good matches for the type of S. pruinosum
and have been our basis for interpretation of the name. —
Extreme dates of flowering are April 27 and July 24. The
flowering period is more prolonged than in subsp. typicum, also
it is earlier at any given latitude. Undoubtedly this feature may
be explained, at least partly, on a basis of the lower altitude at
which subsp. pruinosum occurs and the consequent milder climate.
Principal variations are in the color of the leaves and stems
and in size. Praeger (12, p. 239) has described a var. purpureum
with the leaves purple. We have grown this in Ithaca, having
obtained specimens from two different horticultural sources. The
plants are definitely referable to subsp. pruinosum, under which
the variety should be listed. At least two other unpublished
varietal names, likewise referable to subsp. pruinosum, are also in
the trade.
Plants of two collections have been studied cytologically:
W. C. Muenscher 7930 from Lummi Island, Washington; and
Clausen and Trapido 4937 from Crescent City, California. Both
collections are n= 15. Previously, Hollingshead (9) had reported
2n = 30 in root-tips of the Muenscher collection (7930).
The subsp. pruinosum is probably as common in the horticul-
tural trade as subsp. typicum, but it is usually offered as S. spathuli-
folium. Rarely it is listed as S. pruinosum. The var. purpureum
is available under that or other varietal names indicating reddish
foliage. This subspecies is definitely to be recommended for cul-
ture by eastern gardeners. Particularly desirable are plants of
the type which probably originated at Cape Blanco in Oregon.
We now name this as Sedum spathulifolium subsp. pruinosum var.
“Cape Blanco” var. nov. Hort., with the rosettes 1-2 em. in diam.,
usually densely crowded, and with the leaves small, convex dor-
sally and ascending. As type we designate the specimen (S 266),
174 MADRONO [Vol.7
cultivated at Cornell University, Ithaca, which is shown in figure
4 (pl. 22).
Figure 2 (pl. 22) shows the habit of the plant in nature. A
drawing of a flowering plant, probably to be identified as this
subspecies, appeared in Gardeners Chronicle in 1876 (vol. 5, p.
821). .
le. SepuM spPaTHULIFoLIUM Hooker subsp. anomalum (Brit-
ton) Clausen et Uhl, comb. nov. Gormania anomala Britton, Bull.
N. Y. Bot. Gard. 3: 30. 1903. Sedum yosemitense Britton, ibid.,
p. 44. S. anomalum (Britton) Britton, No. Am. Flora 22: 72.
1905.
The most distinctive characteristic of Sedum spathulifolium
subsp. anomalum, when seen in the field, is the non-glaucous con-
dition of the leaves. This feature is likewise exhibited by culti-
vated specimens, whereas plants of the other two subspecies,
when grown side by side with subsp. anomalum, under the same
circumstances, exhibit the glaucous or pruinose condition. An-
other feature of subsp. anomalum is the usually smaller size, with
the rosettes rarely more than 2.5 cm. in diameter. For the most
part, the stems of the offsets are rather slender, 1-1.5 mm. in
diameter.
The subsp. anomalum, found only in California, ranges along
the west slope of the Sierra Nevada, from the region of Yosemite
National Park southward to the San Bernardino and San Gabriel
mountains. Along the coast it extends as far north as the head-
waters of the Carmel River. It has been collected most fre-
quently in the Sierra Nevada and the San Bernardino and San
Gabriel mountains. The altitudinal range is from a little above
sea-level to 2286 meters. The usual habitats are rocky slopes
and hillsides, ravines and ledges.
Specimens seen. Catirornia. Highest altitude: 2286 meters,
San Bernardino Mts., Crawford 911 (POM); lowest altitude:
30 m. + ?, Gragg’s Canyon, San Luis Obispo Co., Moran 421
(BH); northernmost: 37°50’ N., Hetch Hetchy Valley, Yosemite
National Park, Hall and Babcock 3380 (UC); easternmost: 117°
W., Bear Creek, San Bernardino Mts., Pierce (UC) ; westernmost:
121°30’, Pine Valley, head of Carmel River, Goldman 762 (FM) ;
southernmost: 34°2’ N., pipe line trail on Yucaipa Mts., Reed 2769
(UC); oldest: 1865, Yosemite Valley and Mts., Torrey (NY);
type: sandy hills in the path of strong daily sea winds, San Luis
Obispo Co., June, 1883, Summers (NY); type of Sedum yosemitense
Britton (NY), assigned to synonymy: Yosemite Valley between
Vernal and Nevada Falls, 1677 m., July, 1902, Hall and Babcock
3425. Number of collections seen: 33. Specimens of this sub-
species, examined in connection with the present study, have been
annotated as Sedum spathulifolium Hooker, subspecies C.
A collection from the San Bernardino Mountains (Clausen and
Trapido 4748), is a reasonable match for the type of Britton’s |
1944] CLAUSEN AND UHL: SEDUM 175
Gormania anomala and has been used by us as a standard in our
comparison of living plants. A specimen from the type locality
of S. yosemitense, in Yosemite Valley below Nevada Falls (Clausen
and Trapido 4802), is a good match for the type of that species.
Plants of the collections cited above (4802 and 4748), when grown
under uniform conditions at Ithaca, became very similar and had
the same cultural requirements. There seemed to be no real
differences between these plants.
May and June are the principal months of flowering. Extreme
dates when flowering specimens have been collected are April 30
and July 4, with one record of it being found in flower in the San
| Bernardino Mountains in November.
Two varieties of subsp. anomalum occur, one with white flowers
and the other with the whole plant larger in size. The latter has
been named var. majus by Praeger (12, p. 238). Our basis for
referring it to subsp. anomalum rests on the fact that the leaves are
green, not glaucous. Large specimens, similar to what Praeger has
described, have been seen in the Santa Barbara Botanic Garden.
These came originally from San Luis Obispo County and were col-
lected by Reid V. Moran. This is the region of the type locality,
but other plants from there were not nearly so large. It is possi-
ble that variety majus is only a physiological form, a response to
richer soils, shade and increased moisture. If further collections
from San Luis Obispo County should reveal that most of the
plants from there are large and perhaps slightly glaucous, then
Gormania anomala may need to be referred to subsp. typicum and
the name yosemitense used for the subspecies of the Sierra Nevada
and the San Bernardino and San Gabriel mountains.
Two collections, studied cytologically, are n= 15. These are
from Cascade Canyon, San Gabriel Mountains, California (Clausen
and Trapido 4748), and from below Nevada Falls, Yosemite Val-
ley, California (Clausen and Trapido 4802).
Subspecies anomalum is rare in the horticultural trade. Judg-
ing from experience in growing it at Ithaca, it is dificult to main-
tain and less desirable than either of the other two subspecies.
Praeger’s (12) illustration of S. yosemitense shows fairly well
the aspect of subsp. anomalum.
2. Sepum Purpyi Jepson, FI]. Calif. 2: 110. 1936.
Perennial with the stems procumbent or creeping, bearing flat
rosettes of spatulate leaves; stems 1—4 mm. in diameter, 4—7 cm.
long, light green to brown; leaves of rosettes closely imbricated
with their apices pressed close together, spatulate or orbicular-
spatulate, rounded or truncate at apex, papillose on margins,
0.1-1.8 em. long, 1-10 mm. wide, 1 mm. thick, yellow-green,
granulose ventrally, waxy dorsally, sometimes lustrous; primary
stems with offsets below rosettes, these radiating like the spokes
of a wheel and naked except for the rosettes at their apices, 1.5—7
em. long, green; floral stems erect or slightly decumbent, 0-16
176 MADRONO [Vol.7
em. high, 2-3 mm. thick below the inflorescence, with the leaves
alternate, oblong-spatulate, spreading, 0.3-1.5 cm. long, 1-4 mm.
wide; inflorescence a three-parted cyme of 10 to 39 flowers, with
a central flower; floral bracts oblong-spatulate to linear, 2.5—5
mm. long; flowers sessile or short-stalked on pedicels to 2 mm.
long, usually 5-merous, 1.5 em. in diameter; sepals linear-lanceo-
late, obtuse or subacute, green, 2—2.5 mm. long; petals lanceolate,
acute, 5-7 mm. long, spreading above the base, bright yellow to
white; stamens 4 mm. long, anthers yellow or white; nectar scales
0.8 mm. long, 0.5 mm. wide, transversely oblong; pistils erect or
divergent, gibbous ventrally, 4-7 mm. long, with the styles 1-2
mm. long; seeds pyriform, yellowish-brown, 0.8 mm. long.
The present known range of Sedum Purdy is in the southern
portion of the Klamath Mountains in northern California. We
have seen specimens from altitudes of 183 to 617 meters, but Jep-
son mentioned the upper limit of altitudinal range as 4000 feet
(1219 meters). The usual habitats are ledges and rocky slopes in
shaded situations.
Specimens seen. Catirornia. Highest altitude: 617 meters,
east side of corral of Canyon Creek Camp, 1.6 km. up Canyon
Creek from Scott River, 19 km. south of Hamburg, Siskiyou Co.,
Clausen and Trapido 4983 (CU); lowest altitude: 183 meters,
Somes Bar, Siskiyou Co., Tracy 16271 (UC); northernmost:
41°36’ N., same as highest altitude; easternmost: 122°40’ W.,
near French Gulch, Shasta Co., Rose (CAS) (Jepson has cited the
easternmost locality as Kennett, but we have seen no specimens
from there) ; westernmost: 123°30’ W., same as lowest altitude;
southernmost: 40°42’ N., same as easternmost; oldest: Aug. 21,
1908, Etna Creek, Siskiyou Co., Butler 498 (CAS, UC); type: not
seen (Etna Mills, Calif., Carl Purdy). Number of collections
Seems 74
A collection from ridges along Canyon Creek, Siskiyou County,
California (Clausen and Trapido 4983) agrees for the most part
with characters given in the original description of S. Purdyi and
matches specimens from Mr. Purdy’s nursery. This collection
has been our basis for interpretation of the species. The rosettes
are flat with the leaves closely compressed; the margins of the
leaves are prominently papillose; and the petals are yellow, not
white as stated in the original description. For directions to the
locality at Canyon Creek, the senior writer wishes to express his
gratitude to Mr. J. T. Howell of the California Academy of
Sciences.
The flowering time is late in April. Flowering dates obtained
from herbarium specimens are April 23 and April 29. Fruiting
specimens with ripe seeds have been collected in late July. Plants
obtained in flower in the Marble Mountains.in August were proba-
bly blooming abnormally, since the usual time seems to be in the
spring.
| 1944] CLAUSEN AND UHL: SEDUM 177
papillose.
No noteworthy variations of Sedum Purdyi have come to our
attention. The species seems nearest to the subsp. anomalum of
| §. spathulifolium and, like that, is difficult to grow in the eastern
part of the continent. Some specimens of the subsp. anomalum
have appeared so similar to S. Purdyi that we have had difficulty
in separating them, but usually the rosettes of S. Purdyi are more
compressed and the margins of the leaves are more prominently
Plants of the collection from Canyon Creek Camp, cited above
(Clausen and Trapido 4983), have been studied cytologically.
These are n= 15.
Sedum Purdy is occasionally offered in the horticultural trade
and is usually correctly named.
Figure 3 (pl. 22) shows rosettes of the three subspecies of S.
spathulifolium and of S. Purdyi. Note the slender stems of the
secondary rosettes of the latter species.
SEcTION EUGORMANIA CLAUSEN
Gormania Britton, Bull. N. Y. Bot. Gard. 3: 29. 1903. Named
in honor of M. W. Gorman of Portland, Oregon. Type species:
Cotyledon oregonensis Watson (Sedum oregonense).
A detailed discussion of this section has already appeared in
print (3). For convenience, the key to species and the keys to
the subspecies of S. obtusatum and S. larum, with slight emenda-
tion, are repeated here. Otherwise, data are presented only when
they are supplementary to the information in the earlier paper.
Key To THE SPECIES OF EUGORMANIA
A. Inflorescence and upper part of stem glandular
pubescent, strongly reflexed before flowering
time; leaves of rosettes glandular-ciliate; petals
Vi LOU pate ree at os ame pt A Pert Went fey Se eel a 6. Sedum Moranii
AA. Inflorescence and upper part of stem glabrous,
usually erect before flowering time; leaves of
rosettes usually not ciliate; petals yellow, white
OM eaPO UT Keto es ap eae ete es AS ahs neste Dd a, is gok Rave Anti ccteseows B
B. Flowers yellow or pale yellow, sometimes fading
to white or pink in age; inflorescence a
paniculate cyme; leaves of rosettes 1-3.5
CO}00 teal 00) 01.25 ee Se gare a Ne a ana me a 3. Sedum obtusatum
BB. Flowers white, creamy white or pink; inflores-
cence a dense paniculate or corymbose
cyme; leaves of rosettes 1-4.5 cm. long .... C
C. Flowers white or creamy white; sepals
ovate, 2-3 mm. long ................ 4, Sedum oregonense
CC. Flowers pink or pale pink, rarely white;
sepals lanceolate or ovate, 2-5 mm.
ONY ee ee Ae ace hisire as MGs ont aie 5. Sedum laxum
3. SEDUM oBTUsATUM A. Gray
Mr. Jack Whitehead has sent some large specimens collected
originally by Mr. George B. Youngs from along the North Fork
178 MADRONO [Vol.7
of the Feather River, 4 km. north of Belden, Plumas County, Cali-
fornia, and at an altitude of about 915 meters. These are charac-
terized by very large rosette-leaves, up to 5 cm. long. The petals
are pale yellow. These plants seem referable to S. obtusatum,
but match neither subsp. typicum nor subsp. boreale. It is possible
that they may represent an undescribed subspecies, since no
material has previously been available from the Feather River
region. One might expect that Sedum obtusatum from that sec-
tion would be intermediate between the two described subspecies,
but samples of the plants obtained by Mr. Youngs are larger than
either of the other two races.
KEY TO THE SUBSPECIES OF SEDUM OBTUSATUM
A. Basal leaves relatively small, 0.5-2.5 cm. long, usu-
ally broadly rounded or truncate at apex; stems
of rosettes: bright Ted) 6. aa. a ee 3a. S. obtusatum
subsp. typicum
AA. Basal leaves usually larger, 1-3.5 cm. long, usually
retuse at apex; stems of rosettes pale red or
POM es gt 5 iss co tes anh ar GIR Set eee ae ee eed 3b. S. obtusatum
subsp. boreale
3a. SepuM opTusaTuM A. Gray subsp. TYPICUM.
No new data have accumulated regarding this subspecies.
3b. SEpUM ostusatTum A. Gray subsp. BorEALE Clausen.
A plant of the type collection (Clausen, Trapido and Cooke
4952) flowered in a greenhouse at Ithaca on May 7, 1943. The
stamens were red and the carpels were suffused with red. The
petals spread rather widely above the middle. The length of the
basal leaves ranges up to 3.5 cm. long. |
The senior writer wishes to express hearty thanks to W. B.
Cooke for guiding him and Mr. Trapido to the locality on Mt.
Shasta where subsp. boreale occurs.
4. SEDUM OREGONENSE (Watson) M. E. Peck.
The inflorescence is a paniculate cyme, not a true panicle. A
plant, originally from Crater Lake, which flowered indoors at
Ithaca, had yellow anthers.
5. Sepum Laxum (Britton) Berger.
Further study of living plants in cultivation necessitates a few
changes in the key to the subspecies.
Kry TO THE SUBSPECIES OF SEDUM LAXUM
A. Plants tall and robust, 25-45 cm. high; leaves
dark green to glaucous; stems of sterile
rosettes pink, 4-6 cm. long; petals pale to
deep: pink i ie et eee eee 5b. S. lawum
subsp. typicum
1944] CLAUSEN AND UHL: SEDUM 179
AA. Plants not as tall as above, 10-25 cm. high; leaves
yellow-green or blue-green, usually glaucous;
petals pink, pale pink to white suffused with
| OF 1a") hae RS: Rev Aidt ae nan en ees ed ae B
B. Inflorescence congested; rosettes closely
crowded, forming a dense mat, with the
leaves rather thin; stems of sterile
roseltes green to black, 1.5-2 cm. long 5d. S. laxum
subsp. retusum
BB. Inflorescence lax; rosettes not closely
crowded, forming a loose mat, with the
leaves thick and leathery ............. C
C. Rosette-leaves very broad, 2.0-3.0 cm.
wide, triangular, obcordate, sometimes
not glaucous; petals pale pink to white 5c. S.laxum
subsp. latifolium
CC. Rosette-leaves narrow, 0.3-2.0 cm.
wide, oblong-oblanceolate or spatu-
late, usually glaucous; petals pink D
D. Cauline leaves oblong-spatulate,
longer than broad .......... 5a. S.larum
subsp. perplexum
DD. Cauline leaves cordate to subcor-
date, about as broad as long 5e. S. laxum
subsp. Heckneri
5a. SepuM Laxum (Britton) Berger subsp. PERPLExUM Clausen.
No new data have come to hand.
5b. Sepum taxum (Britton) Berger subsp. tTypicum. Culti-
vated specimens originally from the type locality exhibit some
variation in the glaucous condition of the leaves. Some rosettes
are less glaucous than others. A collection of Clausen (5018) is
n=15. Previously, Dr. Hollingshead had reported a diploid num-
ber of 30 for this same collection. A cultivated plant, received as
Sedum Jepsoni, is Sedum laxum subsp. typicum. That binomial was
introduced into the literature by H. M. Butterfield (Desert Plant
Life 8: 7. 1936).
5c. Sepum Laxum (Britton) Berger subsp. Latirotium Clausen.
A specimen of the type collection, cultivated in a greenhouse at
Ithaca, flowered in the spring of 1943. The inflorescence was
15 cm. long and 6 cm. wide. The stamens were 6-7 cm. long with
the anthers red and the filaments white to pink. The pistils were
8 mm. long with the ovaries green and the styles and stigmas pink.
5d. Sepum taxum (Britton) Berger subsp. retusum (Rose)
Clausen. Two additional years of culture of this subspecies have
not affected its characteristic manner of growth. Cytological
study of University of California no. 28.3 from Mendocino
County, California, indicates that the haploid number is 15.
5e. S—EDUM Laxum (Britton) Berger subsp. Hecxnertr (Peck)
Clausen. No new data have become available regarding this
subspecies.
180 MADRONO [Vol. 7
6. Sepum Morani Clausen. No new localities for this rare
species have come to our attention.
SUMMARY
The subgenus Gormania of Sedum includes six species which
together constitute a natural phylogenetic group restricted to the
Pacific Mountain System of Western North America. Taxonomic,
geographical and cytological data all confirm this opinion. The
chromosome number, with one exception, is n= 15, 2n=30. The
exception is S. oregonense, of which the only plants so far studied
are hexaploid. The species may be grouped in two sections,
Rosulata and Eugormania, the former the more primitive. The
section Oreganica, based on Sedum oreganum, does not belong in
Gormania and should be referred elsewhere. The section Rosulata
includes two species, S. spathulifolium and S. Purdyi. The section
Eugormania includes the other four species: S. obtusatum, S. ore-
gonense, S. larum and S. Moranii. New botanical names are 8S.
spathulifolium subsp. pruinosum and S. s. subsp. anomalum. A
possibly undescribed subspecies of S. obtusatum is mentioned from
the Feather River region in Plumas County, California.
Department of Botany,
Cornell University, Ithaca, New York,
July, 1943.
LIvrERATURE CITED
1. Bercer, Atwin. Crassulaceae, in Engler and Prantl, Die Nat. Pflanzenfam.
ed. 2. 18a: 352-485. 1930.
2. Brirron, N. L. and J. N. Ross. Crassulaceae, in No. Am. Flora 22: 7-74.
1905.
3. Cxiausen, R. T. Studies in the Crassulaceae—III. Sedum, subgenus Gor-
mania, section Eugormania. Bull. Torrey Bot. Club 69: 27-40, figs.
1-3. 1942.
4, ————————. and Cuartes Unt. The taxonomy of the subgenus Gormania
of Sedum, in Abstracts of papers presented before the general, paleo-
botanical, physiological, and systematic sections of the Botanical Soci-
ety of America. Am. Jour. Bot. 29: 5s. 1942.
5, ————————- and ———————. Revision of Sedum Cockerellii and related
species. Brittonia 5: 33-46. 1943.
6. Cooxr, W. B. First supplement to the flora of Mount Shasta. Am. Midl.
Nat. 26: 74-84. 1941.
7. Fosperc, F. R. Subspecies and variety. Rhodora 44: 153-157. 1942.
8. Froperstrom, Harotp. The genus Sedum L. Act. Hort. Got. 10 (4).
App. 1-262, pl. 1-115. 1935.
9. HoxtinesHeap, Linrian. Chromosome studies in Sedum, prion Gor
mania, section Eugormania. Bull. Torrey Bot. Club 69: 41-48, figs. 1
1942.
10. [Kermope, F.] Preliminary catalogue of the flora of Vancouver and Queen
Charlotte Islands. Prov. Mus. Nat. Hist., Victoria, B. C., pp. 1-87.
1921.
11. Macoon, C. A. and Etmer Snyper. Grapes for different regions. U. S.
Dept. Agric., Farmers’ Bull. 1936: 1-38. 1943.
12. Prarcer, R. L. An account of the genus Sedum as found in cultivation.
Jour. Roy. Hort. Soc. 46: 1-314, figs. 1-185. 1921.
13. WeatHersy, C. A. Subspecies. Rhodora 44: 157-167. 1942.
1944] HOWELL: J. W. STACEY 181
J. W. STACEY, CARICOLOGIST
JoHN Tuomas Howey
That Mr. J. W. Stacey should have studied western carices was
inevitable. His interest in this dificult group had been aroused
years before by Charles Fay Wheeler when he was scarcely more
than a lad in the Middle West and under Wheeler he made his
first studies. For all the years that he had lived in San Francisco,
systematic botany had been his chief interest outside of his busi-
ness and the Glumiflorae had
been more attractive to him
than any other group. Hence,
in 1988, when Miss Eastwood
asked her long-time friend to
determine for her a formidable
accumulation of carices which
K. K. Mackenzie had just de-
clined to examine because of ill
health, Mr. Stacey gladly con-
sented. A concurrent reor-
ganization of his business al-
lowed him more time to himself
and it was not long before he
was deeply engrossed in a study
of Carex in Western North
America with ambitious plans
for a revision of the genus in
the United States west of the Fic. 1. J. W. Stacey, summer, 1942.
Rocky Mountains. Thus this
chain of events brought to Western Botany its only resident
caricologist and centered his work at the California Academy of
Sciences. For five years he pursued his systematic studies in-
tensively but, due to a heart ailment which finally prevented the
use of the microscope, he was unable to continue after 1939. Not
only did Mr. Stacey contribute a great deal to our knowledge of
western Carex but in this often-neglected genus he awakened a
widespread interest among field workers that for many years will
attest the influence of his vital and infectious enthusiasm.
Mr. Stacey was fundamentally interested in species and favored
all means for detecting them, whether by morphology, cytology,
or physiology. Although he himself relied chiefly on morphology
in his taxonomic work, he welcomed data from the other fields of
botany as a possible source of information for distinguishing
closely related entities. With well-balanced judgment and keen
discernment he made his analytical studies and he derived much
pleasure from searching out the identity of some miserable but
challenging fragment. He believed that species, when once
properly circumscribed, exhibited at least one character (not
182 MADRONO [Vol.7
necessarily a “key-character’) by which they could be recognized
at some period in their life history. When once he had this exclu-
sive character clearly in mind, he knew his species and could
recognize it no matter how variable it might be. Henceforth for
him the species was “fixed” and rarely could he be persuaded to
accept for it any subdivision into varieties.
Although his work with Carex was seriously conceived and
executed, nevertheless the time he spent in study was really his
period of recreation and relaxation away from business. It
pleased him to make a sociable time of it, discussing the plant
he had before him, expounding some theory of relationship or
distribution, criticizing or praising some treatment in the work
of Mackenzie for whom he had great admiration. In the her-
barium or on field trips he talked of little else but Carex, a source
of annoyance to non-botanical acquaintances who sometimes
accused him of being narrow-minded. One of his chief sources
of delight in conversation was to catch up unwary persons by
exposing some scientific fallacy or rhetorical exaggeration in their
speech. If a verbal tilt developed, he was never acrimonious nor
argumentative, always allowing his adversary the satisfaction of
justifying his statement; but from the twinkle in Mr. Stacey’s eye
one knew who had won, who had had fun. To fellow botanists
he was generous of his means and knowledge but only so long as
he felt he was being accorded an honest and true return in science
and friendship.
As an amateur botanist interested in the general systematic
botany of western vascular plants Mr. Stacey was one of the keen-
est and best-informed. Very few are the professional botanists
who could have surpassed him in a knowledge of plants as they
grow, a knowledge gained from methodical study while on recre-
ational strolls and trips over a period of many years. In an
attempt to organize the taxonomic data obtained on these outings,
he prepared lists of plants for each of the counties bordering San
Francisco Bay which are more complete than anything of which
I know and which contain many entries indicative of his keen
power of observation and his scholarly discernment.
The fullest scientific use of these lists cannot be realized
unfortunately because only the exceptional entry is represented
by a herbarium specimen for reference. Although a brilliant all-
around field naturalist, Mr. Stacey was in no sense of the word a
botanical collector. For him it was enough to search out the
rarest plants where they grew, leaving to amateur collectors or
professional botanists the task of preserving scientific herbarium
records. A few specimens of his collecting will be found in the
herbarium of the California Academy of Sciences but they are
frequently represented by mere scraps or fragments sent to Miss
Eastwood for determination or verification. After he seriously
undertook the study of western carices, he made a real effort to
1944] HOWELL: J. W. STACEY 183
try to collect; but his almost-untouched stock of printed labels
shows that his own activity as a collector fell far short of his
intentions. How well I remember the collecting trips that he and
I made to such rich localities as the Ledum swamp on the road to
Point Reyes or to the Pitkin marsh in Sonoma County: while Mr..
Stacey would freely wander about intent on plants in general and
Carex in particular, I would spend most of my time digging and
picking, collecting and pressing, specimens! ‘The superb research
collection of West American Carea in the herbarium of the Cali-
fornia Academy of Sciences, however, proves that one need not
be a collector to build a collection.
The published results of Mr. Stacey’s research in Carex ap-
peared chiefly as a series of “Notes on Carex” in Leaflets of
Western Botany, volumes 1 and 2, from November, 1934, to Feb-
ruary, 1939. There were fifteen contributions in this series which
include descriptions of seven new species besides extended distri-
butional data and notes on the identity of various critical species.
His only other botanical writings of which I know are “Notes on
some plant introductions, mostly Californian” (Leafl. West. Bot.
1: 69-71. 1938), and his treatment of the genus Carex in Kearney
and Peebles Flowering Plants and Ferns of Arizona (U.S. Dept.
Agric. Miscell. Publ. no. 423, pp. 168-175. 1942). A notice of
Mr. Stacey’s death appeared in Science (n. ser. 98: 464. 1943)
and a brief obituary was published in Academy News Letter no.
48 (December, 1943).
When no longer able to give his serious attention to Carez, Mr.
Stacey turned to another field of natural science and devoted his
time and interest to ornithology. To this study he brought the
same enthusiasm, the same critical discernment, which character-
ized his work in botany. Concerning his ornithological activity,
the following was written in Academy News Letter no. 48: “In
two or three years he had developed a field knowledge of birds
that was surprising even to veteran observers. The new interest
he had taken up at sixty-eight he pursued with the same intensity
and thoroughness that had characterized all of his activities
throughout his life; and after his death his executors found among
his papers literally almost half a bushel of carefully written
manuscript notes on birds.”
John William Stacey was born in Galesburg, Michigan, on
February 26, 1871, and died in San Francisco, California, October
16,1943. His wife, the former Mrs. Florence Ward Waite whom
he married in St. Louis shortly before coming to California, sur-
vives him. He attended the University of Michigan at Ann Arbor,
taking courses in medicine and botany, and at least during one
summer did botanical field work with C. F. Wheeler. After com-
pleting his medical course, he went to the Bellevue Hospital as an
interne, but instead of completing his work and practicing, he
specialized in the further study of drugs and took up editorial
184 MADRONO [Vol.7
work for medical publications. He came to San Francisco in
1914 where, shortly after, he became connected with the book
department of The Emporium, first as head of the medical book
department and later as head of the entire book department. In
1923 he founded the business firm of J. W. Stacey, Inc., which
soon attained importance as an institution in medical and scien-
tific circles throughout Western America. Evidence of the high
esteem in which Mr. Stacey’s store is held is observed in the fact
that medical students are urged by their instructors to browse
through his stock of books to acquaint themselves with the most
recent literature of their field.
Although Mr. Stacey was always a scientist at heart, with
particular interests in systematic botany and ornithology, these
fields of scientific endeavor did not attract him professionally.
His success in the business world was due in no small degree to
his deep attachment to science, for it was the rare combination of
his thoroughly scientific background, his naturally keen business
ability, and his happy and animated personality that accounted
for a career that was eminently successful and in some respects
unique.
California Academy of Sciences,
San Francisco, .
January, 1944.
DERMATITIS AND PHOTOSENSITIZATION PRODUCED
BY PTELEA ANGUSTIFOLIA
W.C. MuenscHerR AND BaBetTE I. Brown
In the summer of 1942:and again in 1943 workmen in High-
land Park, Rochester, New York, received a severe dermatitis
believed to have been caused by contact with the leaves of Ptelea
angustifolia Benth. (P. lutescens Greene), a member of the Rut-
aceae. Other plants in this family, Dictamnus albus L. and Ruta
graveolens L. are known to cause a dermatitis in susceptible indi-
viduals but no records could be found of dermatitis caused by any
species of Ptelea. ‘This note is a report of some tests made by us
which demonstrate that Ptelea angustifolia, a native shrub of the
Southwestern United States, also may cause a severe dermatitis
in some individuals.
The fresh material used in these tests was kindly supplied by
Mr. R. H. Horsey, from introduced shrubs growing in Highland
Park. We are indebted to Dr. Joseph N. Frost of Ithaca for his
interest in the tests, for advice, and for the treatment of some of
the more severe cases of dermatitis in two of the subjects.
Eight persons, all volunteers, were used as subjects in the
1944] MUENSCHER AND BROWN: DERMATITIS 185
tests: subjects 1 and 3, adult white females; subjects 2, 4, 7 and 8,
adult white males; subjects 5 and 6, adult oriental males.
EXPERIMENTS WITH PTELEA ANGUSTIFOLIA
The tests are recorded under time of exposure, subjects used,
method of application and reactions or results obtained with each
subject.
Exposure 1. August 4,9 A.M. Subjects 1 and 2. |
Crushed leaves were rubbed, on an area 2 cm. in diameter,
above the elbow of the left arm. Direct sunlight was excluded.
Subject 1. A slight reddening of the skin appeared
over an area 3 by 4 cm. in diameter four lays later. The
color remained the same for four days and then began to
fade. The spot was still visible on the tenth day when a
slight itching was perceived.
Subject 2. No reaction.
Exposure 2. August 4,9 A.M. Subjects 1 and 2.
Crushed immature fruits were rubbed over an area 2 cm. in
diameter below the left elbow. Direct sunlight was excluded.
No reaction occurred in either subject.
Exposure 8. August 7,1 P.M. Subjects 1, 2 and 38.
Crushed leaves were vigorously rubbed, on an area 3 cm. in
diameter, far above the elbow on the left arm of subjects 1, 2 and
3, and also on the right upper arm of subject 1 and on the center
of the upper back of subject 2. The areas were subsequently
exposed to direct sunlight for one hour. A severe dermatitis
developed in subjects 1 and 2 and a moderate dermatitis de-
veloped in subject 3. The development and history of the derma-
titis on subjects 1 and 2 are recorded below; the hours or days
indicate total time elapsed since the original exposure to contact
with the leaves.
Subject 1. Right arm.
18 hrs. A small faintly red area was discernible.
30 hrs. A red area, 5 by 7 cm., was definitely outlined; it
burned much as a sunburned skin.
40 hrs. The inflamed area, now 7 by 7 cm., was very sensi-
tive and pained when touched.
48 hrs. Frank vesiculation appeared and edema began to
develop.
52hrs. Smaller blisters became confluent; one large blister
had increased to 2 cm. in diameter.
54 hrs. Blisters were cut open, drained of yellow fluid and
covered with dressing and kept moist with dilute
boric acid for the following four days. After the
fifth day the inflammation began to subside and
considerable itching was experienced. The maxi-
186 MADRONO [Vol.7
mum size of the blistered area, 6 by 8 cm., was
reached on the sixth day. The blistered area
developed new skin within a few days but it re-
mained reddish-brown for two months.
The exposure on the left arm developed into a similarly severe
dermatitis on an area 5 by 13 cm. within thirty hours. Its general
course ran almost identical with that described for the right arm.
The same treatment was applied.
Subject 2. Left arm.
18 hrs. A reddish spot 6 by 6 cm. was discernible.
30 hrs. Inflamed area red, 7 by 10 em., with small whitish
vesicle near the center.
40 hrs. Inflamed area 7 by 12 cm., with vesiculation espe-
cially near the center.
52 hrs. Frank vesiculation but some blisters becoming con-
fluent, some 2 cm. in diameter and at least 1 cm.
high, edematous; several had ruptured and were
discharging a yellow fluid.
54 hrs. Blisters were cut open, drained and covered with
dressings and kept moist with dilute boric acid
solution for the following four days. After the
fifth day the inflammation began to subside but
considerable itching was experienced from then
on for ten days. Even six weeks after exposure
itching was severe whenever the subject perspired
or became over-heated. The affected area was
still reddish-brown after two months. The expo-
sure on the back ran a course very similar to that
on the arm except that the area was larger, 8 by
10 cm., with two radial streaks 2 by 6 cm. long
extending toward the neck.
Subject 38. Left arm.
A reddish spot 6 by 8 cm. appeared within eighteen hours.
It became more inflamed and small blisters developed
within three days. The general appearance of the derma-
titis was similar to that in subject 1 but much less severe.
The affected area remained dark reddish brown for two
months and then began to fade gradually but it was still
visible fifteen weeks after the beginning.
Exposure 4. August 9,7 A.M. Subjects 1 and 2.
Crushed leaves were rubbed on an area 8 cm. in diameter on
the inner side above the elbow of both arms (subject 1) and on
the right arm (subject 2). Direct sunlight was excluded.
No reactions were observed in either subject.
Exposure 5. August 14,2 P.M. Subjects 1, 2 and 3.
Crushed leaves were rubbed lightly, on areas 1 cm. square, on
the inner side of the right arm both above and below the elbow.
1944] MUENSCHER AND BROWN: DERMATITIS 187
Direct sunlight was excluded.
No reactions were observed in any subject.
Exposure 6. August 9,2 P.M. Subjects 4, 5, 6,7 and 8.
A small piece of crushed leaf was rubbed on an area approxi-
mately 2 cm. in diameter above and below the elbow of the left
arm of each subject. The arm was not exposed to direct sun-
light. Subject 4 developed a slight reddening of the skin of both
areas of exposure four days later. Subjects 5, 6,7 and 8 showed
no reactions.
EXPERIMENT WITH DICTAMNUS ALBUS
August 7. Subjects 1, 2 and 3.
Crushed leaves were rubbed on an area 2 cm. in diameter on
the right upper arm. The treated areas were exposed to diffuse
sunlight for one hour.
No reaction developed in subjects 1 and 38. A severe derma-
titis developed in subject 2. Within eighteen hours a red in-
flamed area, 5 cm. in diameter, had developed. It increased to
6 by 7 cm. and blisters developed. The general course of the
development of the dermatitis and its response to the boric-acid
solution treatment were very similar to that of Ptelea angustifolia.
EXPERIMENT WITH RUTA GRAVEOLENS
August 7. Subjects 1, 2 and 3.
Crushed leaves were rubbed on areas 2 cm. in diameter on the
outside of the lower right arm of subjects 1 and 8, and on the
extreme upper right arm of subject 2. The treated areas were
exposed to direct sunlight for one hour.
A slight reaction was obtained in subjects 1 and 3. After
thirty hours a definite reddening and inflammation of the skin was
discernible over an area 5 by 7 cm. in subject 1 and on an area
7 by 12 cm. in subject 3. This was followed by more severe
inflammation and itching but no vesiculation.
Subject 2 developed a severe dermatitis. Within eighteen
hours after contact a red inflamed area 8 cm. in diameter was
evident. After thirty hours the spot had increased to 10 by 11
em. with frank vesiculation. The blisters soon became confluent
and from then on the general appearance of the dermatitis and
its response to treatment with boric-acid solution were identical
to that caused by Ptelea angustifolia.
Of the eight subjects tested, three were highly susceptible to
contact with the leaves of Ptelea angustifolia, one was slightly sus-
ceptible and four showed no reaction under the conditions of the
tests. Subjects 1 and 2 developed only a slight reaction if not
exposed to direct sunlight after contact with the leaves but de-
veloped a severe dermatitis if exposed to direct sunlight for one
hour after contact. Of the five subjects who were tested without
188 MADRONO [Vol.7
subjecting them to direct sunlight after contact, only one showed
a slight reaction.
The three subjects who were highly susceptible to Ptelea an-
gustifolia were also tested for susceptibility to Dictamnus albus
(gas plant) and Ruta graveolens (rue). Subject 2 was highly
susceptible to all three species. Subjects 1 and 3 showed no
reaction to Dictamnus albus and a slight reaction to Ruta graveolens.
SUMMARY
Contact with the leaves of Ptelea angustifolia causes a derma-
titis in susceptible individuals. Exposure to direct sunlight
subsequent to contact increases the severity of the dermatitis.
This suggests that Ptelea angustifolia has a photosensitizing action.
The dermatitis caused by Ptelea angustifolia is very similar to that
produced by Dictamnus albus and Ruta graveolens. In the experi-
mentally produced dermatitis the first inflammation appeared
eighteen to thirty hours after contact with the leaves. The severe
cases continued for about ten days.
Department of Botany,
Cornell University,
December, 1943.
ON THE SHOOT APEX OF CHLOROGALUM
POMERIDIANUM (DC.) KUNTH?
CLARENCE STERLING
Recent investigations on the structure of apical meristems
have rekindled interest in the cyto-histology of the shoot apex
as contrasted to the more static formulation of cell-wall patterns.
(See reviews by Foster, 5; 6.) One of the main services of
these studies has been to show the essential lability inherent in
plant tissues and the consequent inadmissibility of posing strict
categories and formulae within which plant life is to function.
Probably one of the most rigid “laws” imposed on the angio-
sperm shoot apex is that it have at least one stratum of cells which
experiences anticlinal divisions exclusively, this cell layer being
called variously a “dermatogen” or “tunica.” Very few excep-
tions to this basic rule have been noted. Magnus (7) indicated
divisions in the dermatogen of the tip of the lateral pistillate inflo-
rescence of Secale cereale; Pottier figured such a pericline at the
tip of a branchlet of Ruppia maritima (8, fig. 77) and in the apical
meristem of the shoot of Cymodocea nodosa (8, fig. 197) ; and more
recently Sharman (11, 12) found periclines at the summits of the
1The writer wishes to acknowledge the helpful advice and criticism of Dr.
A. S. Foster in the preparation of the manuscript.
1944] STERLING: SHOOT APEX OF CHLOROGALUM 189
shoot apices of Zea mays and Agropyron repens, respectively. The
apices of Phoenix, as figured by Ball (2), also indicate irregulari-
ties in the uniseriate tunica, possibly involving sporadic peri-
clines. Aside from phylogenetic implications for the monocots,
these observations have a significance in challenging fixed con-
cepts on the structure of the angiosperm shoot apex.
_ The present discovery resulted during experimentation on
shoot apices with various fixatives. One of these experiments
Fic. 1. Longitudinal section in near-median view of shoot apex of Chloro-
galum pomeridianum (x 485).
involved specimens of Chlorogalum pomeridianum, collected in
April, 1941, near Fairfax in Marin County, California. Slides
were made of the shoot apices of two bulbs. The sections were
serial longitudinal, cut eight micra thick. Because of poor fix-
ation of the apices, drawings were made by the camera lucida
technique.
One of the apices, perhaps cut somewhat obliquely, shows
only slight indications of previous periclinal divisions in the sur-
face layer of cells. In the second apex, there is very definitely
no discrete surface layer. Periclinal divisions have occurred at
various places in this layer, both at the summit and on the flanks,
with high frequency. Text figure 1, which is definitely median
or near-median, shows several distinctive features of this apex:
190 MADRONO [Vol.7
derivatives of two periclinal divisions are observable at the sum-
mit of the shoot apex. An anticlinal mitosis on the left flank
indicates that cell division is active. Both leaf primordia show
periclinal divisions at their tips.
In text figure 2, the periclines in the surface layer of the leaf
primordia apices stand out particularly well. This section is only
eight micra removed from that of text figure 1, being the adjacent
cut. In this diagram, part of the wall of one of the periclinal
Fic. 2. Longitudinal section of shoot apex of Chlorogalum pomeridianum,
8 micra removed from section in fig. 1 (x 485).
divisions of the previous section and a different aspect of the
second pericline are recognizable at the shoot apex. Periclines
are visible on the left flank of the apical cone. The four cells here
seem to be the derivatives of a single superficial cell, possibly as
a result of a periclinal division immediately succeeded by anti-
clines in the daughter cells. The heavily-walled, deeply stained
cell at the upper right of the figure is seemingly merely coinci-
dental in occurrence in the apex.
Discussion
Agnes Arber (1) has noted, in her discussion of the morpho-
logical nature of leaf and shoot, the general equivalence of
1944 ] STERLING: SHOOT APEX OF CHLOROGALUM 191
“tunica” and “corpus”’ in the initial regions of stem and leaf in
angiosperms as well as the similarity, and even concurrence, in
the manner of apical growth of these two organs in most of the
general divisions of plants. This thesis is also a point of de-
parture for Catalano’s (4) support of Delpino’s phytonic concept.
These generalizations, however, are merely philosophical deriva-
tions from a group of basic anatomical data.
However, at present considerable data have been accumulated
on the behavior of leaf and shoot apices in the monocots. The
occurrence of periclinal divisions in the surface layer of the leaf
apex of these plants is a well-established fact. As Sharman (11)
points out, “In the Dicotyledons it may be involved in the pro-
duction of the leaf edge, while in the Monocotyledons it is fre-
quently concerned in the initiation of the leaves and often con-
tributes considerably to their inner tissues.” Sharman has cited
most of the investigators who have observed periclines in the
dermatogen of the monocot and dicot leaves. To his list can be
added the works of Renner (9), Buder (3), Pottier (8), and
Schalscha-Ehrenfeld (10).
It appears likely, therefore, that the shoot apices in monocots
might possibly have some tendency toward occasional periclinal
divisions in the surface layer. A further study of Chlorogalum
and related plants would help to cast more light on the present
rigid concept of “tunica” behavior in angiosperms.
Department of Botany,
University of California, Berkeley,
December, 1943.
LITERATURE CITED
1. Arser, A. The interpretation of leaf and root in the angiosperms. Biol.
Rev. 16: 81-105. 1941.
2. Batt, E. The development of the shoot apex and of the primary thickening
meristem in Phoenix canariensis Chaub., with comparisons to Washing-
tonia filifera Wats. and Trachycarpus excelsa Wendl. Amer. Jour.
Bot. 28: 820-832. 1941.
3. Buber, J. Studien an Laburnum Adami II. Allgemeine anatomische
Analyse des Mischlings und seiner Stammpflanzen. Zeits. f. indukt.
Abstammungslehre 5: 209-284. 1911.
4. Catratano, G. La natura morfologica dell’apice vegetativo dei germogli
alla luce della dotrina del fillopodio. Nuovo Giorn. Bot. Ital. 45: 594—
598. 1938.
5. Foster, A. S. Problems of structure, growth and evolution in the shoot
apex of seed plants. Bot. Rev. 5: 454-470. 1939.
6. ————————.. Comparative studies on the structure of the shoot apex in
seed plants. Bull. Torrey Bot. Club 68: 339-350. 1941.
7. Macnus, P. Bemerkungen zu J. H. L. Flégels Priiparaten. Cited in O.
Schiiepp’s Meristeme. 1926. Linsbauer’s Handb. d. Pflanzenanat. B.
IV.1 Abt.2 T. 1878.
8. Portier, J. Contribution a l’étude du développement de la racine de la tige
et de la feuille des phanerogames angiospermes. Les monocotylédones
marines méditerranéennes Ruppia maritima L., Cymodocea nodosa
(Ucria) Ascherson et Posidonia oceanica (L.) Delile de la famille des
potamogétonacées. Besancon. 1934.
192 MADRONO [Vol.7
9. Renner, O. Zur Entwicklungsgeschichte randpanaschierter und reingriiner
Blatter von Sambucus, Veronica, Pelargonium, Spiraea, Chlorophytum.
Flora 130: 454466. 1936.
10. ScHALsSCHA-EHRENFELD, M. v. Spross-vegetationspunkt und Blattanlage
bei einigen Monokotylen Wasserpflanzen (Potamogeton crispus, Heter-
anthera dubia, Typha angustifolia). Planta 31: 448-477. 1940.
11. SuHarman, B.C. A periclinal division in the ‘dermatogen’ at the tip of the
maize growing point. Nature 146: 778. 1940.
12, —_—————_.. A periclinal division in the ‘dermatogen’ at the growing
point of couch grass, Agropyron repens Beauv. Nature 152: 276-277.
1943.
NOTES AND NEWS
OENOTHERA BRACHYCARPA Gray IN Texas. Several collections
of this rare and seldom observed Oenothera in the subgenus
Megapterium have been made recently in north central Texas.
Citations for these are as follows: breaks along north side of
Brazos River, 6.5 miles south of Benjamin, Knox County, May 15,
1941, Cory 37205 (in flower); Camp Barkeley, Taylor County,
April 11, 1943, Tolstead 6960 (in bud), in red clay soil at west
side of camp, July 1, 1943, Tolstead 7537 (in fruit).
In the description of the subgenus Megapterium (North Ameri-
can species of the subgenera Lavauxia and Megapterium of the
genus Oenothera. Amer. Jour. Bot. 17: 363. 1930), Dr. Philip A.
Munz states, “‘Seeds as in Lavauzia, but in one row in each cell of
the capsule, and with corky tubercles.” Examination of the
capsules of the above cited fruiting specimen of Oe. brachycarpa
shows the seeds to be arranged in two rows in the capsules as in
the subgenus Lavauzria. This is also true for Oe. Wright (re-
garded as a variety of Oe. brachycarpa by some). Ocenothera mis-
souriensis and its relatives, also members of the subgenus Mega-
pterium, do, however, have their seeds arranged in one row in the
capsule. V.L. Cory, Texas Agricultural Experiment Station.
The following two items were received too late for review in
this issue of MaproNo but mention is here made to call them to
the attention of our readers. Under the title “The Citrus Indus-
try” by H. J. Webber and L. D. Batchelor (University of Cali-
fornia Press) an amazing lot of strictly botanical information is
hidden. For instance, Chapter IV by Walter T. Swingle, contains
345 pages entitled ‘The Botany of Citrus” and comprises a taxo-
nomic monograph of the Aurantioideae of the Rutaceae. Chapter
VI deals with the general morphology, histology and physiology
of the group and is under the authorship of E. T. Bartholomew
and H. S. Reed. The second item is “The Flowering Plants and
Ferns of Mount Diablo, California” by Mary L. Bowerman, pub-
lished by the Gillick Press, Berkeley, California.
MADRONO
A West American Journal of Botany
A quarterly journal devoted to important
and stimulating articles dealing with
plant morphology, physiology, taxonomy,
and botanical history. These volumes
should be a part of every botanist’s li-
brary and should be made accessible to
students of all universities and colleges.
Volume I, 1916-1929. . . $5.00
Volume II, 1930-1934 .. 5.00
Volume III, 1935-1936 . 5.00
Volume IV, 1937-1938 . 5.00
Volume V, 1939-1940 . . 5.00
Volume VI, 1941-1942 . 5.00
Single numbers....... 0.75
The subscription price of MADRONO
is $2.50 per year. We solicit your pat-
ronage,
Address all orders to:
William Hiesey Bus. Mgr.
Carnegie Institution of Washington
Stanford University, California
y
VOLUME VII NUMBER 7
: MADRONO
A WEST AMERICAN JOURNAL OF
BOTANY
4
: Contents
STRUCTURE AND TAXONOMY oF TAENIOMA, INCLUDING A DISCUSSION OF THE
PHYLOGENY OF THE CERAMIALES, Gleorge F. Papenfuss ............... 193
Nores on THE AtGaL GENUS TaENIoMA, C. K. Tseng ..................4.5. 215
Reviews: Gilbert M. Smith, Marine Algae of the Monterey Peninsula
1 (George F. Papenfuss); LeRoy Abrams, Illustrated Flora of the
a Poucise States (verbert 1.) Mason) yo) oso ee ik oO OW 226
Published at North Queen Street and McGovern Avenue,
Lancaster, Pennsylvania
| July, 1944
|
MADRONO
A WEST AMERICAN JOURNAL OF BOTANY
Board of Editors
Hersert L. Mason, University of California, Berkeley, Chairman.
LeRoy Asrams, Stanford University, California.
Epcar Anperson, Missouri Botanical Garden, St. Louis.
Lyman Benson, Pomona College, Claremont, California.
Hersert F’. Copetanp, Sacramento Junior College, Sacramento, California.
Ivan M. Jounston, Arnold Arboretum, Jamaica Plain, Massachusetts.
Mirperp E. Maruias, University of California, Berkeley.
Basserr Macuire, New York Botanical Garden, N. Y. C.
Marion Ownsey, State College of Washington, Pullman.
Secretary, Editorial Board—ANNeETTA CARTER
Department of Botany, University of California, Berkeley
Business Manager—Witiiam Huirsey
North Queen Street and McGovern Avenue, Lancaster, Pennsylvania
or
Carnegie Institution of Washington
Stanford University, California
Entered as second-class matter October 1, 1935, at the post office at
Lancaster, Pa., under the act of March 3, 1879.
Established 1916. Published quarterly. Subscription Price $2.50 per year.
Completed volumes I to V inclusive, $25.00; each volume $5.00; single numbers
$0.75.
Papers up to 15 or 20 pages are acceptable. Longer contributions may
be accepted if the excess costs of printing and illustration are borne by the
contributor. Range extensions and similar notes will be published in con-
densed form with a suitable title under the general heading “Notes and News.”
Articles may be submitted to any member of the editorial board. Manuscripts
may be included in the forthcoming issue provided that the contributor pay the
cost of the pages added to the issue to accommodate his article. Reprints of
any article are furnished at a cost of 4 pages, 50 copies $4.10; 100 copies $4.50;
additional 100’s $0.85; 8 pages, 50 copies $5.95; 100 copies $6.60; additional 100’s
$1.30; 16 pages, 50 copies $8.35; 100 copies $9.35; additional 100’s $2.00. Covers,
50 for $2.75; additional covers at $1.65 per hundred. Reprints should be
ordered when proofs are returned.
Published at North Queen Street and McGovern Avenue, Lancaster,
Pennsylvania, for the
CALIFORNIA BOTANICAL SOCIETY, INC.
President: E. B. Copeland, University of California, Berkeley. First ‘Vice-
President: L. R. Abrams, Stanford University. Second Vice-President: C. J.
Kraebel, California Forest and Range Experiment Station, University of Cali-
fornia, Berkeley. Secretary: Mary L. Bowerman, Department of Botany, Uni-
versity of California, Berkeley. ‘Treasurer: William Hiesey, Carnegie Institu-
tion of Washington, Stanford University, California.
Annual membership dues of the California Botanical Society are $2.50,
$2.00 of which is for a year’s subscription to Madrofio. Dues should be
remitted to the Treasurer. General correspondence and applications for
membership should be addressed to the Secretary.
1944] PAPENFUSS: TAENIOMA 193
STRUCTURE AND TAXONOMY OF TAENIOMA, INCLUD-
ING A DISCUSSION ON THE PHYLOGENY OF
THE CERAMIALES
GerorRGE F. PAPENFUSS
The genus T'aenioma was founded by J. Agardh (2) in 1863 to
receive a species which he (1) had previously described as Poly-
siphonia perpusilla. He placed Taenioma in the Rhodomelaceae
while Schmitz (41) in 1889 included it in his newly created Sarco-
menieae, a sub-family of the Delesseriaceae. In conjunction with
his monograph on the Rhodomelaceae, Falkenberg (14) also
studied T'aenioma with the specific purpose of demonstrating, by
way of contrast, the differences in construction of the thallus
between the Rhodomelaceae and Delesseriaceae. ‘The majority
of later writers (De Toni, 12, 13; Bérgesen, 5, 6; Kylin, 28; and
others) who have been concerned with this genus have followed
Schmitz and Falkenberg in assigning it to the Delesseriaceae.
Subsequent to the work of Falkenberg, the most important
paper on Taenioma has been that of Thompson (43) in 1910.
Thompson discovered cystocarps for the first time in the genus
and from a somewhat superficial study of them concluded that
Taenioma belonged in the Rhodomelaceae. Recently this view
was also adopted by Hollenberg (19). Thompson and Hollen-
berg, however, overlooked the most significant structural feature
distinguishing the Delesseriaceae from the Rhodomelaceae,
namely, the difference in the order of formation of the pericentral
cells. As was shown by Falkenberg, the pericentral cells in
Taenioma are formed in the manner characteristic of the Deles-
seriaceae. This question will be taken up more fully in later
pages, after the structure of the thallus has been reviewed.
Kylin (24) in 1923 pointed out that the Sarcomenieae should
be united with the Delesserieae, a conclusion borne out by the
work of Papenfuss (35) on Claudea and Vanvoorstia. Knowledge
concerning the structure and reproduction of the majority of
other genera in this old sub-family is meager. Through the work
of Falkenberg (14), Thompson (43), Thuret (9), Bgrgesen (5)
and Okamura (33), Taenioma has become one of the better-known
members of the Sarcomenieae. Although sexual organs were not
present in the writer’s material, certain observations on the struc-
ture of the thallus and the development of the tetrasporangia
seem worthy of record.
The anatomical work in the present study was made on
Hawaiian and South African plants of Taenioma perpusillum. The
South African material was kindly supplied by Dr. Mary A.
Pocock who collected it at Arniston (May 7, 1940) and Port Eliza-
beth (Dec. 8, 1942, with tetrasporangia). The species has been
ManroNo, Vol. 7, pp. 193-232. July 14, 1944
whl 49
ve
194 : MADRONO [Vol. 7
reported as occurring in South Africa. The record is based on
Kiitzing’s (23) Polysiphonia nana; but, as will be pointed out
farther on, it is very doubtful if P. nana is representative of
Taenioma. 'The Hawaiian material was collected by the writer
during the years 1940—42 at the following localities on the island
of Oahu: Hanauma Bay (March 380, 1941); 1.8 miles northwest
of Nanakuli (May 16, 1942, with tetrasporangia) ; Waikiki (Nov.
17, 1940, and Aug. 21,1941). Additional Hawaiian material was
kindly furnished by Mrs. D. Abbott who obtained it on the
ascidian Pyura momas, collected by members of the Department
of Zoology of the University of Hawaii, at Kaneohe Bay, Oahu
(Feb. 11, 1941, with tetrasporangia). All the material collected
by the writer was obtained in the intertidal zone, where the spe-
cies occurs as an epiphyte on Bornetella and a variety of small
turf-forming algae. T'. perpusillum had been reported from Ha-
waii in 1880 by Chamberlain (10) but the record has escaped
notice.
The Hawaiian and South African plants seem to be identical
in every respect. South African specimens received from Dr.
Pocock in 1939, while the writer was working at Lund, were com-
pared with the type material of T. perpusillum (nos. 43342 and
43343 in Herb. Agardh) and found to correspond very well.
Since only dried South African plants were available for the
present study, the following account is based entirely upon obser-
vations on Hawaiian material, which was preserved in formalin.
STRUCTURE OF THALLUS
Taenioma is a small alga, measuring less than three millimeters
in height, and consisting of a prostrate, terete, branched, inde-
terminate main axis, attached by non-septate rhizoids on the’
ventral side and forming erect, terete, secondary indeterminate
axes on the dorsal side. According to Falkenberg (14), the erect
branches are determinate, but from the writer’s observations there
seems to be no evidence for this. Both prostrate and erect axes
are monopodial and grow by means of a single transversely-
dividing initial. The axes are composed of segments consisting
of a central and four pericentral cells. The rhizoids arise as out-
growths from the ventral pericentral cells of the prostrate parts.
The ascending axes are formed alternately at an interspace of
three to eight segments, and by bending upward give the impres-
EXPLANATION OF THE Figures. PLATE 23.
Pirate 23. TAENIOMA PERPUSILLUM. Fic. 1. Portion of thallus showing an
erect axis with alternate determinate branches and with indeterminate branches
on the adaxial side of the latter, x 125. Fic. 2. Portion of a determinate
branch with three young apical hairs, x 600. Fic. 3. Distal ends of determi-
nate branches showing the terminal monosiphonous hairs, X 125. Fic. 4. Basal
region of a determinate branch with mature tetrasporangia, X 600.
195
: TAENIOMA
PAPENFUSS
1944]
icrographs by T. T. McCabe.
Photom
ENIOMA PERPUSILLUM
Ta
PLATE 23.
196 MADRONO [Vol. 7
sion of having arisen in a secund series from the dorsal side of the
prostrate axis.
Starting at the third, fourth or fifth segment from the base,
the erect secondary axes give rise to determinate branches alter-
nately at an interspace of two to nine segments (fig. 1). These
branches, in contrast to the indeterminate axes, are five cells in
width, except at the base, and are therefore flat (figs. 1, 2, 11,
12,14). Atan early stage in development, three hairs are formed
at the tip of each determinate branch (figs. 1-8, 11-14) where-
upon all further growth in length by segment formation ceases
in such a branch.
When a developing determinate branch is two or three seg-
ments long, it frequently initiates on its adaxial side an indetermi-
nate branch which in a manner similar to the secondary axes
forms determinate branches of a higher order (figs. 1, 7; see also
Thompson, 438, fig. 3). This process whereby a determinate or an
indeterminate branch of one order gives rise to a higher order of
branch of the opposite type is repeated over and over.
MetTHop or Brancuine. When a branch is to be initiated, the
apical cell, which ordinarily divides transversely, forming disk-
shaped segments, divides by an oblique wall, leaving a segment
which on one side is higher than the other (figs. 5, 7). Through
enlargement the higher side of this segment forms a protuberance
which is cut off. The cell so formed is the initial of a branch
(fig. 6). This method of branching has been termed exogenous
branching by Falkenberg (14). The original initial retains its
role as the apical cell of the axis. Both apical cells next divide
in the usual transverse fashion, giving rise to two monosiphonous
filaments whose segments eventually undergo division. In the
case of the erect axes, the branches formed are always alternate
and of the determinate type while the prostrate main axis forms
alternate and commonly erect, indeterminate branches. As stated
above, a determinate branch usually forms an indeterminate
branch on its adaxial side (figs. 1, 7). These branches are also
exogenous in origin.
Although indicated in Falkenberg’s (14) figure 23 on plate 15,
and described and figured by Thompson (48, p. 100, fig. 4), the
exogenous manner of branching in Taenioma has not received
recognition commensurate to its importance. This method of
branching is of common occurrence in the Rhodomelaceae, Dasy-
aceae, and Ceramiaceae but in rarely met with in the Deles-
seriaceae. In other members of this family, the branches are
formed in one or more of the following ways: (1) marginal as in
Membranoptera alata (Kylin, 24, p. 110, fig. 70), (2) from a corti-
eal cell on the midrib as in Apoglossum ruscifolium (Kylin, 24, pp.
85-86, fig. 55b), (3) endogenously as in Claudea and Vanvoorstia
(Papenfuss, 35), or (4) entirely adventitiously from cortical cells
1944 ] PAPENFUSS: TAENIOMA 197
as in Membranoptera alata (Phillips, 37) and Claudea multifida
(Papenfuss, 35).
The only other member of the Delesseriaceae which is known
to show exogenous branching is Caloglossa. Nageli (32) as long
ago as 1855 showed that in C. Leprieurii the branches are initiated
by segments which are formed as the result of an oblique division
of the apical cell of the parent branch (Nageli, op. cit., p. 71, pl. 8,
figs. 9-10). He also pointed out that the axis is monopodial and
that the branches are alternate in position. In manner of growth
and branching Caloglossa thus agrees with Taenioma. It should be
mentioned, however, that in Caloglossa branches also occur on the
midrib, but the exact method of their initiation is unknown. In
general, Nageli’s work has not been correctly interpreted, since
it is usually stated that in Caloglossa the lateral branches are mar-
ginal in origin.
STRUCTURE OF DETERMINATE BRANCHES. The determinate
branches are formed alternately at an interspace of two to nine
segments on the terete, erect, indeterminate branches (fig. 1).
The apical cell of the young branch by transverse division forms
segments in the manner characteristic of the Delesseriaceae.
When a determinate branch is but two or three segments long,
its apical cell frequently divides by an oblique wall, cutting off
_ a segment which gives rise to an indeterminate branch on the
adaxial side (fig. 7). Following the initiation of this branch, the
apical cell by transverse divisions forms from 12 to 20 segments.
There are then formed by alternate oblique divisions two seg-
ments (fig. 8) which are similar to those which initiate branches.
Each of these two segments gives rise to an apical cell (figs. 9,
10). At this stage, the branch apex is thus crowned with three
initials. The latter by transverse division give rise to the three
monosiphonous hairs which adorn the tip of the mature determi-
nate branch (figs. 10-14, 2-3).
As has been pointed out by Falkenberg (14) and Thompson
(43), the hairs terminate all growth in length by cell formation
and are responsible for the determinate character of these
branches. According to Bgrgesen (5) and Okamura (33), in-
tercalary divisions occur at the base of a hair, but the writer can
find no evidence of this. The cells of the hairs are formed in
acropetal succession by division of the apical cell. Those at the
base merely fail to elongate as much as the distal ones and give
the false impression of having been formed by intercalary
divisions.
While the hairs are in course of formation, the part of the
determinate branch posterior to them continues its development.
This is best described in connection with figures eight and eleven
to thirteen.
At the time that the hairs are initiated, the segments immedi-
ately below the two hair-forming ones are still undivided (fig. 8).
198 MADRONO [Vol. 7
The division of the segments to form pericentral cells is shown
in figure eleven. It will be seen from this figure that the second
and third segments below the large triangular cell, representing
the segment which initiated the first hair, have each formed a
lateral pericentral cell, while in the following three segments both
lateral pericentral cells have been formed. In the seventh and
subsequent segments down from the triangular cell, both the
transverse pericentral cells have been cut off, but only one in each
segment is indicated in the figure. At this stage the segments
thus consist of a central and four pericentral cells.
Mature determinate branches, in contrast to indeterminate
ones, are flat, except for the segments below and the three to five
immediately above the place of insertion of the daughter inde-
terminate branch, which remain cylindrical. This flat form is
produced as the seem of the formation of two fe ents cells by
each of the lateral pericentral cells.
The transverse pericentral cells do not divide and the branch
consequently remains only three cells in thickness at the midrib.
The method of formation of the flanking cells is shown in the
lower three segments in figure twelve. It will be seen that the
pericentral cells first divide by an oblique wall, more or less trans-
verse to the branch axis, to form a flanking cell toward the branch
apex. The next division is also in a vertical plane and cuts off the
second flanking cell from the portion of the pericentral cell below
the first-formed flanking cell. After the four flanking cells have
been cut off, no further divisions occur in the segments of a
determinate branch.
EXPLANATION OF THE FicurEsS. PLATE 24.
Pirate 24, Tarnioma. Fics, 5-16, Taenioma perpusillum. Fie. 5. Oblique
division of the apical cell of an indeterminate branch to form a branch-initiating
segment. Fic. 6. Division of a branch-initiating segment to form an apical
cell, the cell to the right. Fic. 7. An indeterminate axis with two alternate
young determinate branches, each of which has initiated an adaxial indetermi-
nate branch, while the apical cell of the axis has divided by an oblique wall in
preparation for the formation of a determinate branch to the right (the oldest
of the determinate branches, the one to the right, is in an early stage of hair-
formation, as seen in a side view of its apex). Fics. 8-10. Early stages in the
formation of the three terminal hairs, the central initial representing the origi-
nal apical cell of the branch. Fuies. 11-13. Young determinate branches show-
ing the further development of the hairs and the division of the segments of a
branch initial to form a central cell, four pericentral cells, and the two cells
which flank each of the lateral pericentral cells. Fic. 14. Optical longitudinal
section parallel to the surface of a determinate branch showing early stages in
the formation of tetrasporangia (cover cells were present in the two lower seg-
ments but are not indicated in the figure). Fic. 15. Optical vertical section
through a row of fertile lateral pericentral cells of a tetrasporangia-bearing
branch showing the formation of two cover cells by the stalk cell in the two
older segments. Fic. 16. Optical longitudinal section parallel to the surface
of a determinate branch with mature tetrasporangia. Fic. 17. Taenioma
macrourum, terminal portion of a young determinate branch with its two apical
hairs in the course of development (material from the Bahamas, Howe 5708).
(Figs. 5-15, 17, x 780. Fig. 16, x 340.)
199
PAPENFUSS: TAENIOMA
1944]
(t
> Cy :
ot Same crag
Ui CaSO
yy Saya ane ane
Puate 24, TAarniIoMa.,
200 MADRONO [Vol. 7
The division of the two hair-forming segments is similar to
that of other segments (figs. 12-14), except that in the distal one,
the lateral pericentral cell on the side opposite to the point of
attachment of the hair fails to form flanking cells (figs. 2, 14).
No intercalary cell divisions occur anywhere in the thallus of
Taenioma. Secondary pit connections are formed between neigh-
boring cells, except in the case of the central cells, which do not
become secondarily connected to each other nor to other cells.
Although the thallus of Taenioma is extremely simple, the con-
struction of the determinate branches is nevertheless in accord-
ance with the plan basic to all Delesseriaceae (cf. Kylin, 24, pp.
67—69, 81-82). Three categories of cells may thus be recognized:
(1) a cell-row of the first order, represented by the central cells;
(2) a cell-row of the second order, represented by the lateral
pericentral cells and the upper of the two flanking cells; and (3)
a cell-row of the third order, represented in the determinate
branches of T'aenioma by a single cell, the lower of the two-flank-
ing cells (figs. 2, 12-14).
It was stated earlier that Taenioma is to be included in the
Delesseriaceae rather than in the Rhodomelaceae on the basis of
the sequence in which the pericentral cells are formed.
That there is a fundamental difference in the order of for-
mation of the pericentral cells in the Rhodomelaceae and Deles-
seriaceae was first pointed out by Nageli (31) in 1847. From his
studies of Hypoglossum Woodwardii, species of Polysiphonia and
other representatives of both Delesseriaceae and Rhodomelaceae,
Nageli (29, 30, 31) established that in the Delesseriaceae the two
lateral pericentral cells are formed before the two transverse ones
(Nageli, 29, pl. 1, fig. 9). In the Rhodomelaceae, on the other
hand, the pericentral cells are formed in a progressive alternate
sequence, with reference to the first one, until the circle is com-
pleted. The last-formed pericentral cell thus occupies a position
diametrically opposite the first-formed (Nageli, 30, pl. 7, figs.
Seno
Since the time of Nageli, hundreds of species of both Deles-
seriaceae and Rhodomelaceae have been studied by a large num-
ber of workers but not a single exception to these plans has been
found. In the manner typical of the Delesseriaceae, the two
lateral pericentral cells are also the first to be found in T'aenioma
(fig. 11; see also Falkenberg, 14, p. 710, pl. 15, figs. 25-26; and
Thompson, 43, p. 101).
In the Delesseriaceae, with the exception of Taenioma and
Caloglossa, lateral organs are not initiated by a segment prior to
the formation of its pericentral cells. In the majority of Rho-
domelaceae, to the contrary, the lateral organs are initiated
before the segment cells have formed pericentral cells. Subse-
quent to Nageli, many workers have consequently been concerned
with problems relating to the position of the pericentral cells
1944] PAPENFUSS: TAENIOMA 201
with reference to the lateral organs and to one another. The
following are some of the more important questions which have
been studied: (1) the position of the first pericentral cell, (2)
whether the second pericentral cell is formed to the left or to the
right of the first, (3) whether or not the position of the second
pericentral cell is constant in a given species, and (4) whether
or not the second pericentral cell follows the spiral described by
the lateral organs. For a summary of knowledge concerning
these interesting questions, the reader is referred to the excellent
paper by Rosenberg (38, pp. 5-9, 31-32).
In this connection, it may be mentioned that in the Dasyaceae,
according to Rosenberg (38), the pericentral cells are formed in
a progressive left-hand sequence, with reference to the first one,
as seen from the outside, so that in the completed circle the
youngest lies to the right and next to the first-formed. The
Dasyaceae, Delesseriaceae and Rhodomelaceae thus differ very
sharply from one another in regard to the plan in which the peri-
central cells are formed. In the remaining family of the Cerami-
ales, the Ceramiaceae, typical pericentral cells are of course not
formed.
In addition to the difference in plan of pericentral cell for-
mation, the Delesseriaceae may also be distinguished from the
Rhodomelaceae, as was pointed out by Falkenberg (14, p. 713),
by the fact that in the Delesseriaceae the lateral pericentral cells
always form two flanking cells, which in the majority of forms
function as the mother-cells of initials which give rise to lateral
rows of cells. Each of these flanking cells has a primary pit-
connection with the pericentral cell and their combined length
equals that of the pericentral cell.
In the few examples of Rhodomelaceae where a similar forma-
tion of flanking cells occurs, only one flanking cell is formed by
each pericentral cell. This cell is cut off by a longitudinal division
and is therefore as long as the pericentral cell. It may later
divide transversely so that there are two (or more ?) flanking
cells. Only one of these cells, however, will be united to the
parent pericentral cell by a primary cytoplasmic connection.
In regard to the manner of flanking cell formation Taenioma
thus also shows itself to be a member of the Delesseriaceae.
REPRODUCTION
TETRASPORANGIA. Several of the earlier workers have recorded
tetrasporangia in Taenioma. The development of these organs
has, however, been studied only by Thompson (43). The
writer's observations are in agreement with those of this author.
The sporangia are cut off in acropetal succession from the
lateral pericentral cells in the expanded distal portion of the
determinate branches (fig. 14). Each pericentral cell forms only
one sporangium. It is separated by a concave wall from the
202 MADRONO [Vol. 7
terminal end of the pericentral cell and is in cytoplasmic com-
munication with this cell only (fig. 14).
After a sporangium has been initiated, the remaining portion
of the pericentral cell, at this stage referred to as the stalk cell,
divides by two walls parallel to the surfaces, forming a small
cover cell on each (fig. 15). In Taenioma the cells so formed never
enlarge or divide to form a protective layer over the sporangium,
but remain as rudimentary cover cells on either surface of the
stalk cell. The sporangia thus remain exposed on two sides dur-
ing their entire existence (figs. 4, 16). In the mature sporangia-
bearing branch, the central cells of the branch and the stalk cells
of the sporangia are considerably stretched by the enlarging spo-
rangia, and the marginal cells become connected by secondary
pit-connections to neighboring marginal and stalk cells (fig. 16).
The method of sporangium-formation in Taenioma is com-
parable to that in Vanvoorstia (Papenfuss, 35, figs. 67, 68). In
both genera the sporangia are formed by pericentral cells which
have not yet become corticated, and the cover cells are formed
after the sporangia have been initiated. The subsequent develop-
ment in the two genera is different, however, in that the primary
cover cells grow and divide in Vanvoorstia, forming a complete >
protective tissue over the sporangia, whereas in Taenioma the
cover cells serve no protective purpose and the sporangia remain
exposed.
By the method of sporangium-formation, Taenioma thus again
shows itself to be a member of the Delesseriaceae. In the Rho-
domelaceae, the pericentral cells destined to produce sporangia
first form two cover cells by longitudinal divisions and then divide
by a transverse wall to form the sporangium (cf. Kylin, 28, p.
173). The situation in the Rhodomelaceae is thus the reverse
from that in T’aenioma and Vanvoorstia where the sporangium is
the first cell to be cut off by a fertile pericentral cell.
In those Delesserieae and Rhodomelaceae in which the spo-
rangia are initiated by cells other than the pericentral cells this
' sequence is maintained (cf. Kylin, 28, pp. 170, 174-175). The
significance of this distinction between the Delesserieae and
Rhodomelaceae will appear later.
SPERMATANGIA AND CysTocarps. Sexual organs have been but
rarely observed in Taenioma. None were present in the writer's
material. Spermatangia were recorded for the first time by
Schmitz and Hauptfleisch (42, p. 415) in 1897, and for a second
time by Thompson (438) in 1910. From the descriptions of these
authors and from the figures of Thompson, it is clear that the
spermatangia are formed directly on the branches (the determi-
nate branches) as they are in all other Delesseriaceae. In the
Rhodomelaceae the spermatangia are usually formed on tricho-
blasts, although, as pointed out by Falkenberg (14), there are
exceptions to this rule. In a few genera such as Bostrychia and
1944] PAPENFUSS: TAENIOMA 203
Rhodomela, the spermatangia are formed on ordinary polysipho-
nous branches.
According to Thompson, the lateral pericentral cells become
divided by vertical walls prior to the formation of spermatangial
mother-cells. This statement requires verification. In. other
monostromatic Delesseriaceae the primary cells become divided
by two walls parallel to the surface, forming a layer of cells on
each surface. The superficial cells so formed then become divided
by vertical walls to form the spermatangial mother cells. The
spermatangia are formed on the surface of these cells. The
mature spermatangial sorus is thus composed of five cells in thick-
ness. Thompson describes and figures the mature sorus of
Taenioma as consisting of four cells in thickness.
Cystocarps have been observed with certainty only by Thomp-
son (43). The record by Collins and Hervey (11) of a mature
cystocarp in material from Bermuda (Phyc. Bor.-Amer. no. 1935)
is very doubtful, since duplicate material of this collection which
was examined by Howe (20) proved to be a rhodomelaceous alga.
The writer has found the same to be true of the duplicate material
in the Herbarium of the University of California.
Nothing is known of the development of the procarp in
Taenioma and very little concerning the cystocarp. According to
Thompson, the cystocarps occur on the terete branches. This
may seem unusual in view of the fact that the sporangia and
spermatangia occur on the flattened determinate branches. From
the position of the cystocarp Thompson concluded that this organ
is a modified branch. This assumption is logical. It seems likely,
however, that the cystocarps are initiated on young determinate
branches but that such branches do not complete their normal
growth and in the course of development of the cystocarp become
incorporated in its wall. Comparable conditions obtain in Claudea
and in Vanvoorstia (cf. Papenfuss, 35, pp. 26, 48, figs. 28, 42,
43,51).
From the general shape of the cystocarp and the gross struc-
ture of the gonimoblast, Thompson concluded that Taenioma be-
longed in the Rhodomelaceae. Through the researches of Falken-
berg and of Kylin and his students, we now know, however, that
there are no sharply defined and constant differences between the
Rhodomelaceae and Delesseriaceae with respect to the develop-
ment and structure of the cystocarp.
TAXONOMY
In addition to JT. perpusillum, which was described by J.
Agardh (1, 2) from material collected by Liebmann at St.
Augustin on the Pacific coast of Mexico, two other species, T.
macrourum Thuret (9) and T. Clevelandii Farlow (15), have been
credited to T'aenioma.
204 MADRONO [Vol. 7
Taenioma macrourum was established by Thuret in 1876 upon
material collected by Schousbee at Tangier, Morocco. In con-
sidering his plant as distinct from T. perpusillum, Thuret seems
to have been influenced more by the widely separated stations of
the two species than by morphological differences. Without
being aware of it, he did, however, describe and figure a feature
whereby T. macrourum can readily be distinguished from T. per-
pusillum as characterized by J. Agardh. According to Thuret the
tetrasporangia-bearing branches of T’.. macrourum terminate in two
hairs while those of T. perpusillum, according to J. Agardh, end in
three hairs. This character is of the first importance in sepa-
rating these two species as will be shown farther on.
Bornet (8) in 1892 reduced T. macrourum to the synonymy of
T. perpusillum, and this point of view has been accepted by the
majority of workers, including De Toni (12), Howe (in Thomp-
son, 43, p. 98, note), Bgrgesen (5, 6), Okamura (38) and others.
Falkenberg (14) and Schiffner (39, 40), on the other hand, have
retained 7’. macrourum as an independent species. Schiffner does
not give his reasons for so doing but Falkenberg, upon comparing
T’. perpusillum from the Pacific with material of JT. macrourum from
the Mediterranean, concluded that the former species lacked the
long monosiphonous terminal hairs characteristic of the latter.
The branches in T. perpusillum were also separated by an inter-
space of more segments, causing the plants to be less compact
than in T’. macrourum.
Although it is not possible to uphold the characters relied
upon by Falkenberg, the writer’s study nevertheless favors sepa-
ration of the two species. As mentioned earlier the plants at hand
of T. perpusillum came from Hawaii and South Africa. The obser-
vations on 7’. macrourum are based upon material from the Ba-
hamas (Howe 5708, as T. perpusillum—Herb. Univ. Calif. no.
207218) and from the Adriatic (Schiffner, Alg. mar. no. 860—
Herb. Univ. Calif. no. 495029).
From the account of the structure of the determinate branches
of T. perpusillum it is seen that the terminal hairs are formed in
avery precise manner. In this species the apical cell of a develop-
ing determinate branch forms a number of disk-shaped segments
by transverse divisions and then, by alternate oblique divisions,
two segments which on one side are higher than the other. These
two segments form initials which together with the original apical
cell of the branch give rise to the three terminal hairs character-
istic of this species. In 7. macrourum only one terminal segment
is formed by an oblique division of the apical cell of a determinate
branch and as a result the tips of these branches are crowned with
only two hairs (fig. 17; also see Thuret, 9, pl. 25, fig. 1; Falken-
berg, 14, pl. 15, figs. 21, 22; Bgrgesen, 5, fig. 337).
Since none of the plants studied by the writer showed part
of the branches ending in three hairs and the rest in two, there
1944] PAPENFUSS: TAENIOMA 205
is reason to believe that the number of hairs is constant in a spe-
cies: in JT. perpusillum the determinate branches invariably end in
three hairs while in JT. macrourum they end in two.
Another feature which may be of importance in separating
T. perpusillum from T. macrourum is that in the former the deter-
minate branches usually form an adaxial indeterminate branch
near their base, while in the latter such branches are rare. Conse-
quently in JT’. perpusillum the thallus is more profusely branched
than in JT’. macrourum. Before it could be conclusively stated,
however, that this latter character is of systematic value, it would
be necessary to examine more material of T’. macrourum than has
been available to the writer.
Bornet (8) when uniting T. perpusillum and T.. macrourum
listed as a synonym the South African species described by
Kiitzing (23) in 1863 as Polysiphonia nana, while Falkenberg (14)
gave the latter as a synonym of T'. macrourum. Kiitzing figured
P. nana as having some branches ending in a large inflated, apical
cell and others ending in two hairs. If P. nana were a species of
Taenioma, it would thus be logical to consider it as representative
of T'. macrourum; and since nana is the older specific name it would
have priority over macrourum. The writer has not had the oppor-
tunity of examining Kiitzing’s material nor has he been able to
identify with certainty a South African plant with Kiitzing’s spe-
cies. Judging from Kiitzing’s description and figure, however,
there is little reason for believing that Polysiphonia nana is repre-
sentative of the genus Taenioma. It seems more likely that the
species is a rhodomelaceous alga. The segments show four
parallel, vertically elongated, cells of the same length, which sug-
gest pericentral cells as seen in surface view.
The third species of Taenioma, T. Clevelandii, was described
by Farlow (15) in 1877 from material collected by Cleveland at
San Diego, California. Ina recent paper Hollenberg (19) records
having again found the species; and he also refers to specimens
in the Herbarium of the University of California. According to
Hollenberg T. Clevelandii was reduced to the synonymy of T. per-
pusillum by De Toni (13). The cited work of De Toni, however,
contains no statement to this effect.
From a study of the material in the Herbarium of the Univer-
sity of California and from the account of Hollenberg, it is clear
that T'. Clevelandii is not a species of T'aenioma but belongs to the
genus Platysiphonia Bgrgesen (7). Weber-van Bosse (45) already
in 1896 remarked upon the great similarity in structure between
TL. Clevelandit and Sarcomenia miniata, which is now Platysiphonia
miniata. T'. Clevelandiu differs from the other two species of
Taenioma in the following important features which it shares with
Platysiphonia: (1) The branches are endogenous in origin. (2)
No terete branches are formed, that is, the lateral pericentral cells
in all branches divide to form two flanking cells. (3) It does not
206 MADRONO [Vol. 7
have determinate branches which end in hairs. (4) In the tetra-
sporangia-bearing branches, the sporangia on one surface are
covered by a large cell and on the other by a rudimentary cover
cell. These branches thus have a dorsiventral organization. In
Taenioma both cover cells are rudimentary.
In habit 7. Clevelandii resembles Platysiphonia intermedia. The
relation of these species to each other can, however, only be
established from a detailed comparative study based on preserved
material. Pending such a study, it seems best to retain them as
distinct entities.
The nomenclature and the geographical distribution of the
species of T'aenioma may be summarized as follows:
TAENIOMA PERPUSILLUM (J. Ag.) J. Agardh, Sp. Alg. 2(3):
1257. 1863. Polysiphonia perpusilla J. Agardh, Ofvers. Kgl.
Svenska Vetensk.-Akad. Forhandl. 4: 16. 1848.
Geographical distribution. Paciric Ocean: St. Augustin, west
coast of Mexico (type locality, J. Agardh, loc. cit.) ; Hawaiian
Islands (Chamberlain, 10, p. 33; Papenfuss in the present article) ;
Japan (Okamura, 33, p. 26, in part) ; Tonga Islands (Grunow, 17,
p- 50); Molucca Islands (Heydrich, 18, p. 295). Inptan Ocean:
Dirk Hartog Island, Western Australia (Askenasy, 8, p. 54);
South Africa (Papenfuss in the present article). ATLANTIC
Ocean: Puerto Rico (Thompson, 43, p. 97).
TAENIOMA MACROURUM Thuret, in Bornet and Thuret, Notes
Algologiques, Fasc. 1: 69. 1876.
Geographical distribution. MerpirerraNreaNn Sea: Tangier, Mo-
rocco (type locality, Thuret, loc. crt.; Bornet, 8, p.. 297, as T-epene
pusillum) ; Balearic Islands (probably this species, De Toni, 13,
p-. 358, as T. perpusillum) ; Naples (Berthold, 4, p. 528; Falken-
berg, 14, p. 709) ; Dalmatia, Adriatic (Schiffner, 39, p. 158—Alg.
mar. no. 860!; 40, p. 302). AtTiantic Ocean: Canary Islands
(Bgrgesen, 6, p. 148, as T. perpusillum). CariBBEAN: Caracas,
Venezuela (Bornet, 8, p. 297, as T. perpusillum), Barbados (prob-
ably this species, Vickers, 44, p. 62, as T. perpusillum), Bahamas
(Thompson, 43, p. 97; Howe, 21, p. 564, as T. perpusillum, Howe
5708 !), Virgin Islands (Bérgesen, 5, p. 338, as T. perpusillum).
Paciric Ocran: Japan (Okamura, 33, p. 26, as T.. perpusillum, in
part).
Platysiphonia Clevelandii (Farlow) Papenfuss, comb. nov.
Taenioma Clevelandii Farlow, Proc. Am. Acad. Arts and Sci. 12:
236. 1877.
Geographical distribution. Catirornia: San Diego (type
locality, Farlow, loc. cit.); San Pedro (Herb. Univ. Calif. no.
96445, Mrs. H. D. Johnston, Jan. 27, 1900, with tetrasporangia ;
no. 315651, H. P. Johnson, Dec. 28, 1895); Carmel Bay (Herb.
1944] PAPENFUSS: TAENIOMA 207
Univ. Calif. no. 274026, N. L. Gardner, May, 1916, with tetraspo-
rangia) ; near Pacific Grove (Hollenberg, 19, p. 584),
DIscuUSssION
From the preceding account of the structure of the thallus
and the development of the tetrasporangia it is clear that Falken-
berg (14) was justified in removing T'aenioma from the Rho-
domelaceae and placing it in the Delesseriaceae. Within this
family the genus belongs in the sub-family Delesserieae; and from
a comparison with other genera, it is apparent that Taenioma is
the simplest of known Delesserieae. This is shown both by the
structure of the thallus and the exposed condition of the spo-
rangia. In the indeterminate branches, the lateral pericentral
cells do not function as the mother-cells of lateral initials, while
in the determinate branches, where they do act as such, the cell-
row of the second order is composed of but two cells and that of
the third order is represented by only one cell. As in other Deles-
serieae, the sporangial mother-cells first form a sporangium and
later the cover cells; but in contrast to other members of this
sub-family the cover cells in Taenioma are of a rudimentary char-
acter and at best can only be classed as incipient cover cells.
They have no protective value at any stage in the development of
the sporangia. These organs consequently always remain exposed
on two sides.
In his monograph on the Delesseriaceae, Kylin (25) divided
the family into a number of groups. To these was added the
Claudea-group by Papenfuss (35). Taenioma possesses certain of
the characters of the latter group but differs from it and all other
groups in one very important feature, namely, the exogenous
origin of the branches. This method of branch initiation is, how-
ever, also shown by Caloglossa, which Papenfuss included in the
Claudea-group. Caloglossa was placed in this group on the basis
of the structure of the blade and the formation of procarps on
only one surface of the blades; and since it has generally been
supposed that the branches in Caloglossa are marginal in origin,
the Claudea-group was circumscribed so as to include forms with
this method of branching. From the work of Nageli (32) it is
obvious, however, that the branches in Caloglossa are exogenous
in origin. It thus becomes necessary to remove this genus from
the Claudea-group, and to amend the group so as to exclude forms
showing marginal branching.
Since Taenioma and Caloglossa differ from all other known
Delesseriaceae by the exogenous origin of their branches, it seems
likely that these genera will prove to be closely related. This
question could be considered more profitably, however, after the
development of the procarp and the cystocarp had been studied
in both genera. It may be noted that structurally the thallus of
Caloglossa is more complex than that of Taenioma.
208 MADRONO [Vol. 7
A question which may here be considered is whether Taenioma
exhibits a primarily simple or a reduced condition. Since the
structure of the thallus, especially that of the indeterminate
branches, is comparable to that of many Rhodomelaceae, the most
highly evolved Florideae, it may be argued that the thallus of
Taenioma has been reduced. None the less, the exposed state of
the sporangia, coupled with the simplicity of the thallus, favors
the view that T'aenioma actually represents a primitive condition
within the Delesseriaceae, that is, a genus which has retained
certain relatively simple features characteristic of Ceramiales
lower than the Delesseriaceae and other characters which have
become elaborated or which have been eliminated in the higher
Delesseriaceae.
Within the Delesserieae the origin of tetrasporangia from
pericentral cells probably is a feature which in itself is indicative
of a primitive condition. But it is difficult to evaluate this char-
acter in T’aenioma and related genera, since the thallus is very
narrow and the only other cells which conceivably could form
sporangia would be the flanking cells. In none of the Deles-
seriaceae, however, do marginal cells form sporangia. In con-
trast to Taenioma, the sporangia in higher Delesserieae are formed
by cortical cells and not by pericentral or other primary cells.
In view of the primitive features exhibited by Taenioma, it
becomes of interest to know whether the genus throws light on
the relationships of the Delesseriaceae. Although it is not possi-
ble to point to any particular transitional type which could be
conceived as forming a link between Taenioma and any other
family of the Ceramiales, yet certain facts have come to be recog-
nized which have a bearing on the phylogeny of the order and
which necessitate a change in the accepted view regarding the
relative positions of two of the families, namely, the Delesseria-
ceae and the Dasyaceae.
In works on the algae, the Dasyaceae are usually placed above
the Delesseriaceae. From a review of the literature and the
results of the present study it is apparent, however, that the
Dasyaceae are phylogenetically lower than the Delesseriaceae.
Certain facts furthermore suggest that the Dasyaceae evolved
from Ceramiaceae-like ancestors and that the Delesseriaceae and
the Rhodomelaceae developed from Dasyaceae-like plants.
It is commonly agreed that the Ceramiaceae include the most
primitive Ceramiales. This view is supported by the following
facts: (1) The thallus in general is relatively simple, consisting
in lower forms of branched monosiphonous filaments but be-
coming polysiphonous or corticated in higher forms. Typical
pericentral cells, that is, cells which from the beginning are as
long as the central cells, are, however, not formed. (2) In the
majority of forms the sporangia and the gonimoblasts are naked.
1944] PAPENFUSS: TAENIOMA 209
(3) The auxiliary cell in lower forms is supplied with a diploid
nucleus via an intermediary connecting cell.
Although Kylin (24, 26, 27) has on several occasions ex-
pressed the view that the Delesseriaceae and the Rhodomelaceae
represent two parallel lines of development, with the Rhodo-
melaceae occupying a somewhat higher level than the Deles-
seriaceae, he has, nevertheless, always placed the Dasyaceae
above the Delesseriaceae. Phycologists in general have adopted
the arrangement of Kylin.
From the work of Rosenberg (38) and others on the Dasyaceae
several facts may be cited which indicate that this family is less
advanced than the Delesseriaceae and more closely related to the
Ceramiaceae than are the Delesseriaceae. The most significant
single fact supporting this view is that in the Dasyaceae the
auxiliary cell receives a diploid nucleus from the fertilized carpo-
gonium via a connecting cell. This feature is characteristic of
the lower Ceramiaceae, but has been lost in the Delesseriaceae.
Other primitive features of the Dasyaceae are: (1) The spore-
lings remain monosiphonous for a long time (Killian, 22) as con-
trasted with those of the Delesseriaceae and Rhodomelaceae in
which pericentral cells are formed at an early stage in develop-
ment. (2) The pericentral cells are of a rudimentary character
in certain genera. (3) The sporangia remain partially exposed.
It may also be added that Falkenberg (14) considered the
sympodial method of branching of the Dasyaceae as a character
which is primitive in comparison with the monopodial branching
characteristic of the Rhodomelaceae. Whether the sympodial
habit actually is a primitive feature or whether it is a derived con-
dition which has become established in the Dasyaceae would,
however, be difficult to decide. It would seem that in exogenous
branching a reversal from the monopodial to the sympodial habit
or vice versa is one which would not entail profound change.
Furthermore, in the Ceramiaceae, which are more primitive than
the Dasyaceae, some forms show monopodial and others sym-
podial branching (Feldmann-Mazoyer, 16, p. 123).
According to Kylin (28, p. 134) the sporelings of the Dasya-
ceae show monopodial branching. If this were correct it would
indicate that the sympodial habit of the older thallus was acquired
in the course of evolution of the family. However, from the
observations of Killian (22) on the sporelings of Dasya arbuscula,
upon which work Kylin’s statement is based, it seems evident that
sympodial branching is also characteristic of the sporelings of the
Dasyaceae, or at least of those of D. arbuscula. This is the inter-
pretation which Oltmanns (34, p. 322) also gives of Killian’s
work.
As to the relative position of the Delesseriaceae and the Rho-
domelaceae the available facts favor the conclusions of Kylin
(24, 26, 27) that these families represent two parallel lines of
210 MADRONO [Vol. 7
evolution, with the Rhodomelaceae occupying a somewhat higher
level than the Delesseriaceae. In support of this view Kylin (26,
27) cites three facts: (1) The Rhodomelaceae are in a more active
state of speciation. (2) In the Rhodomelaceae the pericarp is
initiated prior to fertilization whereas in the Delesseriaceae it is
formed after fertilization. (3) In the Rhodomelaceae, the cover
cell of the procarp is a specialized cell which degenerates if
fertilization fails to occur while in the Delesseriaceae it is com-
parable to an ordinary vegetative cell (Kylin, 24, p. 102; 28,
p. 286).
It may be pointed out, however, that the two latter distinctions
only hold when the Rhodomelaceae are contrasted with the
Delesserieae. In the higher Delesseriaceae, that is, in the Nito-
phylleae, the cover cell of a procarp behaves like that of the
Rhodomelaceae (Kylin, 28, p. 286); and in certain members of
this sub-family (e.g., Phycodrys sinuosa Kylin, 24, figs. 46h, 47d—g;
Acrosorium acrospermum Papenfuss, 36, fig. 16) the pericarp is
initiated prior to fertilization.
To the points cited by Kylin may now be added a fourth which
lends support to the view that the Rhodomelaceae are phylo-
genetically higher than the Delesseriaceae. In the Rhodomelaceae
the sporangial mother-cells cut off the cover cells before the spo-
rangia are initiated whereas in the Delesserieae the cover cells
are formed after a sporangium has been initiated. In the latter
group the young sporangia are thus exposed while in the Rho-
domelaceae they are protected. In Taenioma, the most primitive
of known Delesseriaceae, the cover cells are rudimentary and
serve no protective purpose at any stage in the development of
the sporangia, while in the related Platysiphonia the cover cells are
well developed on one surface of the fertile branches and rudi-
mentary on the other. Thus Platysiphonia may be said to form
a link between Taenioma and Vanvoorstia in which latter genus the
cover cells are well developed on both surfaces of the fertile
blades (Papenfuss, 35).
As to the probable ancestors of the Delesseriaceae, there is
reason to believe that they may have evolved as an off-shoot from
the stock which gave rise to present-day Dasyaceae. Three facts
especially may be cited in support of this view: (1) In the Deles-
serieae as in the Dasyaceae the sporangial mother-cells first form
a sporangium and then the cover cells. (2) In both the Deles-
serieae and the Dasyaceae the cover cell of the procarp is com-
parable to a vegetative cell and functions as such if fertilization
fails to occur. (8) In the Delesseriaceae in general, but more
particularly in the Delesserieae, as well as in the Dasyaceae the
pericarp is not initiated until after fertilization.
It is of interest to note that in these three features the Rho-
domelaceae have advanced beyond both the Dasyaceae and the
Delesseriaceae. It may be mentioned, however, that the higher
1944] PAPENFUSS: TAENIOMA 211
Delesseriaceae, for example, certain members of the Nitophylleae,
not only share some of these advances with the Rhodomelaceae
but have attained a degree of simplicity which surpasses that
shown by any of the Rhodomelaceae. It is thus found that: (1)
In the Nitophylleae growth of the mature thallus is usually mar-
ginal whereas in the Rhodomelaceae it is primarily apical. (2)
In the Nitophylleae intercalary cell divisions are of common
occurrence while such divisions, to the writer’s knowledge, are
unknown in the Rhodomelaceae. (3) In the Nitophylleae the
reproductive organs are formed in parts away from the central
axis of the thallus whereas in the Rhodomelaceae they generally
are localized on the axis, that is, the pericentral cells. It is of
interest to note that an advance in the same direction is shown
by some of the higher Rhodomelaceae, such as Laurencia and
Ricardia. In these genera the sporangia are no longer formed by
pericentral cells but by cells farther away from the central cells
(cf. Kylin, 26, p. 101).
Finally, in regard to the probable prototypes of the Rhodo-
melaceae, it seems likely that this family also may have evolved
from Dasyaceae-like algae, although at a time subsequent to the
separation of the Delesseriaceae. It is well known that the
Rhodomelaceae have a number of characters in common with the
Dasyaceae. The similarity in the habit of the thalli, the forma-
tion of the sporangia from pericentral cells, and the correspond-
ing position of the spermatangia are some of the more important
points of agreement which may be mentioned. It is of interest
to note in this connection that in Heterosiphonia coccinea (Dasya-
ceae) the pericentral cells are formed in the manner characteristic
of the Rhodomelaceae except in the fertile segments of female
plants where they are formed in typical dasyacean fashion
(Rosenberg, 38). In this particular species the sporangia and
their cover cells are also formed in the sequence characteristic of
the Rhodomelaceae. This member of the Dasyaceae thus shows
certain morphological features which have become established in
present-day Rhodomelaceae. It may further be mentioned that
Falkenberg (14) considered the Dasyaceae, as now recognized,
as a sub-family in the Rhodomelaceae and regarded them as the
ancestors of all other Rhodomelaceae.
A schematic representation of the probable interrelationships
and lines of development of the families of the Ceramiales is given
in text figure 1.
SUMMARY
The results of this study show that Taenioma does not belong
to the Rhodomelaceae, in which it has been placed by certain
writers, but to the Delesseriaceae. This is shown by the plan of
pericentral cell formation, the manner of division of the lateral
212 MADRONO 7 [Vol. 7
pericentral cells, and the method by which the sporangia are
formed.
From the simple structure of the thallus and the exposed con-
dition of the sporangia, it is concluded moreover that Taenioma is
the simplest of known genera of the Delesseriaceae.
The branches of Taenioma are exogenous in origin. Although
characteristic of the other families of the Ceramiales, this method
of branching is rarely encountered in the Delesseriaceae and in
addition to T’aenioma is known to occur in Caloglossa only.
The sporangia are initiated by the lateral pericentral cells and
are formed before the mother-cells have cut off cover cells. The
latter feature is recognized as one whereby the Delesseriaceae
RHODOMELACEAE
DELESSERIACEAE
DASYACEAE
CERAMIACEAE
Fic. 1. Diagrammatic representation of the probable interrelationships of
the families of the Ceramiales.
may be distinguished from the Rhodomelaceae. In the Rho-
domelaceae the mother-cells first form cover cells and then a
sporangium.
Taenioma perpusillum and T. macrourum are considered as sepa-
rate species. The primary distinguishing feature lies in the fact
that the determinate branches of the former form three and those
of the latter two terminal hairs. These hairs are initiated in a
precise manner by the apical cell and do not vary in number.
Taenioma Clevelandii, the only other species which has been
credited to T'aenioma, is transferred to the genus Platysiphonia with
which it is shown to share various morphological features.
The phylogeny of the Ceramiales is discussed. Evidence is
produced to show that the Delesseriaceae are more highly evolved
than the Dasyaceae. Additional evidence is furnished in support
of the view of Kylin that the Rhodomelaceae are the most highly
evolved Ceramiales. It is concluded that the Dasyaceae evolved
1944] PAPENFUSS: TAENIOMA 213
from Ceramiaceae-like and the Delesseriaceae and Rhodomelaceae
from Dasyaceae-like ancestors.
Grateful acknowledgment is made to Professor Lee Bonar for
permission to work in the Department of Botany of the University
of California and to Professor Herbert Mason for kindly placing
at the writer’s disposal during a period of two years the excellent
facilities of the Herbarium of this University. Thanks are due
to Dr. M. A. Pocock for the South African material of T'aenioma
and to Mr. T. T. McCabe for his reading of the manuscript and
for kindly preparing the photomicrographs reproduced in this
paper. The study was made during the tenure of a Fellowship
granted by the Carnegie Corporation of New York.
After the study had been completed, it was learned from Dr.
C. K. Tseng that he had also been working on Taenioma, and had
arrived at the same conclusions as the writer with respect to the
systematic position of the genus and the taxonomy of the species.
Department of Botany,
University of California, Berkeley,
March, 1944.
LITERATURE CITED
1. AGarpu, J. G. Nya alger fran Mexico. Ofvers. Kgl. Svenska Vetensk.-
Akad. Forhandl. 4: 5-17... 1848.
2. ————_—_——.. Species, genera et ordines algarum 2(3). Lund. 1863.
3. AsSKENASY, EK. Algen. Forschungsreise S. M. S. Gazelle, Theil 4, Botanik:
1-58. Berlin. 1888.
4. Bertuoip, G. Uber die Vertheilung der Algen im Golf von Neapel nebst
einem Verzeichnis der bisher daselbst beobachteten Arten. Mitt. Zool.
Stat. Neapel 3: 393-536. 1882.
5. Bgrorsen, F. The marine algae of the Danish West Indies, 3, Rho-
dophyceae (5). Dansk Bot. Arkiv 3(le): 305-368. 1919.
6. ————————.. Marine algae from the Canary Islands, 3, Rhodophyceae,
pt. 3, Ceramiales. Kgl. Danske Vidensk. Selsk., Biol. Medd. 9(1):
1-159. 1930. :
7, ———————. Sur Platysiphonia nov. gen. et sur les organes males et
femelles du Platysiphonia miniata (Ag.) nov. comb. [Sarcomenia
miniata (Ag.) J. Ag.]. Recueil de Travaux Cryptogamiques dédiés 4
Louis Mangin. 9 pp. 1931.
8. Bornet, E. Les Algues de P. K. A. Schousboe, récoltées au Maroc & dans
la Méditerranée de 1815 a 1829. Mém. Soc. Natl. Sci. Nat. et Math.
Cherbourg 28: 165-376. 1892.
9. Bornet, E., and G. TuHuret. Notes algologiques, Fasc. 1. Paris. 1876.
10. CHAMBERLAIN, J. E. Algae of the Hawaiian Islands. Hawaiian Almanac
and Annual for 1881: 32-33. 1880.
11. Coxuins, F. S., and A. B. Hervey. The algae of Bermuda. Proc. Am.
Acad. Arts and Sci. 53: 1-195. 1917.
12. De Toni, J. B. Sylloge Algarum 4(2). Patavii. 1900.
13. ————————.. Sylloge Algarum 6. Patavii. 1924.
14. FaALKENBERG, P. Die Rhodomelaceen des Golfes von Neapel und der
angrenzenden Meeres-abschnitte. Fauna und Flora des Golfes von
Neapel, Monogr. 26. 1901.
15. Farrow, W. G. On some algae new to the United States. Proc. Am.
Acad. Arts and Sci. 12: 235-245. 1877.
16. FetpmMann-Mazoyer, G. Recherches sur les Céramicées de la Méditerranée
Occidentale. Alger. 1940.
214
1%,
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
Si.
38.
39.
40.
Al.
42.
43.
4A.
45.
MADRONO [Vol. 7
Grunow, A. Algen der Fidschi-, Tonga- und Samoa-Inseln. Jour. Mus.
Godeffroy (Hamburg) 3(6): 23-50. 1873-74.
Heypricu, F. Beitréage zur Kenntnis der Algenflora von Ost-Asien
besonders der Insel Formosa, Molukken- und Liu-kiu-Inseln. Hedwigia
33: 267-306. 1894.
Ho.iensBerG, G. J. Phycological notes—I. Bull. Torrey Bot. Club 69: 528-
538. 1942.
Hower, M. A. Algae, in Britton, Flora of Bermuda. New York. 1918.
—____———.. Algae, in Britton and Millspaugh, The Bahama Flora. New
York. 1920.
Kivuian, K. Uber die Entwicklung einiger Florideen. Zeitschr. Bot. 6:
209-278. 1914.
Kurzne, F. T. Tabulae Phycologicae 13. Nordhausen. 1863.
Kyun, H. Studien iiber die Entwicklungsgeschichte der Florideen. Kegl.
Svenska Vetensk.-Akad. Handl. 63(11). 1923.
. Studien tiber die Delesseriaceen. Lunds Univ. Arsskr. N. F.
Avd. 2, 20(6). 1924.
Entwicklungsgeschichtliche Florideenstudien. Ibid. 24(4).
1928.
Uber den Aufbau der Prokarpien bei den Rhodomelaceen
nebst einigen Worten iiber Odonthalia dentata. Kgl. Fysiogr. Sallsk.
Lund Forhandl. 4(9): 1-22. 1934.
—_—_—_—_————. Anatomie der Rhodophyceen, in Linsbauer, Handbuch der
Pflanzenanatomie, Abt. II, Bd. 6(2): Algen (B, g). Berlin. 1937.
N AGELT, C. “Wachsthumsgeschichte von Delesseria Hypoglossum. Schleiden
und Niageli’s Zeitschr. wissensch. Bot. 1(2): 121-137. 1845.
Polysiphonia. Jbid. 1(3 & 4): 207-237. 1846.
Die neuern Algensysteme. Ziirich. 1847.
Wachsthumsgeschichte von Hypoglossum Leprieurii (Mont.)
Kg. Pflanzenphysiol. Untersuch. von Nigeli und Cramer, Heft 1:
69-75. 1855.
Oxamura, K. Icones of Japanese Algae 6(3). Tokyo. Published by the
author. 1930.
Ottmanns, F. Morphologie und Biologie der Algen. Ed. 2, 2. Jena.
1922.
PapenrFuss, G. F. The structure and reproduction of Claudea multifida,
Vanvoorstia spectabilis, and Vanvoorstia coccinea. Symb. Bot. Up-
salienses 2(4): 1-66. 1937.
. The development of the reproductive organs in Acrosorium
acrospermum. Bot. Notiser 1939: 11-20. 1939.
Puitiies, R. W. The development of the cystocarp in Rhodymeniales: II.
Delesseriaceae. Ann. Botany 12: 173-202. 1898.
RosenBeERG, T. Studien tiber Rhodomelaceen und Dasyaceen. Dissertation.
Lund. 1933.
ScHIFFNER, V. Neue und bemerkenswerte Meeresalgen. Hedwigia 71:
139-205. 1931.
Meeresalgen aus Siid-Dalmatien, gesammelt von Franz
Berger. Oster. Bot. Zeitschr. 82: 283-304. 1933.
Scumitz, F. Systematische Ubersicht der bisher bekannten Gattungen der
Florideen. Flora 72: 435-456. 1889.
Scumirz, F., and P. HAauptrieiscH. Rhodophyceae, in Engler and Prantl,
Die natiirlichen Pflanzenfamilien 1(2). Leipzig. 1897.
Tuomprson, EK. I. The morphology of Taenioma. Bull. Torrey Bot. Club
37: 97-106. 1910.
Vickers, A. Liste des algues marines de la Barbade. Ann. Sci. Nat., Sér.
9, Bot. 1: 45-66. 1905.
WeEBER-VAN Bosse, A. Notes on Sarcomenia miniata Ag. Jour. Bot. 34:
281-285. 1896.
1944] TSENG: TAENIOMA O15
NOTES ON THE ALGAL GENUS TAENIOMA!
C. K. Tsene
The algal genus T'aenioma (Delesseriaceae, Rhodophyceae)
was founded by J. G. Agardh (2, p. 1256) on Polysiphonia per-
pusilla which he (1, p. 16) himself previously had described from
material collected by Liebmann at St. Augustin on the Pacific
coast of Mexico. The then monotypic genus was characterized.
by (1) a filiform, dichotomously branched frond with poly-
siphonous, articulate, noncorticate filaments; (2) the presence of
marginal cells half the length of the segments on the flattened
portion of the frond; and (3) dilated, vittiform stichidia bearing
tetrasporangia (then known as sphaerospores) in a double series.
The genus was placed by its author, together with Sarcomenia, in
the tribe Sarcomeniae which was then placed in the family
Rhodomelaceae but later more properly removed by Schmitz (19)
to the Delesseriaceae.
Thuret (7, p. 69, pl. 25) added another species to the genus,
namely T'aenioma macrourum, citing as a synonym Hutchinsia macro-
ura Schousboe in herb. The type material came from Tangier,
Morocco, in the Mediterranean. Not having seen J. Agardh’s
specimen of T'. perpusillum Thuret, in describing his species, ex-
pressed some doubt as to the specific difference between the Medi-
terranean form and the one from the Pacific. In keeping the two
separate, he was probably influenced by the widely separated
regions from which the two plants were collected, some apparent
differences in size and color, and especially the presence in the
Mediterranean plant of two apical hairs which were presumably
absent in the alga from the Pacific. It should be noted, however,
that J. Agardh, op. cit., did mention the presence of apical hairs
in his plant: “Stichidium .. . apice saepe in fila minuta 3...
excurrens.’ Examination of Agardh’s type specimen by Howe
(zn ‘Thompson, 21, p. 98) has confirmed this. The writer has also
examined a fragment of the original specimen deposited in the
herbarium of the New York Botanical Garden and is in perfect
agreement with Howe.
Several years after the publication of Taenioma macrourum
Thuret, Bornet (6, p. 297) examined a specimen of T. perpusillum
J. Ag. and came to the conclusion that the Pacific and the Medi-
terranean plants are not separable specifically. Later, Heydrich
(12, p. 295) and De-Toni (9, p. 732) adopted Bornet’s view.
Schmitz and Hauptfleisch (20, p. 415), however, retained both
species. In his classical work on the Rhodomelaceae, Falkenberg
(10, p. 709, pl. 15, fig. 21-29), devoted some space to the mor-
phology of the Mediterranean plant, which he separated from the
Pacific alga chiefly on the basis of its having long monosiphonous
1 Contribution from the Scripps Institution of Oceanography, New Series,
No. 230.
216 MADRONO [Vol. 7
apical filaments on the flattened “leaves,” presumably absent in
IT. perpusillum J, Agardh. This, as already pointed out above,
is not correct. Falkenberg also mentioned that the two species
differed in the nature of the branches. He remarked that in the
Mediterranean plant the cylindrical segments of the side branches
and also those of the axis between two branches were so short and
compact that it was easily distinguishable from the Pacific plant,
which was much slenderer and more loosely branched. While
it is true that the Pacific alga has much longer side branches, the
shortness and compactness of the segments are similar in both
cases. It is no wonder, therefore, that later authors (cf. Collins
and Hervey, 8, p. 117; Bgrgesen, 4, p. 841; and Okamura, 17,
p. 26) in trying to differentiate the two species on this basis, could
not see the desirability of keeping them apart.
Thompson (21) published a detailed study of the morphology
of Taenioma, using two different sets of materials collected by
Howe at Porto Rico (Puerto Rico) and at West Caicos in the
Bahamas. She was probably the first to note some of the impor-
tant differences between T'. perpusillum and T. macrourum although
she discussed both forms under the latter name. Such differences
between these two collections as the branching and the number
of apical hairs and segments in the flattened shoots have been
correctly emphasized. In an editorial note in Thompson’s paper,
Howe justified that author's identification of the Bahamian plant
with the Mediterranean species. He questioned, however, “the
identity of this JT. macrourum with the previously described Taeni-
oma perpusillum of J. Agardh ... ,’ remarked that his Puerto
Rican plant and the type of J. Agardh’s species from Mexico are
“essentially the same except that the terminal hairs are much
longer and more luxuriant in the Porto Rican plant,’ and con-
cluded that he was “inclined to agree with Bornet . .. with
Heydrich . .. and with De-Toni . . . in considering Taenioma
macrourum (Schousb.) Thur. a synonym of Taenioma perpusillum
J. Ag.”
Howe seemed to have regarded the differences of these two
sets of specimens, which he considered to belong to the same spe-
cies, as a matter of ecological influence. He made the remark that
the Bahamian plants “were found growing in an inland pond or
lake, having evidently a subterranean communication with the
sea—a place where several marine algae of recognizable species
were more or less abnormal and peculiar,’ and the Puerto Rican
plants “were growing where they were well exposed to the surge
of the open sea.” It should be noted, however, that Bgrgesen
(4, p. 841), whose plant was evidently of the same species as
Howe’s from the Bahamas, found it “in an open place upon reefs
of calcareous algae, etc.”
To date, phycologists seem to have regarded this problem of
the Taenioma perpusillum-macrourum complex as_ satisfactorily
1944] TSENG: TAENIOMA 217
settled, have generally adopted the view of Bornet and his fol-
lowers, and have conventionally put T. macrourum as a synonym
of T. perpusillum whenever that species is reported.
The third species of T'aenioma was described by Farlow (11,
p. 236) on material from San Diego, California, and was named
Taenioma Clevelandiu. As described, it “has scattered stichidia, is
four inches high, and has a striking resemblance to Griffithsia
tenuis Harv.,” and its “‘stichidial branches terminate in a more or
less acute apex instead of two hairs.” It was also mentioned that
“in the lower part of the frond, the angles between the primary
cells are filled with a small but irregular number of secondary
cells.”” All these peculiarities are certainly very different from
the characteristics of the Taenioma perpusillum-macrourum complex
and the writer has long doubted its being a member of this group.
He has studied some specimens collected at Moss Beach, Pacific
Grove, California (Hollenberg 3228) and is fully convinced that it
should be separated from Taenioma.
Recently Hollenberg (18, p. 534) has pointed out several
additional differences between Taenioma Clevelandi Farl. and T.
perpusillum J. Ag., although he still preferred to keep the Cali-
fornian plant in the same genus. Dr. Papenfuss who has made a
critical study of Taenioma Clevelandii has agreed with the writer
that it does not properly belong to Taenioma. He has further
decided that it should be transferred to the genus Platysiphonia
Bgrg., because (1) its branches are endogenous in origin; (2) it
does not form terete branches; (3) it does not have determinate
branches ending in hairs; and (4) its tetrasporangial branches
have a dorsiventral organization, since the stalk cell of a tetra-
sporangium forms a large cover cell on one surface and a rudi-
mentary one on the other. These characteristics are so funda-
mentally different from those of T'aenioma, that there should no
longer be doubt among phycologists that the plant from San
Diego should be removed from the genus Taenioma.
Some years ago, the writer gathered from Hong Kong a col-
lection of a Taenioma rich in tetrasporangia. At first, he was
inclined to follow others and consider T’. macrourum Thur. a syno-
nym of T'. perpusillum J. Ag. The more the specimens and litera-
ture were studied, however, the more hesitant he was to do so.
After a thorough study of his collection, which has been preserved
in excellent condition for microscopic examination, and the exten-
sive collections of Howe, including some fragments of the type
specimen of 7’. perpusillum J. Ag., deposited in the Herbarium of
the New York Botanical Garden, the conclusion was finally
arrived at, that the differences, especially those reported by
Thompson and Howe, do exist, and are very constant. Other
differences have also been found. The fact that the same form
has been reported to occur in sheltered and exposed places and
both forms in more or less similar situations naturally eliminates
218 MADRONC [Vol.7
the possibility of their being ecological forms, at least, so far as
the factor of the relative exposure to surf is concerned. Further-
more, the geographical distribution of T. perpusillum and T.
macrourum gives evidence of their distinctness. When funda-
mental differences between these two forms are constant and
without intermediates, and
when these cannot be
traced to ecological or geo-
graphical influences, then
(Hue:
Ti
a Rol 0
Hook there is no alternative
Ach than to accept the two
FONGe forms as separate, inde-
f\ AGHOA pendent species.
I Fe The most important
=a HOTOH difference between the two
IS HOVOH lies in the number of
AZ nOlQH apical hairs. The Puerto
BZ HO Ou Rican and Hong Kong
Z USS specimens always have
ZL) Bb three apical hairs on the
=6 JO flattened branchlets and
= OO are referable to T. per-
000 pusillum J. Ag. There are
many determinate branch-
JOO, lets which at a glance seem
to have only two _ hairs.
Fic. 1. Taenioma macrourum Thuret: a, A eareful study, however,
apex of a branchlet showing the two apical
hairs, X 560; b, upper part of a branch show- reveals the fact that the
ing a stichidium, x 190.
a
<
\
oldest hair in the group of
three has already dropped
off, and the two younger ones are left behind (pl. 25, fig. 5). The
Bahamian, Bermudian, and Mediterranean specimens as well as
the Japanese plants described and illustrated by Okamura (17, p.
26, pl. 244, fig. 17-19, pl. 245, fig. 5-9, on the contrary, have only
two hairs and undoubtedly belong to 7’. macrourum (text fig. la).
To be sure, Okamura (17, p. 27, pl. 265, fig. 8) mentioned and
figured a branchlet with three apical hairs. Whether this was an
abnormal form of the normally two-haired plant, or whether
Okamura had both species, can be settled only by examination of
his specimens which had come from at least two different sources.
The presence of two or of three such apical hairs is neither
accidental nor irregular. It is rather a matter of fundamental
difference traceable to the behavior of the apical cells of the
determinate branchlets. In TJ. macrourum, the apical cell, after
having formed the more or less definite number of segments, gives
rise by an oblique division to a hair-initiating cell. Later, the
apical cell itself assumes the role of hair formation, thus result-
ing in two hairs (text fig. la). In the case of T. perpusillum, the
apical cell by alternate oblique divisions forms two hair-initials.
1944] TSENG: TAENIOMA 219