NOILALILSNI NVINOSHLINS S3IYVYUGIT LIBRARIES

2

SMITHSONIAN

gi

Fam
a
es
as
~
a
~
YY
<i

Saidvdai
INSTITUTION
SatyVvYy

Saluy

NOILNLILSNI NVINOSHLINS S3IYV

-IBRARIES SMITHSONIAN INSTITUTION

NVINOSHLIWS
NVINOSHLIWS
NVINOSHLINS

VOILOLILSNI NVINOSHLIWS SIIMVNalT LIBRARIES INSTITUTIC

WS

IBRARIES SMITHSONIAN

SMITHSONIAN

NOILNLILSNI

NOILALILSNI
NOILNLILSNI

INSTITUTION NOILALILSNI NVINOSHLIWS saiuvug
7 - |

INSTITUTION
INSTITUTION

INSTITUTION

NOLLALILSNI NVINOSHLINS S31YVYE!IT LIBRARIES SMITHSONIAN

NVINOSHLINS Sa1uWvugi LIBRARIES SMITHSONIAN
NVINOSHLINS S3SIYVYHYSIT LIBRARIES SMITHSONIAN

<m Za a a

me oo <

Or ge nt fey’ 2 —

S yy AI Z

= Wye 2 SX Us
Z = a = ‘ = |
-IBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31uvY q

o > 3 or 2p)

= ah 4 + Sh — ” XY ard

x. (9 yy = a = VS =

<x vy YH a. < a 2. AS. ‘* <

a 2 : ‘oes

x | rs) = ra if mee

vel Ps oe 2 SS
VOILALILSNI NVINOSHLIIS ~°4 IYVvVud ne LIBRARI ES_ SMITHSONIAN INS TERUG
oe Oo — o) Lo ze

co = ee w

Fs) Wo = = puss ,

x» = WANE = > rat “>

7 SNe xy = ae)

2 NW E 2 =e 2

m PASS S as uy)? =

w = ow : = wn

IBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS Saluvud

NVINOSHLIWS
NVINOSHLIWS
NVINOSHLIWS

JOLLALILSNI S31YVYGIT LIBRARIES

NVINOSHLIAS SMITHSONIAN INSTITUTIC
*

G:

_IBRARIES SMITHSONIAN
BRARIES SMITHSONIAN

NOILALILSNI

NOILOLILSNI

SY
JOILALILSNI

L

SJIMYVYUGIT LIBRARIES SMITHS

mon

LALILSNI_NWINOSHLINS _ — =
o S - S Gf
2 = 2 5
> = > pas
a = sie -

RARIES SMITHSONIAN INSTITUTION ~NOILNLILSNI
wm ee * io) Pus
= < =< <
a - = ae
= BS oO aL: f- ©
ep) 9) _ &O fama 2
D, me * Oo a
= = & 22
> = rte >’ = !

8 Pa w” + te rd ~”

LNLILSNI NVINOSHLIWS SAiYvVaaiIT LIBRARIES SMITHS NIAN INSTITUTION
= of = wh Z
” = ” oO
= oc a oe , a
Cc : a C oc c
oe) ~S _ ) ea O
rn er 2 J =

-RARIES SMITHSONIAN INSTITUTION NOILNLILSNI SS1uvydl

Zz ~ rat Zz
fo) oo ° - °
=

" LY, tp = S = b-
- QYYYy = b ch

ls ‘Yl m ne m a
= ap) z a) .

ILALILSNI NVINOSHLIWS (S23 luVvuad ryt BRARI ES SMITHSONIAN INSTITUTION
& z= = \S = ze &.
i= = Ze YN =. z \:

O t. DO Vee & ak oO
u Dy ZR 2 2
= Z Z s 2 =
= . >
a a a » 3 3 |
SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS SJ3IYVUEI
Ww = == ae ”
‘a ca = = - ox
<4 <
a = oc = oc
co 5 O =
= 2 a ra _J

ILLALILSNI NVINOSHLINS S3I1YVedl)

co. c z 7

oo : wa w — 10]

pe X. \ 5 x = a

Y~ SS = ve ;

= ~§AAE 2 = a

oa SNS on an D a

” A w se nn :
BRARIES SMITHSONIAN INSTITUTION NVINOSHLINS S3SIuYVvudl
i. n = bs Ww z wn .

& < = < Ss

= ra = a |

aU oO ae t. “G) =e

Y ey) wn w n

O 5 an) ats oO

= ee = y = a

~ ies = ; >’ = >

Fi VY) + Pe a uv) me
MLALILSNI NVINOSHLIWS SAlYVYaIT LIBRARIES SMITHSONIAN _INSTITUTIOR

ra nf Z a Zz

2) ea Ww Ww

= 0 = a: oo

= = = ce S

Cc) pad a. =

55

ty

MADRONO

A WEST AMERICAN JOURNAL OF BOTANY

VOLUME XXIV
1977

BOARD OF EDITORS

Class of:

1977—WIL11AM Louis CuLBERSON, Duke University, Durham, North Carolina
Date M. Smiru, University of California, Santa Barbara

1978—SHERWIN CaARLQUIST, Claremont Graduate School
Lestiz D. Gott ers, University of California, Davis
Dennis R. PARNELL, California State University, Hayward

1979—Puitip W. RuNDELL, University of California, Irvine
IsABELLE TAVARES, University of California, Berkeley

1980—JAMEs R. GrirFin, University of California, Hastings Reservation
FrANK A. LANG, Southern Oregon College, Ashland

1981—DaNIEL J. CRAWForD, University of Wyoming, Laramie At HSONTSS
James HEnRICKSON, California State University, Los Angeles anil o Nig ao
) sd 7 ~~

1982—DEAN W. Taytor, University of California, Davis
RICHARD VOGL, California State University, Los Angeles!

Editor — BARBARA D. WEBSTER ed Ar ae
Department of Agronomy and Range Science a =
University of California, Davis 95616

Associate Editor — Gravy L. WEBSTER
Department of Botany, University of California, Davis 95616

Published quarterly by the
California Botanical Society, Inc.
Life Sciences Building, University of California, Berkeley 94720

Printed by Gillick Printing, Inc., Berkeley, California 94710

Portrait by Imogen Cunningham, 1953

Volume 24 of Madrono is dedicated to Dr. Elizabeth McClintock in
recognition of her long service as Curator of Botany at the California
Academy of Sciences, San Francisco, her extensive knowledge of orna-
mental and economically useful plants which she willingly shared with
others, and her knowledge of the California flora and her efforts towards
its conservation. |

il

TABLE OF CONTENTS

Apgott, IsaBeLtA A. and Grorce J. HoLiensperc, Marine algae of California
(LENA NERY). oie SERS aie ane ele Neda rt PCa ere RO ee

Atwoop, N. Duane and Donatp J. PinKava, A new gypsophilous species of
Phacelia (Hydrophyllaceae) from Coahuila, Mexico -000....0000...2000cccccecceceeeeeeeeee

BARBOUR; MicHire G.(see Grams et ali). sien) ee ee

BrooMgE, C. Roser, Four new species of Centaurium (Gentianaceae) from Mexico
CannE, JupIrH M., A new combination in Cymophora (Compositae: Heli-
amtheae wy iGalnmsOommae)) eis. cis. teswickse vd: ckvsdl etait hee ee A ees

CHAMBERS, KENTON L.. (see Miller, John M.) .occccn.ececcclc tice eee ate leeesesdeenenedbsbee
CoNnsTANCE, LINCOLN -(see Mathias, Mildred E..) 220 22.....0.cc2.eccell ee ceelicee hese eset

CRAWFORD, DanieEL J., On the relationships of Chenopodium flabellifolium and
CRINGE OCIIIENI NPR es airs SDs iy oN DN Oe Ot SE a 8 1 a A ee

CRITCHFIELD, WILLIAM B., Hydridization of foxtail and bristlecone pines ............
ErRRTER, BARBARA J. and JAMES L. REveAL, A new species cf Jvesia (Rosaceae)

irom, southeastern, Oregon: 2::22--...632 ln ee eae
EWAN, JOSEPH, The tactless philosopher (review) -........22..0000.2...eetceeeeceeeeeeeeeeeeeeee
NOUN nem CSeGridiok DNER@) Mix IVE) ils 0ie oo oye. aaa. Saavan ee oo ape Webs as oecabe wen ee
Grams, Harotp J., KENNETH R. McPHERSON, Vircinia V. Kinc, Susan A. Mc-

Lrop, and MicuaEer Bargour, Northern coastal scrub on Point Reyes Pen-
DSTO S UNM cl gpae © ch LT (ONL hg ee 2 dL ciel os Seat eek sca aye Palas sedan 22 - uae seats NGa ad eaieteea tree

HextTneEr, M. M. and T. C. For, Vegetation analysis of a northern California.

coastal prairie: Sea Ranch, Sonoma County, California —.....000..0000..000eeeee eee
HOLLENBERG, GEORGE J. (see Abbott, Isabella A.) 20... ceeeeeeeeeeeeceeeeeeeeeeeeeeeeee
Jamy, S. K. (see Olivieri, A. Mi) cccccccsceccccccsscsseseeecssssteeeeeeeesee ee he at cae
KGE DOAVIbea (see. Pheroux et al. ): 2.2.82 ae Ae ec decd eee om
ISING. | VIRGINIA, V-a.(see Grams: et all). 4. .cpcl check A ee
LyNcH, STEVEN P., The floral ecology of Asclepias solanoana Woods. .......,.........-

Matus, Mitprep E, and Lincor~n Constance, Two new local Umbelliferae
(Apiaceae) from CSP OTT As, ne eek ee te ae les ee en od ee nmes

MCGEE ODOM SANA: (See Grams etal) e028 hikes oe he nce 2 tease ce

McNEAL, DALE W., Jr. and MARIon Ownsey, Status of Allium serratum (Lili-
aceae) ang description of a:New SPeClEs 25. :...cieoas oes scecense ce geccece ts sceessceeeee

McPuHerson, KENNETH R. (see Grams et al.) 22..0.0...ceecceecccceeecceeecceeeeceeeeceeeteeeeeseeeees

MILier, JoHn M. and Kenton L. CHAmBeErs, Chromosome numbers and rela-
tionships of Claytonia saxosa and C, arenicola (Portulacaceae) __.......20....0.....

Mitier, Puitip C. and Epwarp No, Root:shdot biomass ratios in shrubs in
Southern=C@antormia-and=central ‘Chile 225 scocce) coeeecssesecccxecs<seeosaeee nt oss ee setes

Mooney, Harorp A., Frost sensitivity and resprouting behavior of analagous
Shrulossors@almormid. aid: iC ile e.. 2c: 20 eee es sl eee eae

Moran, Reip, New or renovated Polemoniaceae from Baja California, Mexico
(TDOMODSIS ML INGNUINUS, INGUALVEILG ) Bcc eae

ING Hop WARD (see lviiller Palit) 2.55: 2 saeco eo 2c oa cad Stee cere

Oxivier1l, A. M. and S. K. Jarn, Variation in the Helianthus exilis-bolanderi
COMIX) pe HEC NAINA UNO Mots ccsessce8s gate ieeds Savas sets sectere eek tent, caste nee wet ere eee

iii

193

224
127
83

18

83
124
177

13

18
159

78

18

24

18

62
215
74

141
215

OLsEN, JoHN, Re-establishment of the genus Hybridella (Asteraceae: Heli-

antheae:)) fehl osc veosc an oe ae 2 ose Ren Noe 29
OWNBEY, Marion: (see- McNeal, Dale Wo) es ee ee, 24
PatTersoN, Rosert, A revision of Linanthus sect. Siphonella (Polemoniaceae) 36
Pinkava, DonaLp J. (see Atwood, N. Duane; Theroux et al.) 0000000... 193, 13
POWELL; A= MM.’ (see Turner, Bi L,)) ee a i re Pe een 1
RAVEN, PETER HLo(Ssee Seavey et cals) ').2 ln Sr i ie ee ne eee 6
REEDER, JOHN R., The “germination flap” in certain Gramineae ...........................- 123
REVEAL, JAMES L. and Re1p Moran, Miscellaneous chromosome counts of west-

ern American: iplants—=V. .c..2c_.cokc eee ee eee es 22%
REVEAL» JAMES L. (See Errter, Barbara’ |) 9052 es oe ee 224

ScHMi1D, Ruporr, and MEtvin D. Turner, Darmera, the correct name for Pelti-
phyllum (Saxifragaceae), and a new combination in Peltophyllum (Triuri-

CACEAC)) oe esi Sa Bae cS cas eee ee Aah ere St 9 OL eee a 68
SEAVEY, STEVEN R., PHitip WRIGHT, and PETER H. RAveNn, A comparison of

Epilobium minutum and E. foliosum (Onagraceae) .............22:c220c20cc2eeeeeeeeeeeeee 6
SPELLENBERG.” RICHARD: (See: Willson, James) 25.2, eee 104
STROTHER, JOHN L., A gazeteer of the Chihuahuan Desert Region (review) ........ 128
TaNowi1Tz, Barry D., An intersectional hybrid in Hemizonia (Compositae:

WEA CUM A ey ee re aoe wee elec eto ee ee BS)
THEROUX, MicHAEL E., DoNALp J. PInKAvA, and Davin J. Keri, A new species

of Flaveria (Compositae: Flaveriinae) from Grand Canyon, Arizona .......... 13
THOMAS, JOHN H., Introduction of Dr. Reid Moran .....0000.....2000..eeee cece eee 140
Turner, B. L., and A. M. PowEL1, Taxonomy of the genus Cymophora (Aster-

aceae:* Heliantheae) ).r5.2 ccuesteat cc aetecs ee a aa ce dee eee 1
TURNER; Mrtvin D, (see Schmid, Rudolf) 2.222.220.2538 ee ee al 68

WEBSTER, Grapy L., A new species of Drypetes (Euphorbiaceae) from Panama 65
WELLS, KENNETH, California mushrooms, a field guide to the boletes (review) 190

WHALEN, Mottiy, Taxonomy of Bebbia (Compositae: Heliantheae) ...................... 112
WILKEN, DiETER H., A new subspecies of Hulsea vestita (Asteraceae) .................. 48
WILLson, JAMES and RICHARD SPELLENBERG, Observations on anthocarp anato-

my in the subtribe Mirabilinae (Nyctaginaceae) _......0.......eeeceeeeeeeceeeeeeeeeeeees 104
WRIGHT: PHILIP’ (See; Seavey: et<ali) nthe eee eee 6

Dates of publication of Madrono, volume 24

No. 1, pp. 1- 64: 17 January 1977
No. 2, pp. 65-128: 21 April 1977
No. 3, pp. 129-192: 9 August 1977
No. 4, pp. 193-244: 7 October 1977

lv

A WEST AMERICAN JOURNAL OF BOTANY

QK
183
Bot:

MADRONO

VOLUME 24, NUMBER 1 JANUARY 1977

Contents

TAXONOMY OF THE GENUS CYMOPHORA (ASTERACEAE: HELIANTHEAE),
B.L. Turner and A, M. Powell

A COMPARISON OF EPILOBIUM MINUTUM AND E. FOLIOSUM
(ONAGRACEAE), Steven R. Seavey, Philip Wright, and Peter H. Raven

A NEw SPECIES OF FLAVERIA (COMPOSITAE: FLAVERIINAE)
FROM GRAND Canyon, Arizona, Michael E. Theroux,
Donald J, Pinkava, and David J. Keil

NORTHERN COASTAL SCRUB ON PoInT REYES PENINSULA,
CatirorniA, Harold J. Grams, Kenneth R. McPherson,
Virginia V. King, Susan A. McLeod, and Michael G. Barbour

STATUS OF ALLIUM SERRATUM (LILIACEAE) AND DESCRIPTION OF A
New Species, Dale W. McNeal, Jr.and Marion Ownbey,

RE-ESTABLISHMENT OF THE GENUS HYBRIDELLA
(ASTERACEAE: HELIANTHEAE), John Olsen

A REVISION OF LINANTHUS SECT. SIPHONELLA
(POLEMONIACEAE), Robert Patterson

A NEw SUBSPECIES OF HULSEA VESTITA (ASTERACEAE),
Dieter H. Wilken

AN INTERSECTIONAL HYBRID IN HEMIZONIA
(ComposiTaAE: Mapunae), Barry D. Tanowitz

NOTES AND NEWS
CHROMOSOME NUMBERS AND RELATIONSHIPS OF
CLAYTONIA SAXOSA AND C. ARENICOLA (PORTULACACEAE),
John M. Miller and Kenton L. Chambers

ON THE RELATIONSHIPS OF CHENOPODIUM FLABELLIFOLIUM
AND C. INAMOENUM, Daniel J. Crawford

13

18

24

29

36

48

55

62

63

?UBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY

Maprono is published quarterly by the California Botanical Society, Inc., and is
issued from the office of the Society, Herbarium, Life Sciences Building, University
of California, Berkeley. Established 1916. Second-class postage paid at Berkeley.
Return requested.

Editor — BARBARA D. WEBSTER
Department of Agronomy and Range Science
University of California, Davis 95616

Associate Editor — Grapy L. WEBSTER
Department of Botany, University of California, Davis 95616

Board of Editors
Class of:

1977—WIL11AM Louis CULBERSON, Duke University, Durham, North Carolina

Date M. SmirH, University of California, Santa Barbara
1978—SHERWIN CarLQuisT, Claremont Graduate School

LeEsLrE D. GoTr1ies, University of California, Davis

Dennis R. PARNELL, California State University, Hayward
1979—Puitr W. RuNDEL, University of California, Irvine

ISABELLE TAVARES, University of California, Berkeley
1980—JaMEs R. GrirFin, University of California, Hastings Reservation

Frank A. Lano, Southern Oregon College, Ashland
1981—DaniEL J. CRAWForD, University of Wyoming, Laramie

James Henrickson, California State University, Los Angeles
1982—Dean W. Tavtor, University of California, Davis

RicuarpD VocL, California State University, Los Angeles

CALIFORNIA BOTANICAL SOCIETY, INC.

OFFICERS FOR 1976

President: DENNIS R. PARNELL, Department of Biological Sciences,
California State University, Hayward CA 94542

First Vice-President: WAYNE SAVAGE, Biology Department,
California State University, San Jose CA 95114

Second Vice-President: C. LEo Hitcucock, Department of Botany,
University of Washington, Seattle WA 98105

Recording Secretary: CHARLES F. QUIBELL, Department of Biological Sciences,
Sonoma State College, Rohnert Park CA 94928

Corresponding Secretary: RUDOLF ScHMID, Department of Botany,
University of California, Berkeley CA 94720

Treasurer: G. Douctas BarBE, California Department of Food and Agriculture,
1220 N Street, Sacramento CA 95814

The Council of the California Botanical Society consists of the officers listed above
plus the immediate Past-President, JoHN H. THomMas, Department of Biological
Sciences, Stanford University, Stanford CA 94305; the Editors of Madrofo; and
three elected Council Members: Joun M. Tucker, Department of Botany, Uni-
versity of California, Davis CA 95616 (1974-1976) ; L. R. HecKarp, Jepson Her-
barium, Department of Botany, University of California, Berkeley CA 94720 (1975-
1977) ; JAMEs R. GrirFin, Hastings Reservation, Star Route Box 80, Carmel Valley
CA 93924 (1976-1978).

TAXONOMY OF THE GENUS CYMOPHORA
(ASTERACEAE: HELIANTHEAE)

B. L. TURNER
Department of Botany, The University of Texas, Austin 78712
A. M. PowELi
Department of Biology, Sul Ross State University, Alpine, Texas 79830

Anderson and Beaman (1968) gave an excellent account of the mono-
typic genus Cymophora of southern Mexico. They compared it with
Tridax and Eleutheranthera, noting that the single species, Cymophora
pringlei, was sufficiently close to Tridax accedens to be included in
Tridax. They accordingly made the transfer, assigning a new specific
epithet, 7. oligantha, the name T. pringlei having been previously
applied.

Turner, Powell, and Watson (1973) questioned the immersion of
Cymophora into Tridax, noting its seemingly closer relationship to
Sabazia and Galinsoga. Anderson and Beaman’s argument for a posi-
tion in Tvidax rested largely on its strong similarity to TJ. accedens and
their belief that the latter was part of a reduction series out of T. dubia,
a true Tvidax, leading to T. oligantha.

As indicated by our discussion under Cymophora accedens, below, we
do not now accept the position of either C. pringlei or Tridax accedens
in Tvidax. These taxa are very closely related, but presumably are more
distantly related to Tvidax proper than they are to other members of
Galinsoginae.

We also describe here a third species of the genus, Cymophora hintoni
(Fig. 1). Since Anderson and Beaman (1968) have made thorough com-
parative studies of both C. pringlez and C. accedens, we do no more than
provide a somewhat amended generic description of Cymophora, along
with a key and appropriate comments upon the several species recognized.

Cymopuora B. L. Robins., Proc. Amer. Acad. Arts 43:39. 1907.

Plants erect or suberect annuals, 12-65 cm high; stems pilose and
glandular. Leaves opposite, blades sparsely pubescent, 2.0—6.5 cm long,
I—-5 cm wide, progressively smaller upwards, broadly lanceolate, ovate,
to subdeltoid (flabellate), margins subentire, subrepand, serrulate, or
distinctly serrate; petioles 0.1-4.0 cm long. Capitulescence paniculate
or cymose; heads narrowly campanulate to subcylindrical; peduncles
with copious glandular trichomes usually mixed with straight trichomes;
receptacle convex to conical, naked, pilose, or chaffy, 0.5-2.0 mm wide;
phyllaries in 1-3 series, pilose and glandular or glabrous, concave, ob-
long, obovate, or linear, greenish, or membranous, often purplish above.

Maprono, Vol. 24, No. 1, pp. 1-64. January 17, 1977.
1

2 MADRONO LVol. 24

Ray florets absent, or disk-like, 4-5 in number, much like inner disk
florets except that 3 outer lobes may be longer than inner 2, corollas
white, achenes with or without a pappus, or with reduced pappus, rather
densely pubescent; inner disk florets 4—40, corollas white, 2.8-3.5 mm
long, tube 0.5—1.0 mm long, throat tubular or funnelform, 1.0-1.8 mm
long, lobes nearly equal, 0.5—1.5 mm long; style branches subulate, 0.8-
1.2 mm long; achenes black, narrowly obconical or obpyramidal, 1.2-
2.3 mm long, 4—5 angled and sulcate, short pilose mostly in sulci or on
basal half; pappus absent, or of 8-20 scales, the scales lanceolate, fim-
briate-plumose, and 1—2 mm long, or oblong, fimbrillate at the apexes
only, and 0.8 mm long; anthers 1.5—2.0 mm long, often purplish, appen-
dages triangular, bases sagittate. Base chromosome number, x = 8.
Type species: C. pringlei B. L. Robins.

Key to Species of Cymophora

1. Leaves markedly flabellate, the blade 4-5 cm wide; florets 8-10
ee oe et ee oe, Ge oe Se See 2. C. hintonu
. Leaves ovate, the blades 1.0—3.5 cm wide; florets 10/40.

2. Achenes without pappus, or seemingly so; plants of Guerrero
a eS ae es ee, BS Re a pee 1. C. pringlei
2. Achenes, at least those of the disk, with well-developed pappus;
plants of Michoacan and northwestward . . . 3.C. accedens

—

1. CYMOPHORA PRINGLEI B. L. Robinson. Proc. Amer. Acad. Arts 43:39.
1907.—Tridax oligantha Anderson & Beaman, Rhodora 70:241-246.
1968.—TyprE: Mexico, Iguala Canyon, 2500 ft, 22 Sep 1905, C. G.
Pringle 10068 (Holotype: GH; isotypes: LL (2), MICH, MSC, UC,
US).

Anderson and Beaman gave a detailed description of this taxon and
we see no reason to enlarge upon this except to note that the character
that they use as a “ready means” for distinguishing between C. pringlet
and C. accedens (florets 10 in the former; 30—40 in the latter) has been
somewhat vitiated, for recent collections of the latter from Colima (Mc-
Vaugh 18075, LL; Feddema 2730, TEX) have somewhat smaller heads
and fewer florets (14-30) than previously reported. A more detailed
account of the relationship of C. pringlei to C. accedens is given under
the latter taxon. |
2. Cymophora hintonii B. L. Turner & A. M. Powell, sp. nov. Fig. 1.

Type: Mexico, Michoacan, Dist. Coalcoman, Huizontla, alt. 460 m,

17 Nov 1938, G. B. Hinton 12590 (Holotype: TEX; Isotypes:

MICH, UC).—Additional collection examined: Jalisco, 17 mi by

road SW of Autlan, 28 Oct 1970, Webster & Breckon 16051 (MICH).

Plantae 28 cm altae. Laminae foliorum 5—6 cm longae 3.5—5.0 cm
latae subdeltoideae basi cuneatae margine serratae; petioli 2-4 cm longi;
capitulescentia paniculata; capitula subcylindrica; involucrum 2-3 seri-

1977] TURNER & POWELL: CYMOPHORA

Fic. 1. Habit of Cymophora hintonii (Hinton 12590; holotype).

4 MADRONO [Vol. 24

. accedens

. hintonii

C. pringlei

Fic. 2. Collection sites of Cymophora species.

j\4

id plyif
azn ts i
iN Hf hy) Wak Z
NINE VEZ

Yes,

4 H 3
EN 4 ANG | g fs !
oh \ Vs;
ye AO Y
oN

my

=
Im

Fic. 3-5. Achenes of Cymophora species. 3. C. pringlet (Pringle 10068; isotype).
4. C. hintonii (Hinton 12590; holotype). 5. C. accedens (Hinton 12884; isotype).

1977] TURNER & POWELL: CYMOPHORA 5

atum; flores disci externi 4 epapposi interni 4—5 corollis 2.8-3.0 cm

longis styli subulati; achaenia anguste obpyramidalia 1.2-1.5 mm longa

4—5 angulata; pappi squamae 8 oblongae (raro nullae).

Plants ca. 28 cm high; leaf blades 5—6 cm long, 3.5—5.0 cm wide, sub-
deltoid and basally cuneate (flabellate), margins serrate; petioles 2—4
cm long; capitulescence paniculate; heads subcylindrical; receptacle
convex, naked, ca. 0.5 mm wide; phyllaries in 2-3 series, 2-3 mm long,
glabrous, sublinear, rather membranous; ray florets (or outer disk
florets) 4, corolla with 3 outer lobes longer than inner 2, achenes short-
pilose, epappose; inner disk florets 4—5, corollas 2.8-3.0 mm long, tube
0.8-1.0 mm long, throat 1.0-1.2 mm long, subfunnelform, lobes 5, ca.
1.0 mm long, nearly equal; style branches subulate, ca. 1.0 mm long;
achenes narrowly obpyramidal, 1.2—-1.5 mm long, 4—5-ridged and sulcate,
short pilose with hairs mostly in sulci and on basal half; pappus (when
present) of ca. 8 oblong scales, ca. 0.8 mm long, rather closely enclosing
the corolla tube, obtuse and fimbrillate apically; anthers ca. 1.5 mm
long; chromosome number, unknown.

3. Cymophora accedens (Blake) B. L. Turner & A. M. Powell, comb.
nov.—Tvridax accedens Blake, J. Wash. Acad. Sci. 33:270. 1943. See
detailed description, Powell, 1965, p. 95.—Additional specimens ex-
amined: Mexico, Colima, 11 mi SW of Colima on Manzanillo rd,
500 m, 21 Sep 1958, R. McVaugh 18075 (TEX); ca. 20 km SW of
Colima on the rd to Tecoman, 500 m, 25 Nov 1963, C. Feddema 2730
(TEX); Michoacan, Dist. Coalcoman, Ixtala, 22 Sep 1938, G. B.
Hinton 12226 (LL, UC).

Powell (1965) recognized this taxon as belonging to Tridax, relating
this to T. dubia Rose, much as proposed by Blake. However, at the time of
his treatment Powell was unaware of the existence of Cymophora pringlei
and would have surely perceived its closer relationship to the latter taxon
were it known to him. Anderson and Beaman (1968) correctly noted
this relationship, but instead of disengaging Tridax accedens from its
seemingly comfortable position and placing this in Cymophora, they
preferred to transfer the monotypic Cymophora to the larger genus. This
seemed reasonable enough except that none of these workers was aware
at that time of the chromosome number of Cymophora pringlei (2n =
16; Turner et al., 1973), which is on a base of x = 4 or 8, much as its
presumably more closely related genera Galinsoga and Sabazia (includ-
ing Tricarpha Longpre; Urbatsch and Turner, 1975). Tvidax (including
T. dubia, the taxon upon which both Powell and Anderson and Beaman
hinge the relationship of T. accedens) is dibasic with x = 9 and 10
(Powell, 1965), and we believe Cymophora (including C. accedens) is
thus not correctly placed in Tridax. While the chromosome number is
not known for C. accedens, considering its very similar habital and floral
features in common with C. pringlei (Anderson and Beaman, 1968), we
have little doubt that it too will be a base of x = 8.

6 MADRONO [Vol. 24

Finally, it should be noted that with the description of the quite dis-
tinct Cymophora hintoniu (above) the genus now has three well-marked
species (Fig. 3-5), and presumably additional taxa will be uncovered
as this poorly known region of Mexico in which they occur becomes
better known.

ACKNOWLEDGMENTS
Professor M. C. Johnston provided the Latin description and Dr.
Judith Canne made several helpful emendations to the original draft;
for these contributions we are grateful.

LITERATURE CITED

ANDERSON, C. E. and J. H. BEAMAN. 1968. Status of the genus Cymophora (Com-
positae). Rhodora 70:241-246.

PowELt, A. M. 1965. Taxonomy of Tridax (Compositae). Brittonia 17:47—96.

Turner, B. L., A. M. PoweELt, and T. J. Watson. 1973. Chromosome numbers in
Mexican Asteraceae. Amer. J. Bot. 60:592-596.

UrpatscH, L. E. and B. L. TurNEr. 1975. New species and combinations in Sabaziza
(Heliantheae, Galinsoginae). Brittonia 27:348-354.

A COMPARISON OF EPILOBIUM MINUTUM AND
E. FOLIOSUM (ONAGRACEAE)

STEVEN R. SEAVEY
Lewis & Clark College, Portland, Oregon 97219

PHILIP WRIGHT
California State College, Bakersfield, California 93309

PETER H. RAVEN
Missouri Botanical Garden, St. Louis, Missouri 63110

All but six of the approximately 200 species comprised in the genus
Epilobium have a gametic chromosome number of 2 = 18, with some
polyploidy in sect. Chamaenerion Tausch. Among the unusual species,
E. minutum Lindl. ex Lehm. (Fig. 1A) has 2 = 13 (Kurabayashi et al.,
1962; Taylor and Mulligan, 1968), and what has generally been re-
garded as a trivial form or variety of it, E. foliosum (T. & G.) Suksd.
(Fig. 1B), has n = 16. These two closely related species together com-
pose one of the sections of the genus, sect. Crossostigma (Spach) Raven
(Raven, 1976). The purpose of this paper is to present the cytological
and morphological evidence that distinguishes them and to outline their
respective distributions in western North America. They were treated
in a preliminary fashion, on the basis of information we provided, by
Munz (1965, p. 208-9).

1977] SEAVEY ET AL: EPILOBIUM 7

\

Fic. 1. The 2 species of Epilobium sect. Crossostigma. A, E. minutum. B, E.
foliosum.

Both Epilobium minutum and E. foliosum are annuals, a habit they
share with only one other species in the genus, FE. paniculatum Nutt. ex
T. & G. Unlike most of the other members of the genus, these three
species are found in xeric habitats. We do not believe that there is a
direct relationship between the spring-blooming sect. Crossostigma, con-
sidered here, and the summer-blooming Epilobium paniculatum. Conse-
quently, we hypothesized that the annual habit has been derived inde-
pendently in each line (Raven, 1976).

8 MADRONO [Vol. 24

It is difficult to suggest a hypothesis for the derivation of the unique
chromosome numbers of Epilobium minutum (n = 13) and E. foliosum
(n = 16). In view of the absence of intermediate chromosome numbers,
however, we believe that they are most likely derived independently
from presumably extinct diploid ancestors with m = 8, 7, and 6 (Steb-
bins, 1971, p. 193).

Both Epilobium minutum and E. foliosum are highly autogamous,
with £. foliosum often actually cleistogamous; as a result of this, many
highly inbred morphologically distinguishable lines are apparent in both
species. Several morphological characters, however, consistently distin-
guish the two species. The flowers of £. minutum are larger than those
of EL. foliosum (the petals up to 5 mm long), and the buds are broadly
ovoid, often nodding, and borne in relatively uncrowded inflorescences.
The narrowly elliptic cauline leaves are subacute at the apex and flat
or nearly so. Epilobium foliosum, on the other hand, has erect, narrowly
ovoid buds in a crowded inflorescence. The flowers are often cleistoga-
mous and the petals rarely exceed 2 mm in length. The narrowly elliptic
leaves are acuminate and often folded along the midrib. Axillary fas-
cicles of leaves are more common than in EL. minutum.

Additional differences between these two species include seed size and
seed coat architecture. The seeds of EL. foliosum are smaller, 0.64—0.85
mm long, and very slightly papillose when viewed with a dissecting
microscope (X 20). When viewed with the scanning electron micro-
scope, the seed coat is seen to be composed of a series of smooth raised
areas, one over each cell lumen, with these raised areas separated by a
reticulate network of minutely papillose side walls (Fig. 2, D-F). The
larger seeds of E. minutum, 0.86—-1.20 mm long, are smooth when viewed
with a dissecting microscope and generally minutely papillose with a
depressed area over each cell lumen when viewed with the scanning
electron microscope (Fig. 2, A-C). Both of these seed coat types appar-
ently are confined to sect. Crossostigma and unique to the respective
species.

The distributions of these two species are similar (Fig. 3). Not only
do they overlap geographically, but they are also often found growing
sympatrically; a number of collections we examined included both spe-
cies. Epilobium minutum is the more common northward, whereas E.
foliosum is more frequent in the south. The prevalent autogamy of both
species, coupled with their small, tufted seeds, makes them excellent
colonizers, and they are often found in disturbed areas such as roadcuts
and recently burned areas.

Epilobium foliosum has been found in California as far south as Los
Angeles Co., and has been collected at least twice in Arizona (Gila Co.,
Peebles & Smith 13273, ARIZ, CAS, POM, US; Pinal Co., Nelson 1905,
ARIZ, GH, NY, UC, US). Its northward distribution follows the coastal
foothills from Ventura Co., California, through Oregon and Washington

1977 | SEAVEY ET AL: EPILOBIUM 9

ESS

Fic. 2. Scanning electron micregraphs of seeds. A-C, Epilobium minutum;
D-F¥, E. foliosum.

into British Columbia, with northeastern limits in Idaho and northern
limits on Vancouver Island, British Columbia (Taylor 46108, UC).
The species is also scattered in the foothills of the Sierra Nevada of
California, where it has been collected as far south as Tulare Co.

Epilobium minutum is less frequent in southern California, but ranges
through the coastal foothills from Santa Barbara Co., California, to
British Columbia. It is scattered in the Sierra foothills from Mariposa
Co., California, northward and, like E. foliosum, is relatively poorly rep-
resented in this part of its range. The northern distribution of this species
is more extensive than that of E. foliosum, and it reaches Stuie (40 mi E
of Bella Coola), British Columbia (52° N, McCabe 1546, UC), and the
Queen Charlotte Islands at nearly 54° N (Calder and Taylor, 1968, p.
438). The northeastern limits of H. minutum are in Montana (Flathead
Co., Holzinger & Blake 33, US; Missoula Co., Hitchcock 1662, POM).

A collection from Guadalupe Island, Baja California (Palmer 4217,
in 1875, K, MO, NY, UC) is difficult to assign to either species on the
basis of the available material. No plants of either species have been
collected on Guadalupe Island subsequently, and perhaps the plant is
extinct there. For the present, we place this collection in E. foliosum,
mainly on the basis of flower size. Additional material, if it were to be-
come available, might cause us to alter this determination.

10 MADRONO [Vol. 24

EPILOBIUM MINUTUM Lindl. ex Lehm., in Hook., Fl. Bor.-Am. 1:207.
1833.—Crossostigma lindleyi Spach, Nouv. Ann. Mus. Paris 4:404.
1835; based on Epilobium minutum Lindl. ex Lehm.—Epilobium
lindleyi (Spach) Rydb., Fl. Rocky Mts. 586. 1917.—Epilobium ad-
scendens Suksd., Deutsch. Bot. Monats. 18:87. 1900; illeg. subs. for
E. minutum Lindl. ex Hook.—LEcToTyPE: Washington, Skamania
Co., abundant on a mountain near Stevenson, 3-5 Sep 1825, David
Douglas (Hololectotype: K, isotype: NY; selected by Munz, 'N.
Amer. FI. IT. 5:208. 1965).

Epilobium minutum var. canescens Suksd., Deutsch. Bot. Monats. 18:87.
1900. Type: Washington, Klickitat Co., Falcon Valley, 27 June 1892,
N. Suksdorf (WS).

Annual, simple or diffusely branched, (3—) 5—25(40) cm tall, suberect;
leaves mostly alternate, narrowly elliptic, entire or remotely and ob-
scurely denticulate, 1-2(—2.5) cm long, the lowest ones opposite, lanceo-
late or oblanceolate; buds broadly ovoid, not crowded, often nodding
individually; petals (1.5—)2—4(—5) mm long; stigma 4-lobed or entire;
seeds obovoid, smooth, 0.86-1.20 mm long, the coma cay) deciduous;
gametic ehronie one ecener n = 13.

Chromosome count vouchers, alla = 13: US. emo Butte Co.:
Magalia, seeds from Howell 3745 7, CAS, grown at Stanford Univ., Raven
19088 (MO). Del Norte Co.: Patrick Ck., Raven 18380 (MO): 7.6 mi
NE of Crescent City along Hwy. 199, seeds from Breedlove 3087 (MO),
grown at Stanford Univ., Raven 19761 (MO). Lake Co.: 0.2 mi E of
Lakeport on road to Hopland, Raven 18236 (MO); 5.9 mi E of Houghs
Springs, Raven 18402 (MO). San Benito Co.: 5.4 mi from Hernandez
on road to New Idria, Raven 10585 (MO). San Mateo Co.: Jasper
Ridge, Stanford Univ., Raven 18258 (MO). Santa Barbara Co.: road
from Davey Brown Public Camp to Sargent Cypress Forest, seeds from
Hardham 3381 (CAS), Raven 18774 (MO). Santa Clara Co.: near Red
Mt., on J17, Seavey 1090 (MO). Sonoma Co.: intersection of Joy and
Bittner Roads, SW of Occidental, Seavey 1092 (MO). Stanislaus Co.:
Arroyo del Puerto, 6 mi E of Junction in San Antonio Valley, Raven
18229 (MO). Trinity Co.: 6.3 mi S of Trinity Center on road to Miner-
ville, Raven & Snow 13571 (RSA; Kurabayashi et al., 1962). OREGON.
Douglas Co.: bluffs of the Umpqua, opposite Roseburg, Raven 15383
(MO). WasHIncTON. Kittitas Co.: DeRoux Forest Camp, upper N
Fork of Teanaway R., seeds from Kruckeberg, Raven 18990 (MO);
Denny Moore Forest upper Cle Elum R., seeds from Kruckeberg, Raven
18991 (MO).

CaNADA. BRITISH COLUMBIA: Queen Charlotte Islands (Taylor and
Mulligan, 1968, p. 92; three localities). South Pender I., near Victoria,
seeds from Calder 29886 (DAO, MO). Lillouet Dist., above Piebiter
Ck., Bralorne, seeds from Kruckeberg 5733 (WTU), Raven 19090
(MO); N end of Butte L., Vancouver Is., seeds from Calder 30566
(DAO).

1977]

Hig, 3:

SEAVEY ET AL: EPILOBIUM 11

oO Epilobium minutum

a Epilobium foliosum

Ranges of Epilobium minutum and E. foliosum.

12 MADRONO [Vol. 24

EPILOBIUM FOLIosUM (T. & G.) Suksd., Deutsch. Bot. Monats. 18:87.
1900.—Epilobium minutum var. foliosum T. & G., Fl. N. Am. 1:490.
1840. Type: Dry rocks, Oregon and the Rocky Mountains of Cali-
fornia, 1834-5, Thomas Nuttall (Holotype: NY; isotopes: BM, K).

Epilobium minutum var. bioletti Greene, Pittonia 2:296. 1892. Type:
California, Marin Co., Sequoia Canyon above Mill Valley, Mt. Tam-
alpais, May 1892, Bioletti (Holotype: UC).

Epilobium foliosum var. glabrum (T. & G.) Suksd., Deutsch. Bot. Mo-
nats. 18:87. 1900. Type: Washington, Klickitat Co., moist, stony
places, rare, on western plains, 17 May 1892, NV. Suksdorf 2108 (Holo-
type: WS; isotypes: NY, UC, US).

Like EL. minutum, but leaves more pointed, very narrowly elliptic, the
lowest ones narrowly lanceolate, often folded along midrib; axillary fas-
cicles more frequent; flowers smaller, often cleistogamous, pointed in
bud, usually crowded, erect; petals 1-2(—2.5) mm long; seeds slightly
papillose, 0.64—0.85 mm long. Gametic chromosome number, ” = 16.

Chromosome count vouchers, all 7 = 16. U.S. CALIForNIA. Fresno Co.:
Simpson Meadow, Middle Fork Kings R. (seeds from Howell 33792,
CAS), Raven 19083 (MO). Mariposa Co.: 3.7 mi NW of Coulterville,
along Hwy. 49, Raven 18348 (MO). Monterey Co.: Twin Valley, El
Piojo (seeds from Hardham 5650, RSA), Raven 17525 (MO). Tuol-
umne Co.: ridge above Twain Harte (seeds from Lewis & Snow 1243,
LA), Raven 16025 (MO). Ventura Co.: Rose L. (seeds from Hardham
6126, RSA), Raven 17741 (MO). Ivano. Valley Co.: E side of Big
Payette L., Raven 18518 (MO). Orecon. Douglas Co.: 20 mi E of
Roseburg (seeds from Gratkowski, OSC), Raven 19089 (MO).

ACKNOWLEDGMENTS
This study has been supported by a series of grants from the U.S.
National Science Foundation to P. H. R., and a California State College
Bakersfield Foundation grant to S. S. We are grateful to the curators of
herbaria who made material of these species available to us: ARIZ, CAS,
DS, GH, ID, JEPS, K, LA, MO, NY, ORE, OSC, POM, RSA, UC, US,
WS, WTU.
LITERATURE CITED
Caper, J. A. and R. L. TAytor. 1968. Flora of the Queen Charlotte Islands. Part 1.
Systematics of the vascular plants. Res. Branch, Canada Dept. Agric. Monogr.
4, Pt. 1:i-xiii, 1-659.
Kurapayasui, M., H. Lewis, and P. H. Raven. 1962. A comparative study of
mitosis in Onagraceae. Amer. J. Bot. 49:1003-1026.
Muwz, P. 1929. New plants from Nevada. Bull. Torrey Bot. Club 56:163-167.
Raven, P. H. 1976. Generic and sectional delimitation in Onagraceae, tribe Epilo-
bieae. Ann. Missouri Bot. Gard. 63(2): (in press).
SrepBins, G. L. 1971. Chromosomal evolution in higher plants. Addison-Wesley
Publ. Co., Reading, Mass.
Taytor, R. L. and G. A. Muttican. 1968. Flora of the Queen Charlotte Islands.
Part 2. Cytological aspects of the vascular plants. Res. Branch, Canada Dept.
Agric. Monogr. 4, Pt. 2:i-ix, 1-148.

A NEW SPECIES OF FLAVERIA (COMPOSITAE:
FLAVERIINAE) FROM GRAND CANYON, ARIZONA?

MiIcHAEL E. THEROUX
Department of Biology, Museum of Northern Arizona, Flagstaff 86001

DonatLp J. PINKAvA and Davi J. KEIL
Department of Botany and Microbiology
Arizona State University, Tempe 85281

Recent collections from remote areas of the Grand Canyon National
Park, Coconino and Mohave Counties, Arizona, include a hitherto unde-
scribed endemic species of Flaveria Juss. Flaveria is a widespread genus
of mostly American distribution. Our new species, Flaveria mcdougalli,
is the second member of the genus known from Arizona and extends
the range of the genus into the Mohave Desert region.

We are naming the new species in honor of Dr. Walter B. McDougall,
Curator of Botany at the Museum of Northern Arizona since 1955 and
a long-time student of the flora of the National Parks System and
northern Arizona. His many contributions are summarized by Hamann
(1974).

Flaveria mcdovgallii Theroux, Pinkava & Keil, sp. nov. (Fig. 1) Suf-
frutex robustus usques ad 1 m diametro, basi crassa rhizomata lignea 1-3
cm diametro, caulibus numerosis erectis vel ascendentibus, usque ad 1
m altis, glabris, fragilibus, in dimidio superiore herbaceis virentibus,
porcis decurrentibus leviter angulosis, in dimidio inferiore aetate ligne-
scentibus et tephro-brunnescentibus. Folia opposita, carnosa, basi con-
nata, linearia vel anguste lineari-lanceolata saepe falcata, 5-14 cm longa,
2-8 mm lata, apice acuta, basi cuneata, margine integra, 1—5-nervata,
glabra. Capitula discoidea, sessilia vel brevipedunculata (1-2 mm), in
fasciculis oligocephalis in cymis corymbiformibus 4—5 cm diametro aggre-
gatis; ramuli capitulaque a bracteolis subtenti; bracteolae infimae foli-
aceae, supernae reductae et squamiformes, 1-2 mm longae. Involucrum
turbinatum; phyllaria 3-6, oblongo-lanceolata, 2.0-3.5 mm longa, 0.6-
0.8 mm lata, dorsaliter 3-5 nervata, apicibus acutis erectis vel vix paten-
tibus, basibus truncatis, in marginibus et adaxiale prope apicem tricho-
matibus multicellularibus 0.02—0.04 mm longis puberulenta vel glabres-
centia. Flosculi ligulati nulli. Flosculi tubuliflori 3-6; corollae flavae,
3.5-4.0 mm longae, glabrae, tubis 1-2 mm longis, faucibus 1.0-1.5 mm
longis, lobis 1 mm longis ad maturitatem reflexis; antherae exsertae the-
cis 1.0-1.5 mm longis, basaliter rotundatae vel truncatae, appendicibus
terminalibus oblongis, 0.3 mm longis et 0.1 mm latis, subacutis; pollenis
granum circa 23 um diametro; styli basaliter bulboso-incrassati, glabri,
altitudine fere antheras aequantes, ramulis exilibus recurvatis apice peni-

1 Publication No. 16, Colorado River Research Series, Grand Canyon National Park.

13

14 MADRONO [Vol. 24

cillatis, extus papillosis. Achenia 1.5—2.5 mm longa, brunnescentia vel
nigrescentia, cylindrica, circa 0.4 mm diametro, 10-costata, trichomati-
bus duplicibus 0.05—0.08 mm longis puberulenta vel glabra, carpopodio
eburneo obliquo 0.1 mm longo et 0.3 mm lata; pappus coroniformis,
squamellis fimbriatis 0.2-0.4 mm longis, plus minusve connatis. Chromo-
some numerus: 2” = 18).

Fic. 1. Flaveria mcdougallii. a, Habit; b, Head; c, Corolla; d, Style; e. Stamens;
f, Achene; g, Pappus. c, d, e, f, same scale.

1977] THEROUX ET AL: FLAVERIA i)

TyprE: United States, Arizona, Mohave Co., Grand Canyon National
Park, Cove Canyon (174.2 mi below Lee’s Ferry), 27 Jan 1976, M. E.
Theroux 1675 (Holotype: US; isotypes: ARIZ, ASC, ASU, DES,
GCNP, GH, MNA, NY, RSA, SRSC, TEX, UC).

PARATYPES: United States, Arizona, Coconino Co., Grand Canyon
National Park, Matkatamiba Canyon (148.8 mi below Lee’s Ferry), 1
Oct 1975, M. E. Theroux 1519 (MNA); Mohave Co., Grand Canyon
National Park, Cove Canyon, 5 Oct 1975, M. E. Theroux 1567 (MNA).

The type locality of Flaveria mcdougalli is a permanent saline spring
situated near the mouth of a remote tributary canyon in the Grand
Canyon. The spring, located at ca. 550 m on an ESE-facing slope, is fed
by seepage from the Muav Limestone at its contact with the underlying
Bright Angel Shale. Direct sunlight reaches the site for a period of only
a few hours per day because of shading by surrounding canyon walls.
The soil where F.. mcdougallii grows is a saturated, highly organic vary-
ing mixture of sand and silt bound together by roots and rhizomes of
two grass species, Muhlenbergia asperifolia (Nees & My.) Parodi and
Sporobolus airoides (Torr.) Torr. The A-horizon averages about 30 cm
deep and overlies an unstable, shifting aggregation of coarse shale talus.
Flaveria mcdougallii grows in soils ranging from a loose mud to consoli-
dated travertine oxides. The other site where PF. mcdougallii has been
collected is very similar to the type locality.

Both collection localities for F. mcdougallii are situated in the narrow
strip of Mohave Desert vegetation that parallels the Colorado River
through the western two-thirds of the Grand Canyon. The most common
woody species represented in the area are Acacia greggii A. Gray, Lycium
palidum Miers, and Zizyphus obtusifolia (T. & G.) A. Gray. The new
species is the dominant shrub at the springy sites. In addition to the
grasses and shrubs mentioned above, associated species include Brick-
ellia longifolia S. Wats. and Mentzelia pumila (Nutt.) T. & G. in wet
soil and Adiantum capillus-veneris L., Mimulus cardinalis Dougl., and
Petrophytum caespitosum (Nutt.) Rydb. on seepage rock-faces.

Reproduction in the known populations of Flaveria mcdougallii is
apparently predominately vegetative. Woody horizontal rhizomes 4—5
cm below the soil surface connect adjacent clumps. Seed set is low
with many ovaries shrunken and apparently abortive. According to Dr.
A. M. Powell (pers. comm.) perennial Flaveria taxa are self-sterile. The
low seed set of F. mcdougallii may indicate a low incidence of cross-
pollination. Our success in germinating a few achenes on moistened filter
paper indicates that at least some of the seeds that do form are viable.

Flaveria is morphologically similar in many respects to both Sartwellia
A. Gray and Haploésthes A. Gray (Turner, 1971, 1975). All three ge-
nera are characterized by opposite, often fleshy, basally connate leaves;
yellow-flowered, cymosely-clustered heads with distinct, uniseriate or
biseriate involucral bracts; penicillate-tipped, recurved style branches;

[Vol. 24

MADRONO

16

sjtos snoosdAs Aq

UOIsdyY Jlasaq UeNYyenyIYyD

siTey-3qnoqd

snoiqeyr)

sop}stiq AURA

sos snoasdAs Aq

UOIsdyY Jlasaq] UeNYenyIYy?)

(SZ6T ‘ouIny, pue yWWS

Aq poyodar satzads auo) *+)
siley-3[qnoq

snoiqe*)
(soreds auo

UI) SsayeoS pa}UN JO UMOII 10
sayeos pure sof}sliq SUTVeUII}Y

ssutids aul[es
UOIZIY Jasaq, VALYO]T

(O'0E— = 9% Der) °O
siTey-a[qnop 10 snoiqeys)

snoiqe[s)

sajeos po}lun JO UMOID

winsddA3 Sutpnypsut
Sd}BIJSqNS JSIOW SNOTIPA
(yAasaq, BAPYOTT
jou ynq) peoidsapi
(S261 ‘touIny,
pue yywsg) *) pue *D
snoiqey‘)
(uOIda1 [e}sBvOd WOT}
syur[d ouios ul snoiqrys
IO) SOWOYIT} Te[N{[I17[N J

(satseds au0 UL) soles
ayeredas 10 sndded oN

sAemyyed

yeyqeH
uOTNgLysI

dTJ9Yy}UASO}OY g

aouadsaqnd [eusyoy

aousoseqnd eI[0107

adA} snddeg

OOT — 02 Or -OT OS Slt sja1Op IsIpP Jo JaquinN,
gro Ss Ce re) I-O syoioy Avi Jo 1aquinN
9} [09DUP
3}VAO 3}2AO a}e[OIUL[-3UO[GO 0} BuO[GO A[MOIIeN adeys Arey[Aqd
3}€A0 10
d}e[OINUL]-IvIUT] IO IvIUTT IVIULT d}e[OIUL[-IvIUT] 10 AIUTT ‘gy e[OINUR[-Iv9UT] ‘TeaUl'T adeys je]
say4saojdv DIY] aMNIADG 22]JDSNOPIU DIAIAD]A DIAIUD] Jopeieyy
AVNIMIAVIY FHL JO SUATWATAT TO SOLLSINALOVAVHD JO NoSIivdWO) “T ATavL

1977] THEROUX ET AL: FLAVERIA 17

and brown to black, cylindrical, 8—10-ribbed achenes. All apparently
have the same chromosome base number, x = 18 (Turner, 1975). Recent
authors (Turner and Johnston, 1961; Turner, 1971, 1975; Powell, pers.
comm.) have maintained the three as separate genera. The principal
morphological differences among the three genera are number of flowers
per head, pappus structure, and pubescence of florets (Table 1).

The features of Flaveria mcdougalli are to some extent intermediate
between those of Flaveria and Sartwellia as traditionally delimited
(Table 1). Flaveria mcdougalli has slightly broadened leaves and eradi-
ate, few-flowered heads with narrow phyllaries, characteristics that it
shares with other Flaveria taxa. It also has a coroniform pappus and
glabrous corollas, features that have until now been characteristic of
Sartwellia. Some plants from one population of F. mcdougalli have
glabrous achenes whereas other members of the same population and
representatives of the other population have achenes puberulent with
double-hairs (Zwillingshaare). According to Powell (pers. comm.), the
double-hairs are characteristic of Sevtwellia and until now have not
been known in Flaveria.

If morphology alone were the deciding factor, the continued accep-
tance of Flaveria and Sartwellia, and possibly even Haploésthes as dis-
tinct genera might be questioned. Fortunately, systematists have at their
disposal a battery of experimental techniques for gathering information
about relationships. Powell (pers. comm.) is currently carrying on an
experimental investigation of taxa in the Flaveriinae that hopefully will
clarify the biological relationships among the genera. We have made
material of F. mcdougalli available to Dr. Powell for his study.

ACKNOWLEDGMENTS

We thank the Superintendent of the Grand Canyon National Park
for granting permission to collect specimens. Carbon isotope ratio was
determined by Dr. Austin Long and Dr. Juan Carlos Lerman of the Uni-
versity of Arizona Laboratory of Isotope Geochemistry. We thank Dr.
A. M. Powell and Dr. B. L. Turner for reviewing our manuscript and for
their many helpful comments. Illustrations were prepared by Miss
Wendy Hodgson.

LITERATURE CITED
Hamann, S. 1974. Ninety-one years of enthusiasms. Arizona Highways 50(10) :2-4.
SMiTH, B. N. and B. L. Turner. 1975. Distribution of Kranz syndrome among
Asteraceae. Amer. J. Bot. 62:541-545.
Turner, B. L. 1971. Taxonomy of Sartwellia (Compositae-Helenieae). Sida 4:
265-273.
. 1975. Taxonomy of Haploésthes (Asteraceae-Senecioneae). Wrightia 5:
108-115.
, and M. C. Jounston. 1961. Chromosome numbers in the Compositae
III. Certain Mexican species. Brittonia 13:64-69.

NORTHERN COASTAL SCRUB
ON POINT REYES PENINSULA, CALIFORNIA

Harotp J. GRAMS, KENNETH R. MCPHERSON, VIRGINIA V. KING,
SusAN A. MacLeop, and MicHAEL G. BARBOUR
Botany Department, University of California, Davis 95616

According to Munz (1959) and Ornduff (1974), northern coastal
scrub extends in a narrow coastal strip from southern Oregon to Point
Sur, Monterey County. The vegetation type has recently been reviewed
by Heady et al. (1977), who showed it to be a discontinuous community
within a grassland matrix, dominated either by Baccharis pilularis var.
consanguinea or Lupinus arboreus. Few quantitative studies for the type
exist. The grassland-scrub mosaic is complex, and Howell (1970) pointed
out that the scrub “. . . is perhaps the least definite in its boundaries

” of any Marin County community.

Our objective was to document quantitatively the composition and
structure of northern coastal scrub for an area representative of its
range. Point Reyes Peninsula was selected as the study site because of
its proximity to Davis, its topographic and soil diversity, its protected
(National Park) status, and its relatively well-known land-use history.
Within the Peninsula, major sample areas were on the slopes of Mt.
Vision and near Laguna Ranch (Fig. 1). The elevation of Mt. Vision
sites made the vegetation typical of scrub to the north, in the center of
its distributional range. Additional, minor sample areas were near To-
males Po'nt and in Tomales State Park.

DESCRIPTION OF THE AREA

The climate of Pt. Reyes Peninsula is Mediterranean, modified by a
maritime influence. It is characterized by cool, foggy but otherwise dry
summers and cool, wet winters.

Temperature and rainfall data were obtained from unpublished Na-
tional Park Service records from the Bear Valley Weather Station
(24 m elevation) for 1964-75. Due to topography, annual rainfall varies
markedly within short distances, but only sporadic climatic records exist
for the Mt. Vision area (395 m), and this prohibits comparison between
our high and low elevation sample sites. Mean annual precipitation at
Bear Valley was 1,052 mm over the 11-yr. period, with the greatest
amount occurring October through April. The driest months are July-
September and total rainfall then averages 7 mm. According to Howell
(1970), the average annual temperature is 11.4°C. The range.of average
temperature from warmest to coldest months is less than 2°C, although
the extremes have ranged from 38.9°C (September, 1971) to — 9.4°C
(December, 1972) at Bear Valley. Sub-freezing temperatures occur any-
time from October to May but are infrequent and seldom include hard
frosts. Prevailing winds are WNW. 1 OT mies)

18

1977] GRAMS ET AL: SCRUB 19

Pt. Reyes is gently rolling to hilly; elevation ranges from sea level to
435 mon Mt. Wittenburg. Inverness Ridge, averaging 300 m, runs paral-
lel to Tomales Bay from Tomales Point south to the Park boundary and
beyond. Ocean-facing cliffs, about 20-50 m high, are common along most
of Drakes Bay and the Pacific Ocean coast.

The vegetation was mapped by Donald Lauer in 1973 (Point Reyes
National Seashore, 1973), and his map shows a mosaic of coastal prairie,
northern coastal scrub, closed-cone pine forest, mixed evergreen forest,
and localized patches of dune scrub close to shore.

Point Reyes Peninsula is joined to the mainland along the San An-
dreas Fault. According to Howell (1970) and Feray et al. (1968), the
peninsula is composed of three major geologic formations. The NE
quarter and the tip of the point is Montara granitic formation of Meso-
zoic origin. The SE half of the peninsula is Monterey shale formation of
Miocene origin. Some stabilized sand dune deposits overlay the Mon-
terey shale in the central coastal portion.

The U.S. Soil Survey is in the process of mapping soil types of Point
Reyes Peninsula. Through personal communication with Gordon Ship-
man of the Marin County Soil Survey staff, we were able to define the
general soil types of the primary study locations. The soil of Mt. Vision
sites is a clay loam, over a clay hardpan, about 50-100 cm deep, over
quartz diorite parent material. It is similar to the Auberry series as de-
fined by the National Cooperative Soil Survey (NCSS). Soil at the
Laguna Ranch sites is a shaley clay loam, over shaley clay, about 50-100
cm deep, over shale parent material. It is similar to the NCSS Santa
Lucia series in color, and in shallowness it is much like the NCSS Lopez
series. The soil at Tomales State Park is a coarse sandy loam up to 1.5
m deep. It has developed on a stabilized sand dune deposit and is similar
to the NCSS Sheridan series. We have no data for soil beneath the
Tomales Point site.

SELECTION OF SAMPLE SITES AND SAMPLING METHOD

Sampling areas were chosen because they represent, with one excep-
tion, undisturbed, extensive, distinct examples of northern coastal scrub.
The exception was a site on Tomales Point in an area actively grazed
by cattle. The areas provided a variety of slope, aspect, soil type, and
elevation which seemed to include the full range of habitat occupied by
the vegetation type.

The method of sampling was by line transect. A total of 20 transects
was established at 15 different sites (Fig. 1). Transects were 20 m long,
but sometimes they were subdivided into two smaller units and placed
apart from each other, depending on homogeneity of the site. Prelim-
inary observations indicated that a length of 15 m included most vari-
ability and gave a representative sample. Within a given site, transects
were subjectively located after reconnaisance. Plant cover of each species

20 MADRONO [Vol. 24

a, —

A
oO
a
a 10 km
e S*
TOMALES
POINT er
iy
PACIFIC
TOMALES
OCEAN x4 STATE @
PARK
123°W “ul @
VISION
@ LAGUNA
@ RANCH
DRAKES BAY

38°N

Fic. 1. Point Reyes Peninsula and location of northern coastal scrub sampling
sites (dots).

1977] GRAMS ET AL: SCRUB 21

TaBLE 1. ABSOLUTE COVER (%) OF ALL SPECIES ENCOUNTERED ON 15 NORTHERN
CoasTaL ScrUB SITES ON Pont REYES PENINSULA. t < 0.1% cover.

Taxa Average “N-facing”’ “S-facing”
Anaphalis margaritacea 0.3 0.5 re
Artemisia californica 2.7 ft 6.0
Baccharis pilularis var. consanguinea 31.2 Zoo 38.0
Corylus cornuta Sut 5.5 nee
Elymus condensatus 2.0 335 0.1
Erechtites prenanthoides 0.3 0.3 0.2
Fragaria californica 0.4 0.6 0.3
Galium nuttallii 0.5 0.2 0.8
Gaultheria shallon 3.9 7.0 s
Heracleum lanatum 1.6 1.6 1.6
Holodiscus discolor 0.1 G2

Lupinus arboreus t 0.1 a
Mimulus aurantiacus 7.0 1.0 14.5
Picris echioides t onl a
Polystichum munitum 28.6 44.5 8.4
Pteridium aquilinum 6.4 Be) 10.0
Rhamunus californica 20.1 4.3 40.0
Rhus diversiloba 2:2 0.2 4.6
Rubus ursinus and/or R. vitifolius 5.8 3.6 8.7
Satureja douglasii 5.9 2.4 10.4
Stachys rigida 1.6 1.4 1.9
Total 123.8 110.2 145.5

along the transect was measured to the nearest 0.5 dm. Additional data
recorded for each site were slope, aspect, average vegetation height, and
a subjective evaluation of soil moisture conditions (very dry, dry, moist,
wet, very wet). Any species not on the transect but occurring nearby
was also noted. Since the study was conducted January-March, maxi-
mum cover of Rhus and some herbs could not be measured. Nomencla-
ture follows Munz (1959) and voucher specimens are deposited in DAV.

DISCUSSION AND CONCLUSIONS

Our 15 transect sites ranged in elevation from 50 to 340 m, in slope
aspect from 90° (due E) to 350° (NNW), and in steepness of slope
from 26 to 80%. Inspection of cover data indicated that slope aspect
had the greatest effect on community composition. Table 1 summarizes
absolute cover by slope aspect category. ‘“North-facing” here includes
all slopes from 270° (due W) to 90° (due E) in orientation, and “south-
facing” includes all other orientations.

North-facing slopes were dominated in the overstory by a rather open
(26% cover) canopy about 1.5 m tall of Baccharis pilularis var. consan-
guinea. The understory, about 0.3 m tall, was dominated by Polystichum
(45% cover), with Gaultheria, Corylus, Pteridium, Rubus, and Elymus
species accounting for another 27% cover. It was essentially a two-
layered community. There was less than 2% bare ground. South-facing

22 MADRONO [Vol. 24

slopes had shared dominance by Baccharis (38% cover) and Rhamnus
californica (40% cover) in an overstory canopy about 1.0 m tall, with
an understory dominated by Mimulus, Satureja, and Pteridium species.
The southern affinity of this phase is shown by modest cover values for
Artemisia californica and Rhus diversiloba, dominants of coastal sage
scrub in the Monterey area (Heady et al. 1977). North-facing slopes
exhibited less than 4% cover by deciduous species while south-facing
slopes exhibited more than 13%, also illustrating affinity to the predom-
inantly deciduous coastal sage scrub. Species richness was greater on
north-facing slopes (total of 20 species encountered) than on south-
facing slopes (15 species). There was twice as much bare ground beneath
south-facing stands. Only a few species were completely restricted by
slope aspect: Anaphalis margaritacea, Corylus cornuta, Gaultheria shal-
lon, Holodiscus discolor, Lupinus arboreus, and Picris echioides were
found only on north-facing slopes, and Artemisia californica was found
only on south-facing slopes.

Additional, stand to stand variation was revealed by ordination. A

X

Fic. 2. Ordination of stands. For convenience, “north-facing” stands are indi-
cated as o’s, “south-facing” stands as dots. The actual aspect is indicated by the
direction of the line away from each dot or o. Stands on Mt. Vision, Tomales
Point and Tomales State Park are marked with a V, TP, TSP, respectively. Laguna
Ranch sites are unmarked.

1977] GRAMS ET AL: SCRUB 23

similarity index (S) for every pair of stands was calculated as S =
RCs/200, where RCg is the sum of the smaller of the pair of relative
cover values for each taxon, for all taxa common to the two stands being
compared, and 200 is the sum of relative cover values for all taxa in
stands A and B. This equation is Motyka’s modification of Sorenson’s
Index (Mueller-Dombois and Ellenberg, 1974).

The matrix of similarity indices was displayed by a polar ordination
method of Bray and Curtis (Cottam et al. 1974), with three mutually
orthogonal axes, X, Y, and Z.

The ordination of stands on X and Y axes is shown in Fig. 2 (these
two axes accounted for 79% of stand-stand variation, and the Z axis
gave very little additional spread of stands). The X axis correlates with
our subjective moisture assessments of stands. Relatively mesic, north-
facing stands on the left grade into relatively dry, south- facies lower
elevation stands on the right. The Y axis, however, which adds consider-
able spread to the south-facing stands, does not correlate well with any
of the other environmental factors that we recorded (steepness of slope,
land-use history, soil type, elevation). Degree of exposure to prevailing
winds (a combination of aspect, nearness to the Pacific Ocean, and open-
ness of the site to air movement) accounts for some of the spread on the
Y axis. (Subjectively, we noticed that canopy height was lowest on the
windiest sites. )

We found no evidence that northern coastal scrub is seral to forest.
In the Berkeley Hills, McBride and Heady (1968) and McBride (1974)
reported that a Baccharis scrub was being invaded by elements of oak
or mixed evergreen forest. At Point Reyes National Seashore, Research
Biologist Richard M. Brown believes the scrub may be seral to closed-
cone pine forest (pers. comm.). Although we did find an occasional tree
of Umbellularia californica, Pseudotsuga menziesii, Quercus agrifolia, or
Pinus muricaia in the midst of scrub, we saw no seedlings or saplings.

LITERATURE CITED

Cottam, G., F. G. Gorr, and R. H. Wuirtaker. 1974. Wisconsin comparative
ordination. Jn R. H. Whittaker (ed.). Ordination and classification of com-
munities. W. Junk, Hague.

Feray, D. E., P. OETKING, and H. B. RENFRo. 1968. Geological highway map of
the Pacific Southwest Region. Amer. Assoc. Petroleum Geologists, Tulsa,
Oklahoma.

Heapy, H. F., T. C. Forn, M. M. Hextner, D. W. Taytor, M. G. Barsour, and
W. J. Barry. 1977. Coastal prairie and northern coastal scrub. In J. Major
and M. G. Barbour (eds.) Terrestrial vegetation of California. Wiley-
Interscience, N.Y. (in press).

Howett, J. T. 1970. Marin flora, 2nd ed. Univ. Calif. Press, Berkeley.

McBripe, J. 1974. Plant succession in the Berkeley Hills, California. Madrofio
22:317-329.

and. H. F. Heapy. 1968. Invasion of grassland by Baccharis pilularis DC.
J. Range Managem. 21:106-108.

24 MADRONO [Vol. 24

MUELLER-DompBots, D. and H. ELLENBERG. 1974. Aims and methods of vegetation
ecology. Wiley, N.Y.

Mowz, P. A. 1959. A California flora. Univ. Calif. Press, Berkeley.

OrnpburFr, R. 1974. Introduction to California plant life. Univ. Calif. Press,
Berkeley.

Point REYES NATIONAL SEASHORE. 1973. Vegetation types for Point Reyes mapping.
Unpub. map + 2 p mimeo. legend in PRNS files.

STATUS OF ALLIUM SERRATUM (LILIACEAE)
AND DESCRIPTION OF A NEW SPECIES

Date W. McNEAL, Jr.
Department of Biological Sciences
University of the Pacific, Stockton, California 95211

MARION OWNBEY
Department of Botany, Washington State University, Pullman 99163

In 1871, Watson described ‘Allium serratum” from California. In
the protologue he cited ten collections and listed Allium amplectens
Torr. (1865) as a synonym, indicating that its type (Sonoma, 3 May
185-, Bigelow s.n., NY) “is a very young undeveloped state and the
name is inapplicable to the mature plant”. Application of the current
rules of nomenclature (Arts. 62 and 63) makes “A. serratum” superflu-
ous and illegitimate. Later Watson (1879) reconsidered his circum-
scription of “A. serratum”. Here he cited A. amplectens Torr. as a syno-
nym of A. attenuifolium Kell. (1863). (We agree that the types of these
names are conspecific; however, the correct name is A. amplectens
Torr.) “Allium serratum” then apparently referred to a taxon repre-
sented by the syntypes originally cited in 1871.

During investigation into the taxonomy of the Allium acuminatum
alliance to which “‘A. serratum” has been referred (Saghir et al., 1966)
we studied all but four (Bolander s.n., Douglas s.n., Kellogg s.n., and
Wallace s.n.) of those ten collections. We believe the six collections
studied represent two species. Five (Hartweg 1991, GH, NY; Fremont
469, GH; Bridges 345, NY, US; Stillman s.n., NY; Rich, s.n., NY)
correspond closely to Watson’s original description and he annotated
them “Allium serratum’”’. The other specimen (Benecia, 1853-4, Bige-
low s.n., GH) does not match the original description and Watson seems
to have recognized this since he annotated it ‘Allium serratum, form”.
This is reinforced by a specimen (Kellogg 1012, GH) that he annotated
in the same manner but that was not cited among the syntypes. Later
authors incorrectly applied the name “A. serratum” to these latter speci-
mens disregarding the fact that they do not correspond to Watson’s de-
scription nor his interpretation of what was “A. serratum” and what was
a form of this species.

1977] McNEIL & OWNBEY: ALLIUM 25

The name Allium peninsulare Lemmon ex Greene (1888, Holotype:
Lemmon s.n. NDG) is the correct name for Watson’s typical ‘“‘A. serra-
tum” and has long been used in this sense. We consider Watson’s “A.
serratum, form” to represent a distinct species as follows:

Allium serra McNeal & Ownbey, sp. nov. (Fig. 1).

Bulbus ovoideus vel subglobosus, 8-12 mm longus, tunico exteriore
plerumque brunneo, manifeste celluloso-reticulato, maculis transverse
elongatis, deorsum angulatis, regularibus verticalibus ordinibus dispo-
sitis, tunicis interioribus albis; folia 2-4, anguste concavo-convexa vel
subteretia, scapo aequilonga vel quarta parte breviora; scapus teres,
gracilis, 15-30 (50) cm longus; bracteae 2/3, lanceolato-ovatae vel
ovatae, acuminatae, 1-2 cm longae; umbella 10-35 (vel pluribus) flori-
bus, compacta, pedicellis 6-15 mm longis; pedicelli et flores maturi simul
decidui; segmenta perianthii integra rosea, lanceolata vel lanceolato-
ovata, acuta, obtusa vel emarginata, erecta, papyracea maturescentibus
fructibus et conniventia super capsulam; segmenta perianthii exteriora
8—11 mm longa et 3.0—5.5 mm lata, interiora breviora et angustiora; sta-
mina inclusa, antheris luteis vel rubris, apiculatis; capsula triloba, cris-
tata 3 minutis bilobis processibus circa styli basin; stigma capitatum,
trilobum; semina atra, alveolis minute asperis.

Type: California, Stanislaus Co., 20.5 mi W of Patterson in Canyon
Del Puerto, dry rocky hillside above the road, 11 Apr 1968, McNeal 397
(Holotype, WS!; isotype, CPH! )

The specific epithet refers to the outer bulb coat which, when broken,
has a serrate edge resulting from its typical herringbone reticulation
pattern (Fig. 1,f).

Representative specimens (for a complete list of specimens see McNeal,
1970): Alameda Co., Corral Hollow, Eastwood & Howell 5293 (CAS);
Butte Co., Chico, May 1918, Rixford (CAS); Colusa Co., Rt. 20, 5 mi E
of the Lake Co. line, Benson 4327 (ND, POM); Contra Costa Co., 2 mi
inside the N gate of Mt. Diablo, Hoffman 2796 (WS); Glenn Co., hills
W of Willows, Eastwood 11149 (CAS) ; Lake Co., 6 mi N of Lower Lake,
Eastwood & Howell 5583 (CAS, UC, WTU); Merced Co., 10 mi S of
Los Banos, Hoover 2891 (UC, US, WS); Napa Co., 1.3 mi S of Knox-
ville, Keck 2374 (CAS, DS, POM, UC); San Joaquin Co., E end of
Corral Hollow, Eastwood & Howell 2094 (CAS); Santa Clara Co., Eden-
vale, Thomas 8990 (DS, OSC, RSA); Solano Co., foothills of the Vaca
Mts., W of Vacaville, Heller 15559 (DS, MO, NY, UC, US, WS, WTU) ;
Stanislaus Co., Del Puerto Canyon, 8.3 mi W of Patterson, McNeal 394
(CPH, WS).

Allium serra grows on heavy clay or serpentine soils in the Inner Coast
Range of California from Butte Co. south to central Merced Co. at 300—
600 m. It is associated with such plants as Pinus sabiniana, Heteromeles
arbutifolia, Quercus sp., Rhus diversiloba, Dodocatheon hendersonii, and
several species of introduced annual grasses common throughout the
range.

26 MADRONO [Vol. 24

Allium serra belongs to the A. acuminatum alliance on the basis of its
thick, cellular reticulate bulb coat, which develops from the inner epi-
dermis of the inner leaf base (McNeal and Ownbey, 1973). It appears
to be the most closely related to A. peninsulare and A. amplectens. These
are widespread species; A. peninsulare occurs south and east of A. serra
except for two records in the southern part of its range, while A. amplec-
tens is sympatric with A. serra throughout its range. The resemblances
to these two species are in an unrelated combination of morphologic
characters. In addition to characteristics in the key below, A. serra re-
sembles A. amplectens in that the perianth segments, which are erect
at anthesis, become papery and connivent over the capsule as it matures.
Both species have short pedicels and compact umbels with the pedicel
falling as a unit with its flower when the capsules are mature. In 4A.
peninsulare the pedicels are comparatively long and the umbels are open
with the pedicels and flowers persisting after the capsule matures. In A.
peninsulare flowers are deep reddish purple while in A. amplectens they
are white or sometimes flushed with pink; A. serra has bright pink
flowers.

Chromosome numbers of both Allium serra and A. peninsulare are
2n = 14 while A. amplectens has two chromosomal races, 2n = 3x = 21
and 2n = 4x = 28, both of which occur in the range of A. serra (Table

1)

TABLE 1. CHROMOSOME COUNTS FoR Allium. All collections are from California;
vouchers are in WS. Our counts were all made during first meiotic metaphase.
* indicates previously unpublished counts by Dr. Hannah C. Aase.

Allium serra McNeal & Ownbey
Alameda Co., Livermore—Tesla Rd., 15.4 mi from Livermore, Hoffman 2799,
7 1I*; Colusa Co., above Bear Creek, ca 15 mi N of Rumsey, Henry s.n. 7 II;
Lake Co., Hwy. 20, 12 mi E of its jct. with Hwy. 53, McNeal 408, 7 II; Stanislaus
Co., Del Puerto Canyon, 20.5 mi W of Patterson, McNeal 397, 7 II; Yolo Co.,
Cache Creek, 3 mi N of Rumsey, McNeal 407, 7 II.

Allium amplectens Torr.

Colusa Co., Rt. 20, 21 mi SW of Williams, Ownbey and Ownbey 2951, 21 I*
(Achiasmatic) ; Lake Co., 6 mi S of Hwy. 20 on Scotts Valley Rd. to Lakeport,
McNeal 409, 1411; 2.8 mi N of Middleton, Ownbey and Ownbey 2954, 14 IL*;
Marin Co., Big Rock Ridge, 2-3 mi W of Hamilton AFB, Robbins s.n., 21 I*
(Achiasmatic) ; San Mateo Co., Jasper Ridge experimental area, edge of Stanford
U., Raven s.n., 1411; Stanislaus Co., Mt. Hamilton-Livermore Rd., 1.0 mi N of
Canyon Del Puerto Rd., McNeal 575, 21 I (Achiasmatic).

Allium peninsulare Lemmon ex Greene
Butte Co., Chico-Paradise Rd., 14.8 mi E of Chico, Hoffman 3774, 7 I1*; Kern
Co., Rt. 178, 1.5 mi E of Onyx, McNeal 389, 7 II.

Fic. 1. Allium serra. a, Habit. b, Flower. c, Outer perianth segment with anther
in erect position. d, Inner perianth segment with anther in versatile position. e,
Older flower with perianth segments connivent over the capsule. f, Portion of
outer bulb coat with herringbone reticulation pattern. From 35 mm transparency
of living plant and type collection.

1977 | McNEAL & OWNBEY: ALLIUM 2d

i yd
xe
‘i;

Ws
|

=
ea 4
=
———
——

HAN

28 MADRONO [Vol. 24

The following key distinguishes Allium serra from other members of
the A. acuminatum alliance that have a similar herringbone reticulation
pattern on the bulb coats.

a. Ovary crested with 6 lateral processes; meshes of the reticulum
wavy, not in sharply serrate transverse rows, forming a more or less
indistinct herringbone pattern, or contorted; inner and outer peri-
anth segments approximately equal in length and breadth.

Allium ampblectens Torr.

aa. Ovary crested with 3 minute, 2-lobed central processes; meshes of
the reticulum in sharply serrate transverse rows, forming a herring-
bone pattern; inner perianth segments shorter and narrower than
the outer.

b. Perianth segments connivent after anthesis, becoming papery;
umbel shattering, each flower with its pedicel deciduous as a
unit. 3. 4 . . . . Allium serra McNeal & Ownbey

bb. Perianth eroment not connivent after anthesis, texture dull or
shiny, never papery; flowers persisting.

c. Plants low, fleshy; leaves 3—6, the shorter ones arcuate, the longer
tortuous; bracts broadly ovate, abruptly acuminate; umbel
compact, pedicels short. Sea cliffs from San Mateo Co. to Men-
docino Co., California . . . Allium dichlamydeum Greene

cc. Plants taller, slender or stout, not appearing fleshy; leaves 2-4,
straight or curved; bracts lanceolate to ovate, acuminate; um-
bels loose, pedicels spreading. Interior California from Butte
Co. south along the Sierra Nevada foothills and Coast Range
into Baja California, also in the Santa Cruz Mountains of San
Mateo Co.

d. Inner perianth segments crisped. . Allium crispum Greene

dd. Inner perianth segments with margins entire or cae
serrulate, never crisped. A cane

Allium peninsulare emer ex Grecne

ACKNOWLEDGMENTS

We thank curators of herbaria (CAS, DS, GH, JEPS, MO, ND, NY,
OSC, POM, RSA, UC, US, WTU) from which material was borrowed
during the course of this investigation. We also thank Dr. Robert Smut-
ny of the Classics Department, University of the Pacific, for assistance
in preparing the Latin diagnosis of Allium serra; Sandra McNett for
preparing the drawings; and Dr. Hannah C. Aase for permission to use
some of the chromosomal data included in this paper.

Dr. Marion Ownbey died 6 December 1974 without seeing the final
drafts of this manuscript. As senior author, I am solely responsible for
any errors that appear.

1977] OLSEN: HYBRIDELLA 29

LITERATURE CITED

GREENE, E. L. 1888. New or noteworthy species II. Pittonia 1:159-178.

Kettocc, A. 1863. Allium attenuifolium. Proc. Calif. Acad. Sci., I, 2:110.

McNEAL, D. W. 1970. Comparative studies of the Allium acuminatum alliance.
Ph.D. Thesis, Washington State Univ., Pullman.

and M. Ownsey. 1973. Bulb morphology of some western North American
species of Allium. Madronio 22:10-24.

SacHir, A. R. B., L. K. Mann, M. Ownsey, and R. Y. BERG. 1966. Composition of
volatiles in relation to taxonomy of American alliums. Amer. J. Bot. 53:
477-484.

Torrey, J. 1856. Description of the general botanical collections of the route near
the 35th parallel explored by Lieut. A. W. Whipple 1853-1854. U.S. War Dept.
Rep. on explorations and surveys for a railroad from the Mississippi River to
the Pacific Ocean, 4(Botany).

Watson, S. 1871. Report of the geological exploration of the Fortieth Parallel. Vol.
5. Govt. Print. Off., Wash., D.C.

. 1879. Contributions to American botany. IX. Revision of the North
American Liliaceae. Proc. Amer. Acad. Arts 14:213-288.

RE-ESTABLISHMENT OF THE GENUS HYBRIDELLA
(ASTERACEAE: HELIANTHEAE)

JOHN OLSEN
Department of Botany, University of Texas, Austin 78712

Hybridella is a small genus of three herbaceous perennial taxa native
to Mexico. It was treated as a subgenus of Zaluzania by Robinson and
Greenman (1899) and by Sharp (1935), although it was described as
a genus by Cassini in 1821. During a study of Zaluzania (Olsen, 1977),
I became aware that Hybridella comprises a cohesive unit phyletically
remote from Zaluzania. Based on morphological, cytological, and eco-
logical data, it should be positioned elsewhere. Table 1 lists major
differences between the two genera.

There are only two chromosome counts available for Hybridella: H.
globosa var. globosa (n = 16, Powell and Turner, 1963) and H. globosa
var. myriophylla (n = 16, Olsen 265, LL, published here). These counts
suggest a base number of x = 16 for the genus.

The most likely relationships of Hybridella are with Heliomeris, a
Viguiera segregate (Yates, 1967). The base chromosome number of
Heliomeris is x = 8 (Turner, 1976), presumably one of the ancestral
numbers in the Heliantheae (Stuessy, 1976). It is likely that the an-
cestral base number for Hvbridella is x = 8, with stabilization occurring
at the tetraploid level. This coupled with the obvious floral similarities
between Hybridella and Heliomeris (Table 1), suggests a close relation-
ship between the two taxa.

30

MADRONO

[Vol. 24

TABLE 1. COMPARISON OF HyBRIDELLA, ZALUZANIA, AND HELIOMERIS

HYBRIDELLA

ZALUZANIA

1. Base chromosome
number, x = 16
2. Leaves pinnatisect

3. Receptacle globose to
hemispheric

4. Tube of disc corolla
pubescent with
glandular hairs

Base chromosome
number, x = 18

Leaves entire to tripar-
tite, never pinnatisect

Receptacle conical

HELIOMERIS

Base chromosome
number, x = 8

Leaves entire or
serrate, never
pinnatisect

Receptacle hemispheric

Tube of disc corolla
glabrous or with
simple multicellular
hairs

5. Ray florets (always
present) 15-20, ligule
supplied by 5-7
principal veins.

Tube of disc corolla
may be glandular

Ray florets (when
present) 8-10, ligule
supplied by 10-12
principal veins

6. Ligule of ray floret
2-lobed, usually with
3 basal lobes

. Plants generally
occupying wet
habitats; marshy
soils; sandy river-
banks, etc.

~sT |

Ligule or ray floret
(2—)3-lobed, without
basal lobes

Ray florets (always
present), ca 13, ligule
supplied by usually
7 principal veins

Ligule of ray floret
entire to 2-lobed
without basal lobes

Plants generally occu-
pying dry habitats;
Larrea flats, pine
oak forests, etc.

Plants generally
occupying dry rocky
habitats; exposed
mountain slopes,
dry plains, etc.

Hybridella may also be related to Viguiera sect. Chloracra series

Pinnatilobatae. The habit and leaf morphology, especially of H. anthe-
midifolia, resemble that found in V. stenoloba, a highly variable taxon
with respect to leaf morphology (Butterwick, 1975); however, the ab-
sence of a pappus and the presence of fertile ray florets in Hybridella
serve readily to distinguish the two.

In summary, I consider Hybridella to be more closely related to
Heliomeris and elements of Viguiera than to Zaluzania; however, it is
sufficiently distinct from both to deserve generic status.

TAXONOMIC TREATMENT
HyYBRIDELLA Cass., Dict. Sci. Nat. 22:86. 1821. Type: Anthemis globosa

Ort.

Herbaceous perennials with 1—4 stems arising from a woody caudex.
Plants less than 1 m tall, stems striate, usually glabrous at the base,
becoming pubescent above. Leaves alternate, sessile, usually obtrullate
in Outline, pinnatifid, glabrous on upper surface or with a few hairs on
the veins, pubescent beneath. Heads one to several; solitary on well
developed peduncles. Involucre of 2-3 series; bracts ovate-elliptic to
lanceolate; outer series larger than the inner, pubescent with simple
multicellular hairs. Receptacle globose or hemispherical. Chaff present,

1977] OLSEN: HYBRIDELLA 31

the pales oblanceolate to linear, herbaceous or subherbaceous. Ray
florets pistillate and fertile, 15-20 per head; corollas yellow, glandular
along the tube, ligule usually 2-lobed, often with 2 lateral lobes at the
base with a smaller central lobe (Fig. 1b). Disc florets numerous, bi-
sexual, fertile; corollas yellow, base of the tube expanded. Ray achenes
three-angled in cross section, black, glabrous to sparsely pubescent, epap-
pose. Disc achenes four-angled in cross section, black, glabrous to
sparsely pubescent, epappose. Base chromosome number, x = 16.

Fic. 1. Hybridella globosa var. myriophylla. A, habit. B, ray floret illustrating
the lateral and basal lobing of the corolla. C, disc floret. D, receptacular chaff.
Drawings from Olsen 265, LL.

32 MADRONO [Vol. 24

Key to Hybridella
Plants decumbent; leaves 2—3-pinnatifid, ultimate segments
1-2 mm brosd: or ast . . . . 1. A. anthemidifolia
Plants erect; leaves 4- pinnatifid, mlnaaace segments less
than 1 mm broad.
Ultimate leaf segments acute, more than

2mm long. . . . . . . 2a. A. globosa var. globosa
Ultimate leaf segments obtuse or rounded,
less than 2mm long. . . . 2b. H. globosa var. myriophylla

1. Hybridella anthemidifolia (Rob. & Greenm.) Olsen, comb. nov.
Zaluzania anthemidifolia Rob. & Greenm., Proc. Amer. Acad. 34:531.
1899, TypE: Mexico, Jalisco, wet sandy riverbanks near Guadala-
jara, 23 Sep 1891, Pringle 5156 (Holotype: GH! ).

Herbaceous perennials with one to few stems arising from a decum-
bent woody caudex. Plants to 0.6 m tall; stems striate, glabrous below to
sparsely pubescent with simple multicellular hairs above. Leaves sessile,
elliptic in outline, 2—3-pinnatifid, ultimate segments 1-2 mm broad.
Lower leaves usually absent; upper leaves 4.3—7.0 cm long, 1.8-2.6 cm
wide. Upper surface of the leaves glabrous, lower surface with scattered
long, thin multi-cellular hairs. Heads 1-3, solitary on peduncles 1.2—4.0
cm long; 2.0—2.6 cm wide including the rays. Involucre of 2—3 series of
ovate elliptic bracts; outer series 4.2—7.0 mm long, 2.2-3.1 mm wide;
inner series 3.6-3.7 mm long, 1.1-1.7 mm wide. Pales oblanceolate with
base wrapped around the base of the disc achenes, 2.9-3.0 mm long,
0.5-0.6 mm wide. Ray florets ca 20 per head; corollas 10.0-11.0 mm
long, 2.0-3.8 mm wide, 2- or 3-lobed with no basal lobes present. Disc
florets ca 50 per head; corollas 4.1—4.2 mm long, 0.6—-0.9 mm wide, glan-
dular. Ray achenes 1.0-1.6 mm long, 0.6-0.7 mm wide, glabrous. Disc
achenes 1.5—1.6 mm long, 0.5—0.8 mm wide, glabrous.

Distribution (Fig. 2): Known from only two collections, from bar-
rancas in the Guadalajara region, along Rio Grande de Santiago (Pringle
in 1891 and 1895). I have looked on two occasions for this taxon, both
in the vicinity of the type locality and in the general region of the Rio
Grande de Santiago, and have not been able to find it. Neither has
McVaugh collected this species, even though he has collected extensively
in the region. Flowering Sept.—Oct.

Additional specimens examined: JALisco: wet sandy banks of Rio
Grande de Santiago, barranca near Guadalajara, 12 Oct 1895, Pringle
7367 (F, MO, UC).

2. HYBRIDELLA GLoBosa (Ort.) Cass., Dict. Sci. Nat. 22:86. 1821.
Herbaceous perennials with one to few stems arising from a woody

caudex. Plants to 0.6 m tall; stems striate, glabrous below to hispidulous

or hirsute above. Leaves sessile, obtrullate in outline, 4-pinnatifid, ulti-

1977] OLSEN: HYBRIDELLA 33

% H. globosa var. myriophylla
O 4H. globosa var. globosa
a

‘ anthemidifoli
lf ss ia a AY ee
ri } | ra. Ve)

25

Fic. 2. Distribution of Hybridella.

mate segments less than 1 mm broad. Lower leaves 7.4-18.5 cm long,
1.5—6.0 cm wide; upper leaves 1.6—-5.3 cm long, 0.5—2.5 cm wide;
pubescence of upper surface usually limited to the veins, lower surface
hispidulous to strigose. Heads one to several, solitary on peduncles 2.5—
13.5 cm long; 1.6-3.2 cm wide including the rays. Involucre of 2-3
series of lanceolate to ovate-lanceolate bracts; outer series 3.8-6.7 mm
long, 1.3-2.8 mm wide; inner series 3.3-5.9 mm long, 0.7—2.4 mm wide;
both pubescent with simple, multicellular hairs. Pales linear-lanceolate to
oblanceolate, 2.0-3.9 mm long, 0.2-0.5 mm wide. Ray florets ca 20 per
head; corollas 7.2-11.7 mm long, 1.5—2.8 mm wide, ligules 2-lobed, usu-
ally with two small lateral lobes and one central lobe at base, tube glan-
dular. Disc florets ca 100 per head; corollas 2.2-3.1 mm long, 0.4—1.5 mm
wide, tube glandular. Ray achenes 0.9—2.2 mm long, 0.4-0.7 mm wide,
glabrous to sparsely pubescent along the angles. Disc achenes 1.0—-2.4 mm
long, 0.4-0.8 mm wide, glabrous to sparsely pubescent. Chromosome
number nz = 16.

34 MADRONO [Vol. 24

Distribution (Fig. 2): Mexico, in moist soils from just north of the
city of Durango, south and east into the Federal District, Hidalgo,
Mexico, and Queretaro. Flowering Jun—Sep.

2a. HYBRIDELLA GLOBOSA (Ort.) Cass. var. GLoBosA.—Anthemis globosa
Ort., Nov. Rar. Plant. 46. 1797.—Hybridella globosa (Ort.) Cass.,
Dict. Sci. Nat. 22:86. 1821—Chiliophyllum globosum (Ort.) DC;
Prodr. 5:554. 1821.—Zaluzania globosa (Ort.) Sch.-Bip., Flora 44:
564. 1861.—TyYPE: not seen, presumably a Sessé collection in MA.

Variety globosa is characterized by its glabrous or hispidulous stems,
hispidulous to strigose lower leaf surface, acute ultimate leaf segments,
more than 2 mm long, linear-lanceolate receptacular pales, and the inner
series of involucral bracts, which are pubescent over the entire abaxial
surface. This variety is found growing on moist soils of the Federal Dis-
trict, Hidalgo, Mexico, and Queretaro.

Specimens examined: Distrito FEDERAL: Valley of Mexico, champs
incultes, 12 Jun 1865, Bourgeau 385 (GH, US); Valley of Mexico, 24
Jun 1887, Pringle 2925 (GH); low meadows, Valley of Mexico, 27 Jun
1890, Pringle 3204 (F, GH, LL, MO, NY, UC, US); damp meadows,
Valley of Mexico, 7300 ft, 25 Jun 1897, Pringle 7440 (US); wet mead-
ows, Valley of Mexico, 7300 ft, 7 Jun 1901, Pringle 9395 (GH, MO,
US); near Tlalnepantla, 20 Jun 1901, Rose & Hay 5247 (US); near
Tacuba, 30 Jul 1901, Rose & Hay 5816 (US). Hipatco: Real del Monte,
Jul 1946, Martinez 15308 (MO); bosque bajo ladera calcarea, C. de los
Pitos, 2690 m, 22 Jul 1951, Matuda 21494 (NY); swales and ditches,
vicinity of Tulancingo on road to Pachuca, 13 May 1947, Moore 2811
(MICH); ditches and roadsides between Tepetates and Acopinalco on
road from Tepeapulco to Apan, 21 Jul 1947, Moore 3443 (GH); Telles,
21 Sep 1910, Orcutt 4139 (F, GH, MO); along rt. 85 ca 20 mi N of
road to the pyramids (Rt. 30), 10 Aug 1961, Powell & Edmondson 597
(F, LL); Pachuca, Jul 1903, Purpus 77 (MO, UC, US); Sierra de Alca-
parrosa, 5 km al NW de Tepozolan, 29 Jul 1971, Rzedowski 28247
(LL, NY, US); 2 mi W of Tulancingo, 29 Aug 1965, Torres 1709
(MICH); N shore of Laguna de Apan, 4 km NW of Apan, 19 Jul 1966,
West R-12 (MICH). Mexico: near Tlalnepantla, 6 Jul 1905, Rose et al.
8417 (US); Mexico, Schmitz 2246 (GH). QUERETARO: 10 mi E of
Palmillas on Mexico 45, 14 Jul 1971, Sanderson 258 (LL).

2b. Hybridelila globosa (Ort.) Cass. var. myriophylla (Sch.-Bip.) Ol-
sen, comb. nov.—Zaluzania myriophylla Sch.-Bip., Flora 44:565.
1861.—Zaluzania globosa var. myriophylla (Sch.-Bip.) W. M. Sharp,
Ann. Missouri Bot. Gard. 22:103. 1935. Type: Mexico, Aguascali-
entes, low places, Hartweg 111 (Isotype: GH! ).

1977 | OLSEN: HYBRIDELLA ois)

Variety myriophylla (Fig. 1 a—d) is characterized by hirsute stems,
hispid lower leaf surfaces, rounded or obtuse ultimate leaf segments less
than 2 mm long, oblanceolate receptacular pales, and the inner series of
involucral bracts, which are pubescent only along the margins. This
variety is a more northern element, being sympatric with variety globosa
only in Hidalgo and the Federal District. Where sympatric, the two
varieties remain distinct with no mixed populations found and no evi-
dence of hybridization.

Specimens examined: AGUASCALIENTES: 31 mi N of Aguascalientes, 24
Aug 1953, Manning & Manning 531252 (GH). DURANGO: in moist low
soil about 16 mi NE of Durango, rt 31, 25 Jul 1958, Correll & Johnston
20163 (LL); 8-9 mi NE of Ciudad Durango near Rio Mesquital, 1 Oct
1948, Gentry 8568 (GH, MICH, UC); in the vicinity of Durango, Apr—
Nov 1896, Palmer 307 (F, GH, MO, NY, UC, US); in grassland 5 mi
SE of Victoria, 26 Aug 1939, Shreve 9172 (ARIZ, GH). Hwatco: 3 km
W of Tezontepec, 14 Jun 1963, Rzedowski 16718 (MICH, US). JAttsco:
near km 57, just E of the Aguascalientes state line, road from Ojuelos,
ca 13 mi W of Paso de la Troje, 16 Aug 1958, McVaugh 17053 (LL,
MICH, NY, US). Mexico: Tlalnepantla, 15 May 1904, Pringle 13092
(ARIZ, F, LL, MO, US); San Juan Teotihuacan, 1 Jun 1973, Rzedow-
ski 30174 (MICH). San Luts Potosi: region of San Luis Potosi, 1878,
Parry & Palmer 527 (F, GH, MO, NY, US); Laguna Seca, km 20 carre-
tera San Luis Potosi-Antigua Morelos, 30 Aug 1955, Rzedowski 6302
(LL); San Miguelita, Sep 1876, Schaffner 345 & 770 (GH, NY, US).
ZACATECAS: 2 mi SW of Zacatecas-San Luis Potosi state line along hwy
80, 23 Jul 1969, Bierner & Turner 116 (LL, NY); 6 mi W of Ojo Cali-
ente, 6-8 Sep 1938, Johnston 7454 (GH); 9 mi S of Fresnillo, 20 Aug
1956, Linsdale 56-F1 (UC); 9 mi S of Fresnillo, 9 Aug 1954, Linsley,
MacSwain, & Smith 1 (LL, UC); 15 mi N of Zacatecas on hwy 45, in
La Joya, 3 Sep 1975, Olsen 265 (LL); Frio, 23 Aug 1934, Pennell 18115
(US); damp hollow plains of Calera Station, 1 Sep 1904, Pringle 8914
(F, GH, LL, MO, NMC, NY, UC, US).

ACKNOWLEDGMENTS
I thank Dr. B. L. Turner for critically reading and discussing the
manuscript and the curators of the following herbaria for loans of ma-
terials: ARIZ, F, GH, LL, MICH, MO, NMC, NY, UC, US.

LITERATURE CITED

Butterwick, M. L. 1975. A systematic treatment of series Pinnatilobatae of
Viguiera. M.A. Thesis, Univ. Texas, Austin.

Cassini, A. 1821. Dict. Sci. Nat. 22:86. (Reproduced in R. M. King and H. D.
Dawson, 1975, Cassini on Compositae Vol. 1, 454. Oriole Editions, Inc., Sentry
Press, New York.)

OLsEeNn, J. S. 1977. A systematic study of the genus Zaluzania (Asteraceae: Heli-
antheae). Ph.D. Thesis, Univ. Texas, Austin.

36 MADRONO [Vol. 24

PowELtL, A. M. and B. L. Turner. 1963. Chromosome numbers in the Compositae
VII. Additional species from the southwestern US and Mexico. Madrofio
17:128—140.

Rosrnson, B. L. and J. M. GREENMAN. 1899. Revision of the genera Montanoa,
Perymenium and Zaluzania. Proc. Amer. Acad. Arts 34:507-534.

SHaArP, W. M. 1935. A critical study of certain epappose genera of the Heliantheae-
Verbesininae of the natural family Compositae. Ann. Missouri Bot. Gard.
22°51=153.

StuEssy, T. F. 1976. A revised subtribal classification of the Heliantheae. In V. H.
Heywood, J. B. Harborne, and B. L. Turner, eds. The biology and chemistry
of the Compositae. Academic Press, London (in press).

TurRNER, B. L. 1976. Fossil history and geography. Jn V. H. Heywood, J. B. Har-
borne, and B. L. Turner, eds. The biology and chemistry of the Compositae.
Academic Press, London (in press) .

Yates, W. F. 1967. Taxonomic studies in the genus Heliomeris (Compositae). Ph.D.
Thesis, Indiana University.

A REVISION OF LINANTHUS SECT. SIPHONELLA
(POLEMONIACEAE)

ROBERT PATTERSON
Department of Biology, Occidental College, Los Angeles 90041

Linanthus comprises nearly 40 species and is one of the larger genera
in the Polemoniaceae. Four species are suffrutescent perennials, and the
rest are annuals. The genus is distributed throughout much of western
North America, with one annual species indigenous to Chile. In spite of
the size and relatively widespread distribution of the genus, it is one of
the least examined in the family from a taxonomic viewpoint.

The four perennials constitute one of the least understood groups in
Linanthus. These plants were first described by Gray (1870) as two
species, Gilia nuttallai and G. floribunda. Milliken (1904) recognized
these two taxa as members of Linanthus on the basis of their palmately-
lobed leaves. A number of subsequent authors (McMinn, 1939; Munz,
1958; Grant, 1959) were basically in accord with Milliken, while others
regarded this complex as belonging in related genera such as Leptodac-
tylon (Rydberg, 1906; Jepson, 1925; Tidestrom, 1935) Navarretia
(Kuntze, 1891), Siphonella (Heller, 1912; Jepson, 1943), and Linantha-
strum (Ewan, 1942; Wherry, 1945) or retained it in its original genus,
Gilia (Brand, 1907). In addition to these varied generic interpretations, a
number of species, subspecies, varieties, and forms have been recognized
and named. These will be considered in greater detail in the taxonomic
treatment that follows.

1977] PATTERSON: LINANTHUS 37

The perennial species of Linanthus grow predominantly in montane
and subalpine localities in many areas in the western United States and
northern Mexico (Fig. 1). The distribution of these species is a series of
geographically isolated populations corresponding roughly to the moun-
tain ranges. The perennials are characterized morphologically by a
suffrutescent habit, opposite, palmately-lobed leaves, calyces with a nar-
row hyaline membrane between the lobes, and funnelform corollas with
stout tubes. They exhibit little morphological variation within any popu-
lation, but there are populations that differ significantly from each other
in one or more characteristics, such as leaf pubescence, calyx pubescence
(Fig. 2 a-c), leaf lobe shape (Fig. 2 d-f), number of lobes per leaf,
pollen grain diameter, and seed size. In addition, some populations con-
sist only of tetraploids, while others consist only of diploids.

Linanthus laxus (Vasey & Rose) Wherry, an annual, is closely re-
lated to the perennials and is characterized by the same calyx and corolla
morphology. It is a poorly known species and very few specimens are
present in herbaria. Its geographical range is restricted to the coastal
canyons near San Quintin, Baja California (Fig. 1). It does not present
the same kinds of taxonomic problems that are found in the perennial
species; however, it is included in this treatment because of its close
relationship to the perennials.

Grant (1959) placed the perennial species of Linanthus, plus L.
laxus, in sect. Siphonella (A. Gray) V. Grant, but no attempt to elu-
cidate the relationships of the taxa within the section was made. A
revision of sect. Szphonella is presented here. It is based on the examina-
tion of specimens in the field and over 2400 herbarium specimens. Par-
ticular attention was paid to characters whose expressions vary from
population to population.

Representatives of each morpho-geographical group were grown in
the greenhouse to establish whether a genetic basis exists for the charac-
ter differences used to distinguish taxa, especially leaf and calyx pubes-
cence and leaf lobe shape. Each of the diagnostic characters examined
was retained by each taxon when grown under uniform conditions for at
least two growing seasons, and progeny obtained from self-fertilizations
maintained these parental features.

Chromosome numbers were counted for eight of the nine taxa con-
sidered in this treatment. Six were diploid with a somatic number of
2n = 18, two were tetraploid, 2n = 36. Cytological material for L. flori-
bundus subsp. hallii was not available. Voucher specimens are cited in
Table 1 and deposited at UCSB.

Pollen grain diameters were measured for all nine taxa; they range
from 23 to 28 um in diploids, and from 33 to 38 um in tetraploids. The
diameter of pollen grains was used to estimate the chromosome number
of specimens from regions where live material was unavailable. On this
basis, L. floribundus subsp. hallii is considered to be diploid.

38

Oo + Pp OO 89

Jt Ue ie Ie atts ite Gees

{te

MADRONO [ Vol. 24

nuttallii subsp. tenuilobus (2n=18)
pachyphyllus (2n=36)

floribundus subsp. floribundus (2n=18)
floribundus subsp. glabrus (2n=18)
floribundus subsp. hallii (2n=18)
melingii (2n=36)

laxus (2n=18)

Fic. 1. Geographical distribution of Linanthus sect. Siphonella.

1977] PATTERSON: LINANTHUS 39

jane

SS Te ri =
rs f
= ita 4
. \ Py yt

SIS

fal
a.
eee E

ARGON ES
5 Mane:

er. cis yy
WERE:
ae Seen

NH i
fe ir a
arte. ate (iA
\\ Ea $34 co
Be, WS iy yiad
AV ENG % 4, i¢ oy ~
, aN oh Sa Ki oe
¥\ = ON Say “ay
lS “Sy, fe 47
3 iS re “ah ff i Ke

ia

Lx \
i 4 py
ti , INS inthe
£94 Las | ate: 1:
Ye t. § 4 ei a 4 ae Nh g \ 2 oe x ft: iy
See, ea LAF Aye ed pe ae
STR rea fyy SA VY 68
Or te iy LOY MA bag BS , A
ae a pt, ; a GY 4 Fes
a ae Me hy Ages eatin gi ied \
ese Ta ac Od fg oT one anid

Fic. 2. a—c. Calyx pubescence in Linanthus sect. Siphonella. a, L. floribundus
subsp. glabrus (Patterson 1021) showing glabrous calyx. b, L. nuttallii subsp. nuttallii
(Patterson 1032) showing lightly pubescent calyx. c, L. nuttallii subsp. pubescens
(Patterson 1012) showing densely pubescent calyx. d-f, Variation in leaf lobe shape
in sect. Siphonella. d. L. nuttallii subsp. tenuilobus (Patterson 1009). e. L. nuttallii
subsp. nuttallii (Patterson 1030). f. L. pachyphyllus (Patterson 1001), also found
in L. melingii. Drawings by B. Tanowitz.

40 MADRONO {Vol. 24

TABLE 1. CHROMOSOME COUNTS FOR LINANTHUS SECT. SIPHONELLA.

L. nuttalla (A. Gray) Greene ex Milliken subsp. nuttalliti; 2n = 18.
COLORADO: Eagle Co.: Tennessee Pass, Patterson 1031. Lake Co.: 3 mi SW
of Climax, Patterson 1030.
IDAHO: Blaine Co.: Trail Creek Summit, Patterson 1033.
WYOMING: Lincoln Co.: Allred Springs Campground, Patterson 1032,
L. nuttalla (A. Gray) Greene ex Milliken subsp. pubescens R. Patterson; 2” = 18.
CALIFORNIA: Mono Co.: Junction of White Mountain and Silver Canyon
Roads, Patterson 1013.
NEVADA: Clark Co.: Mary Jane Falls Campground, Patterson 1012.
L. nuttalli (A. Gray) Greene ex Milliken subsp. tenuidlobus R. Patterson; 2n = 18.
ARIZONA: Gila Co.: Camp Verde-Strawberry Road, one mi W of Straw-
berry, Patterson 1009.
L. pachyphyllus R. Patterson; 2” = 36.
CALIFORNIA: Inyo Co.: Bishop Creek near South Lake Road, Patterson
1001; two mi E of Onion Valley, on Independence Road, Patterson 1002.
Kern Co.: one mi SW of Piute Mountain, Smith sn. Mono Co.: Lee Vining
Creek, just S of ranger station, Patterson 1000; Rock Creek, Meyers s.n.
L. floribundus (A. Gray) Greene ex Milliken susbp. glabrus R. Patterson; 2n = 18.
CALIFORNIA: Riverside Co.: Santa Rosa Mountains, Meyers s.n. San Diego
Co.: Four mi S of Tenaja Campground, on road to Fallbrook, Patterson
1020.
L. floribundus (A. Gray) Greene ex Milliken subsp. glabrus R. Patterson; 2m = 18.
CALIFORNIA: San Diego Co.: nine mi NE of U.S. Highway 80, on county
road S-1, Patterson 1011.
L. melingi (Wiggins) V. Grant; 2n = 36.
BAJA CALIFORNIA: The observatory, Sierra San Pedro Martir, Haller s.n.
L. laxus (Vasey & Rose) Wherry; 2” = 18.
BAJA CALIFORNIA: Arroyo Socorro, Moran 19330, Moran & Reveal 20281,
Moran 20593.

TAXONOMY
LINANTHUS Benth. sect. SIPHONELLA (A. Gray) V. Grant, Natural his-
tory of the Phlox family 106. 1959.—Giulia sect. Siphonella A. Gray,
Proc. Amer. Acad. Arts 8:266. 1870 (Based on a generic name in
Nuttall’s herbarium).—Siphonella (A. Gray) Heller, Muhlenbergia
8:57. 1912.—TypeE: Linanthus nuttallu (A. Gray) Greene ex Milliken.
Suffrutescent perennials and one annual; leaves opposite, mostly palm-
ately lobed; inflorescence ranging from dense, subcapitate clusters to
sub-panicles, rarely solitary; pedicels present or absent; calyx herba-
ceous with narrow hyaline membrane connecting the lobes to about one-
half their length; corolla funnelform, tube stout, equalling to twice as long
as the calyx; stamens inserted on corolla throat; pollen grains 23-38 wm
in diameter, yellow; style divided into three lobes, each as long as the
united style body, the point of trifurcation stigmatic; fruit a three-
parted septicidal capsule; seeds smooth, light to dark brown, non-
mucilaginous.

1977} PATTERSON: LINANTHUS 41

Key to Linanthus sect. Siphonella
Plants suffrutescent perennials; mostly montane or subalpine, seldom
below 500 m.
Leaves palmately-lobed.
Leaves predominantly 5-lobed.

Leaf lobes linear to linear-lanceolate.

Leaf lobes linear; corolla tube equalling calyx; seeds 1.5—2.0
mm long.

Leaf lobes and calyces subglabrous to lightly pubescent,
never densely so; wide ranging in Rocky Mountains,
eastern Great Basin ranges, ranges of Washington, Ore-
gon, and northern California, and the Sierra Madre
Occidental of Mexico. la. L. nuttallii subsp. nuttalli

Leaf lobes and calyces densely hirtellous; dry mountain
ranges of Nevada and eastern California.

oe Ib. L. nuttalli subsp. pubescens

Leaf ienee mea lanceolate; corolla tube often exceeding the
calyx; seeds 2.5-3.5 mm long; Sierra Nevada. .

: iia! ee wi niin

Leaf fabes agetilhe falcon anountcine of central Arizona and
New Mexico. . . . . Ic. L. nuttalli subsp. tenuilobus

Leaves predominantly 3-lobed.

Leaves and calyces glabrous to moderately pubescent; plants usu-
ally robust, to over 3 dm high; mountain ranges of southern
California and Baja California.

Leaves and/or calyces subglabrous to moderately pubescent.
: 3a. L. floribundus subsp. floribundus
Weavcs mit calyces glabrous. 3b. L. floribundus subsp. glabrus

Leaves and calyces densely hirtellous; plants compact, often
matted; higher elevations of the Siema Juarez, Sierra San
Pedro Martir, and Sierra San Borja, Baja California.

ee . 4. L. melingu

Leaves met re aay een ‘epes of coast ranges of southern
California endl Baja California. 3c. L. floribundus subsp. hallii

Plants annual; occurring below 100 m; coastal areas near San Quintin,
Baja California. . . . ...... . . . +5. ZL. laxus

la. LINANTHUS NUTTALLI (A. Gray) Greene ex Milliken subsp. NUT-
TALLI, Univ. Calif. Publ. Bot. 2:54. 1904.—Gilia nuttallii A. Gray,
Proc. Amer. Acad. Arts 8:267. 1870.—Navarretia nuttallit (A. Gray)
Kuntze, Revisio generum plantarum 2:432. 1891.—Leptodactylon
nuttalli (A. Gray) Rydberg, Bull. Torrey Bot. Club 33:149. 1906.—
Siphonella nuttallii (A. Gray) Heller, Muhlenbergia 8:57. 1912.—

42 MADRONO [Vol. 24

Linanthastrum nuttallu (A. Gray) Ewan, J. Wash. Acad. Sci. 32:139.
1942.—-LEcTOTYPE (here designated): R. Mts. Bear R. Hills, Nuttall
s.n. (PH, seen only as photograph; isolectotype: GH! ).

Gilia nuttallu A. Gray var. montana Brand, Pflanzenreich 4(250):125.
1907.—SyntypeEs: Allen 119, Hall 941, Parish 3686, and Purpus
1443.

Gilia nuttalli A. Gray var. parviflora Brand, Pflanzenreich 4(250):125.
1907.—SYNTvYPES: Baker 1053, Cusick 255, and Jones 1106.

Perennial herb from woody base, stems erect, 1-4 dm high, slightly
puberulent; leaves mostly 5-partite into linear lobes, 6-32 mm long,
subglabrous to moderately pubescent; flowers mostly sessile in upper
axils, in dense subcapitate clusters; calyx narrow-campanulate, 8-10 mm
long, tube 2—4 mm long, lobes lance-subulate, slightly pubescent; corolla
12-16 mm long, tube 8-10 mm long, white, pubescent on outer surface,
throat short, yellow, lobes 5-8 mm long, oval to oblanceolate; stamens
inserted at base of throat or just below, glabrous, barely exserted to
included; pollen 23—28 um in diameter; chromosome number 2” = 18.

Distribution (Fig. 1) and phenology: Cascade range from central
Washington to northern California; high ranges of eastern Washington,
Oregon, and northeastern California; Rocky Mountains from central
Idaho to southern Colorado; higher ranges of Utah; Sierra Madre Occi-
dental of Chihuahua and Durango; Sierra de la Victoria, Baja Cali-
fornia. Flowering July to August (June to September). The populations
of L. nuttallii subsp. nuttalli from the Sierra Madre Occidental and the
Sierra de la Victoria present an interesting disjunction from the majority
of populations of this subspecies. The nearest population of subsp. nué-
tallit is in southern Colorado. The Mexican populations are poorly repre-
sented in herbarium collections, but the regions within the area of dis-
junction (e.g., Arizona and New Mexico) are well known botanically;
hence, the disjunction is not simply the result of insufficient collecting.
Future studies on the nature and significance of this disjunction should
be rewarding from a systematic viewpoint after the range of L. nuttallu
subsp. nuttallii in Mexico is better known.

Representative specimens: CALIFORNIA: Little Lily Lake, Wheeler
3789 (GH, NY, UC, US, UTC). Cotorapo: Tennessee Pass, Clokey
3549 (CAS, DS, GH, MICH, NY, POM, TEX, US, UTC, WSU); Rab-
bit Ears Pass, Gooding 1540 (COLO, DS, GH, MO, NY, RM, UC, US).
IpAHO: Squaw Valley, Macbride & Payson 3388 (CAS, DS, GH, MO,
NY, POM, RM, UC, US); 3.5 mi SW Stanley Lake, Hitchcock & Muh-
lick 9635 (NY, RM, UC, UTC, WSU, WTU); Trinity Lake region,
Macbride 691 (DS, GH, MO, NY, RM, UC, US, WSU, WTU). NeEvapa:
Lamoille Terrace Ranger Station, Maguire & Holmgren 22166 (GH,
NY, UC, UTC). Orecon: near Cornucopia, Thompson 13353 (CAS,
COLO, GH, MO, NY, RSA, UC, UTC, WSU); along Musick Mine Road,

1977] PATTERSON: LINANTHUS 43

Dennis 2313 (ASU, CAS, DS, ID, NY, OSU, RSA, US, UTC, WSU,
WTU). Utan: Alta, Jones 1106 (ARIZ, GH, MICH, NY, OSU, RSA,
US, UTC, WSU). WasHINGcTON: alpine slopes of Mt. Wow, Thompson
12585 (CAS, DS, GH, MO, NY, POM, UC, US, WSU, WTU); Blue
Mountains, Piper 2419 (GH, WSU). Wvyominc: Teton Pass, Cronquist
831 (ID, UC, UTC). Baja Catirornia: La Laguna, Gentry 4372
(ARIZ, DS, MO, UC, US). CutHuanHua: SW Chihuahua, Palmer 398
(MO, NY, US).

1b. Linanthus nuttallii (A. Gray) Greene ex Milliken subsp. pubescens

R. Patterson, subsp. nov.—TyPe: California, Inyo County, White

Mountains, junction of White Mountain and Silver Canyon Roads,

3100 m, 2 Aug 1970, Patterson 1009 (Holotype: UC!; isotypes:

NY!, UCSB!).

Herba perennis, 1—4 dm alta; folia plerumque 5-partita, moderate vel
dense pubescentia, lobis linearis; flores sessiles racemis subcapitatis;
calyces moderate vel dense pubescentes; chromosomatum numerus
2n = 18.

Distribution (Fig. 1) and phenology: Dry mountain ranges of central
and southern Nevada and the White and Inyo Ranges of California,
mostly above 2800 m. Flowering July to August (June to September).

Representative specimens: CALIFORNIA: Wyman Creek, Duran 3048
(ARIZ, CAS, DS, GH, MICH, MO, NY, POM, RM, SD, UC, US,
UTC); 1.5 mi N of New York Butte, Alexander & Kellogg 3083 (ARIZ,
DS, MO, UC, US, UTC). Nevapa: Charleston Park, Clokey 7253
(ARIZ, CAS, COLO, DS, GH, ID, MICH, MO, NY, OSU, POM, RM,
TEX, UC, US, UTC, WSU); 2 mi N of Toiyabe Dome, Hitchcock &
Martin 5598 (DS, POM, UC, UTC, WSU).

Ic. Linanthus nuttallii (A. Gray) Greene ex Milligen subsp. tenui-
lobus R. Patterson, subsp. nov.—Typer: Arizona, Gila County, one
mile east of Strawberry on Camp Verde-Strawberry Road, 1700 m, 22
Aug 1970, Patterson 1013 (Holotype: UC!; isotypes: NY!, UCSB!).
Herba perennis, 1-4 dm alta; folia plerumque 5-partita, parce pubes-

centia, lobis tenuibus; flores sessiles racemis subcapitatis; calyces parce

vel moderate pubescentes; chromosomatum numerus 2” = 18.
Distribution (Fig. 1) and phenology: Mountain ranges of central

Arizona and New Mexico, and in the Santa Catalina Mountains of south-

ern Arizona, mostly above 1500 m. Flowering July to August (June to

September).

Representative specimens: ARIZONA: Strawberry Valley, Toumey 201
(ARIZ, DS, GH, POM, US); 3 mi S of Stray Horse Camp, Parker &
McClintock 7662 (ARIZ, CAS, COLO, MO, RSA, UC). New Mexico:
Mogollon Mountains, Rusby 272 (MICH, MO, NY); White Moun-
tains, Wooton 205 (DS, GH, MO, NY, POM, RM, UC, US).

44 MADRONO [Vol. 24

2. Linanthus pachyphyllus R. Patterson, sp. nov.—TyPeE: California,
Mono County, Lee Vining Grade, 9000 ft, 30 Jun 1934, Keck, Abrams,
& Blake 2872 (Holotype: UC!; isotypes: CAS!, DS!, GH!, POM!,
Us!).

Herba perennis e basi lignea; caules erecti, 1-4 dm alta, moderate vel
dense pubescentes; folia plerumque 5—partita, moderate vel dense pubes-
centia, lobis lineari-oblanceolatis, 3-30 mm longa; flores sessiles in axiles
superioribus ramorum fascientes racemos dense subcapitatos; calyx
superioribus ramorum fascientes racemos dense subcapitatos; calyx tenui-
campanulatus, 8-10 mm longus, dense pubescens, tubo 3—5 mm longo,
lobis lanci-subulatis; corolla 12-18 mm longa, tubo 8-12 mm longo,
albo, externe pubescente, fauce brevi, flavo, lobae 5-8 mm longi, albi,
ovali-oblanceolati; stamina basi faucis vel modo infra inserta, glabrata,
vix-moderate exserta; pollinis granula 33-38 um diametro; semina fulva,
2.5-3.5 mm longa; chromosomatum numerus 2” = 36.

Distribution (Fig. 1) and phenology: Linanthus pachyphyllus is re-
stricted to the Sierra Nevada of California and western Nevada. Its
southernmost limit is near Piute Mountain in Kern County, California,
and its northernmost limit is near Mt. Rose in Washoe County, Nevada.
Flowering June to July (May to September).

Representative specimens: CALIFORNIA: Lee Vining, Benson 3767
(NY, POM, RM, UC, US); Sabrina Lake, Jones s.n. (CAS, DS, GH);
Farewell Gap, Baker 4536 (CAS, DS, GH, MO, NY, POM, US). NE-
vapaA: Zephyr Point, Mason 12161 (ARIZ, CAS, COLO, DS, GH, ID,
MICH, MO, NY, POM, RM, UC, US, UTC, WSU).

3a. LINANTHUS FLORIBUNDUS (A. Gray) Greene ex Milliken subsp.
FLORIBUNDUS, Univ. Calif. Publ. Bot. 2:55. 1904.—Gilia floribunda
A. Gray, Proc. Amer. Acad. Arts 8:267. 1870.—Linanthus nuttallu
(A. Gray) Greene ex Milliken var. parviflora Brand subvar. flori-
bunda (A. Gray) Brand, Pflanzenreich 4(250):125. 1907.—Lepto-
dactylon nuttallii (A. Gray) Rydberg var. floribundum (A. Gray)
Jepson, Manual of the flowering plants of California 808. 1925.—
Leptodactylon floribundum (A. Gray) Tidestrom, Proc. Biol. Soc.
Wash, 48:42. 1935.—Linanthus nuttallii (A. Gray) Greene ex Milli-
ken var. floribundus (A. Gray) McMinn, Illustrated manual of Cali-
fornia shrubs 446. 1939.—Linanthastrum nuttalli (A. Gray) Ewan
subsp. floribundum (A. Gray) Ewan, J. Wash. Acad. Sci. 32:139.
1942.— Siphonella floribunda (A. Gray) Jepson, A flora of California
3:218. 1943.—Linanthastrum floribundum (A. Gray) Wherry, Amer.
Midl. Naturalist 34:218. 1945.—Linanthus nuttalli (A. Gray) Greene
ex Milliken subsp. floribundus (A. Gray) Munz, Aliso 4:96. 1958.—
LECTOTYPE (here designated): California, Coulter 454 (GH!).

Linanthus saxiphilus A. Davidson, Bull. S. Calif. Acad. Sci. 19:10. 1920.
—Linanthastrum floribundum (A. Gray) Wherry forma saxiphilum

1977] PATTERSON: LINANTHUS 45

(A. Davidson) Wherry, Amer. Midl. Naturalist 34:386. 1945._TvPE:
California, San Bernardino County, Seven Oaks, Jul 1901, Davidson
2242 (Holotype: US!).

Suffrutescent perennial, 1-4 dm high; leaves mostly 3-partite, lobes
linear, 6—24 mm long; inflorescence an open subcapitate panicle to nearly
solitary; flowers mostly pedicellate with pedicels 1-6 mm long; calyx
puberulent, narrow-campanulate, 8-9 mm long, lobes lance-subulate;
pollen grains 23—28 «wm in diameter; chromosome number 2” = 18.

Distribution (Fig. 1) and phenology: San Bernardino, Santa Mar-
garita, and Santa Rosa Mountains of southern California, rarely in ad-
jacent ranges. Flowering April to June (March to August).

Representative specimens: BAJA CALIFORNIA: Cerro el Sauco, Moran
8090 (ARIZ, RSA, SD). CALtForNIA: near Seven Oaks, Parish 3686
(ARIZ, CAS, DS, GH, NY, UC, US); Laguna Mountains, Spencer 958
(GH, NY, POM).

3b. Linanthus floribundus (A. Gray) Greene ex Milliken subsp. gla-
brus R. Patterson, subsp. nov.—T ype: California, San Diego County,

9 mi NE of U.S. highway 80 on county road S-1, 1700 m, 3 Apr 1971,

Patterson 1011 (Holotype: UC! isotypes: NY!, UCSB!).

Herba perennis, 1—4 dm alta; caules glabrati vel leviter puberuli; folia
plerumque 3-partita, glabrata, lobis linearis vel filiformibus; calyx gla-
bratus; chromosomatum numerus 2” = 18.

Distribution (Fig. 1) and phenology: Most of the high mountain
ranges of southern California excluding the San Bernardino Mountains;
the higher ranges of northern Baja California, mostly below 200 m.
Flowering April to June (March to July).

Representative specimens: BAJA CALIFORNIA: Rancho El Barril, Mo-
ran 13840 (ARIZ, RSA, SD, UC). Catirornia: Descanso, Epling s.n.
(COLO, DS, MO, NY, RSA, UC).

3c. LINANTHUS FLORIBUNDUS (A. Gray) Greene ex Milliken subsp.

HALLII (Jepson) H. L. Mason, im Abrams, Ill. Fl. Pacific States 3:431.

1961.—Siphonella floribunda (A. Gray) Jepson var. hallii Jepson,

Flora of California 3:218. 1943.—Linanthastrum floribundum (A.

Gray) Wherry forma hallii (Jepson) Wherry, Amer. Mid]. Naturalist

34:386. 1945.—Type: California, San Diego County, Coyote Canyon,

in the lower Sonoran zone, 600 ft, Apr 1902, Hall 2767a (Holotype:

JEPS!; isotype: UC!).

Perennial herb, 1-4 dm high; stems slightly puberulent; leaves entire,
linear, rarely 3-partite on lower branches, glabrous; calyx moderately
pubescent to glabrous; pollen grain diameter 23-28 um.

Distribution (Fig. 1) and phenology: Dry washes below 700 m in
southern Santa Rosa Mountains, Riverside and San Diego Counties,
California. Flowering late March and May. /

46 MADRONO [Vol. 24

Representative specimens: CALIFORNIA: Rockhouse Canyon, Jaeger
1088 (DS, POM); Martinez Canyon, Davidson sn. (UC); Thermal,
Davidson s.n. (DS); Elder Canyon, Woglum 3041 (RSA); canyon east
of Clark Dry Lake, Buechner 694 (RSA); Palm Canyon, Gander 1277
(SD).

4. LINANTHUS MELINGII (Wiggins) V. Grant, Natural history of the
Phlox family 107. 1959.—Le ptodactylon melingu Wiggins, Contr. Dud-
ley Herb. 1:173. 1933.—Linanthastrum melingu (Wiggins) Wherry,
Amer. Midl. Naturalist 34:386. 1945——Type: Baja California, in
gravelly soil along the margins of an open meadow at La Encantada,
Sierra San Pedro Martir, 2100 m, 18 Sep 1930, Wiggins & Demaree
4884 (Holotype: DS! isotype: UC!).

Low suffrutescent perennial, branches 3-15 cm long, often forming
dense mats to 3 dm in diameter; leaves 3-partite, lobes 3-6 mm long,
linear-lanceolate, densely hirtellous; flowers sessile in subcapitate clus-
ters, rarely solitary; calyx narrow campanulate, 4-5 mm long, hirtellous,
tube 1 mm long, lobes 3-4 mm long; corolla white with yellow throat,
ofien tinged with purple, tube 4-5 mm long, lobes 4-5 mm long, ovate;
stamens inserted in throat, barely exserted; pollen grains 33-38 wm in
diameter; chromosome number 2” = 36.

Distribution (Fig. 1) and phenology: High elevations (above 2000 m)
in the Sierra Juarez, Sierra San Pedro Martir, and Sierra San Borja,
Baja California. Flowering July to August (June to September).

In Wiggins’ original description of Leptodactylon melingu, he charac-
terized this species in part as having wingless seeds in differentiating it
from Leptodactylon nuttalli; however, the examination of numerous
herbarium specimens indicates that none of the members of sect. Sipho-
nella possess winged seeds (Patterson, 1975).

Munz (1959) cited Linanthus melingi as occurring in the White
Mountains of California and Nevada, and in the Toiyabe Mountains of
Nevada. There are plants that have the same habit as L. melingu in
these localities; however, these individuals are characterized by 5-partite
leaves and are recognized herein as L. nuttallii subsp. pubescens. In addi-
tion, pollen grain diameters of the Toiyabe and White Mountains speci-
mens vary between 23—28 wm, indicating that they are diploids.

Representative specimens: BajA CALIFORNIA: 8 mi N of Vallecitos,
Breedlove 16306 (CAS, MICH); Hansen’s Ranch, Orcutt 128 (UC,
US):

5. LINANTHUS LAXUS (Vasey & Rose) Wherry, Amer. Midl. Naturalist
34:386. 1945.—Gulia laxa Vasey & Rose, Proc. U. S. Natl. Mus. 11:
531. 1889.—TypE: Baja California, San Quintin Bay, Jan 1889,
Palmer 650 (Holotype: US!; isotype: UC!).

Linanihus wigginsti Mason, Madrono 6:203. 1942.—TyPeE: Baja Cali-
fornia, southern end of Santa Maria Plains, 5 Feb 1935, Wiggins 7557

1977]. PATTERSON: LINANTHUS 47

(Holotype: DS!; isotype: UC! )

Slender annual, 3-15 cm high; stems erect, glabrous to slightly pubes-
cent; leaves 3-partite (rarely 5-partite), lobes 5-20 mm long, linear,
glabrous to slightly pubescent; inflorescence solitary to subpaniculate,
pedicels 2-16 mm long; calyx narrow-campanulate, 4-5 mm long, tube
1 mm long, lobes 3-4 mm long, glabrous to moderately pubescent;
corolla tube equal to or slightly exceeding the calyx, white with yellow
throat, lobes 5 mm long, white, obovate; staments inserted below throat,
1.5-2.0 mm long, barely exserted; pollen grains 23—28 um in diameter;
chromosome number 2” = 18.

Distribution (Fig. 1) and phenology: Coastal canyons and plains,
Rio Santa Maria, San Quintin, and Arroyo Socorro, Baja California.
Flowering January to April.

Representative specimens: BAJA CALIFORNIA: Arroyo Socorro, Moran
19330 (SD, UCSB), Moran & Reveal 20281 (SD, UCSB), Moran 20593
(SD, UCSB); Santa Maria, Orcutt sn. (MO); north of San Quintin
Bay, Raven 12370 (UC).

a ACKNOWLEDGMENTS

Appreciation is extended to curators of the following herbaria for
loans of specimens: ARIZ, ASU, CAS, COLO, CSU, DS, F, GH, ID,
IDS, JEPS, MICH, MO, NY, OSU, PH, POM, RM, RSA, SD, TEX,
UC, UCSB, UNM, US, UT, UTC, WSU, WTU. Dr. Reid Moran has
been particularly helpful in providing specimens of Linanthus laxus. I
am grateful to D. M. Smith and S. L. Timbrook for reviewing the manu-
script; to the Academic Senate of UCSB for a grant-in-aid of research;
and to those who provided collections for me throughout this study.

LITERATURE CITED

Branp, A. 1907. Polemoniaceae. Jn A. Engler, Das Pflanzenreich 4(250).

Davinson, A. 1920. Linanthus saxiphilus. Bull. S. Calif. Acad. Sci. 19:10.

Ewan, J. 1942. Linanthastrum, a new West American genus of Polemoniaceae. J.
Wash. Acad. Sci. 32:138-141.

Grant, V. 1959. Natural history of the Phlox family. Systematic Botany. M. Nijhoff,
The Hague.

Gray, A. 1870. Revision of the North American Polemoniaceae. Proc. Amer. Acad.
Arts 8:247-282.

HEtter, A. 1912. The flora of the Ruby Mountains—V. Muhlenbergia 8:49-58.

Jepson, W. L. 1925. Manual of the flowering plants of California. A.S.U.C. Book-
store, Berkeley. "

. 1943. Polemoniaceae. Jn A flora of California 3:131-222. A.S.U.C. Book-

store, Berkeley.

Kuntze, O. 1891. Polemoniaceae. Jn Revisio generum plantarum 2:432-434.

McMinn, H. E. 1939. Illustrated manual of California shrubs. Univ. California
Press, Berkeley.

MruikeEN, J. 1904. A review of California Polemoniaceae. Univ. Calif. Publ. Bot.
2:1-71.

Muwz, P. A. 1958. California miscellany IV. Aliso 4:87—-100.

————. 1959. A California flora. Univ. Califcrnia Press, Berkeley.

48 ’ MADRONO [Vol. 24

PATTERSON, R. 1975. The systematics of Linanthus section Siphonella. Ph.D. Thesis,
Univ. California, Santa Barbara.

RypBerc, P. A. 1906. Studies on the Rocky Mountain flora—XVI. Bull. Torrey
Bot. Club 33:149.

TIDESTROM, I. 1935. New Arizona plant names. Proc. Biol. Soc. Wash. 48:42.

Wuaerry, E. T. 1945. Two Linanthoid genera. Amer. Midl. Nat. 34:381-387.

A NEW SUBSPECIES OF HULSEA VESTITA (ASTERACEAE)

DIETER H. WILKEN
Department of Botany and Plant Pathology
Colorado State University, Fort Collins 80523

The Hulsea vestita complex of California and Nevada comprises five
currently recognized subspecies of self-incompatible, densely lanate,
caespitose, herbaceous perennials with a dimorphic, quadripartite, pale-
aceous pappus and a uniform diploid complement of 38 chromosomes
(Wilken, 1975). Of particular interest within this complex is the occur-
rence in the western Transverse Ranges of southern California of popu-
lations that have been variously treated in several floristic works. Specific
epithets applied to these populations include H. callicarpha (Hall) Ryd-
berg, H. parryi A. Gray, and H. vestita A. Gray (Rydberg, 1914; Jepson,
1951; Keck, 1959; Ferris, 1960; Munz, 1974). Recently, I postulated
(Wilken, 1975) that these populations represent an intergrading complex
involving ssp. vestita and ssp. parryi. This hypothesis was based on
observations of leaf morphology and ray flower number. Further study
has revealed, however, that populations from the western Transverse
Ranges, particularly the San Gabriel Mountains, display discordant
variation with respect to presently circumscribed taxa. It is the purpose
of this paper to discuss the relationships and status of these problemati-
cal populations by a numerical analysis of morphological variation and
studies of synthetic hybrids.

MATERIALS AND METHODS

Field studies were begun in 1970 and seeds were collected to serve as
a source of parental strains for an extensive crossing program. Methods
with respect to crossing attempts and studies of meiosis and hybrid fer-
tility (= pollen stainability and seed set) were as described elsewhere
(Wilken, 1975). During the spring and summer of 1972, a series of
populations, derived from 18 natural populations, were grown from seed
to flowering maturity in a common garden at Occidental College, Los
Angeles, California. These garden populations represented all recog-
nized subspecies of Hulsea vestita and six populations from widely sepa-

1977] WILKEN: HULSEA 49

rated sites in the San Gabriel Mts. (Table 1). Each of the garden
populations comprised 10-18 plants and represented progenies of at least
five randomly selected plants in each of the natural populations. Twenty
vegetative and floral characters were chosen for measurement (Table 2).
These characters have been shown to be the most useful in delimiting
taxa within the complex (Wilken, 1973). The data were standardized by
range and an unweighted analysis was performed, using Gower’s Corre-
lation Coefficient (Sneath and Sokal, 1973) and a program written for
the CDC 6400 computer. Each of the 18 garden populations was treated
as an OTU in the numerical analysis. A phenogram was constructed by
using single-linkage clustering. Although variation of leaf shape and
blade margin was valuable in assessing relationships, quantification of
these characters for the numerical analysis was not done. Variation of
leaves among garden populations is illustrated in Figure 1. Additional
morphological studies were made of natural population samples, syn-
thetic hybrids, and specimens borrowed from several herbaria (CAS,
GH, JEPS, MO, NY, POM, RM, RSA, UC, UTC, WTU). Common gar-
den and greenhouse populations were assigned the number correspond-
ing to vouchers of natural population samples.

RESULTS
A phenogram of relationships (Fig. 2) reveals the distinctiveness of
San Gabriel Mt. populations and supports the delimitation of accepted
infraspecific taxa of the Hulsea vestita complex. The lowest level of simi-
larity between OTUs representing a recognized taxon is 0.889 for ssp.
inyoensis, whereas the highest level is 0.945 between OTUs and 8200

TaBLe 1. Locations or NatTurAL POPULATIONS USED AS SEED SOURCES FOR THE
CROSSING PROGRAM AND GARDEN POPULATIONS. Plants of all 18 populations were
examined for chromosome number, which was uniformly 2” = 38. All collections,
with the exception of ssp. inyoensis, are those of the author (cited W). The two
collections of ssp. inyoensis were provided by J. Beatley. Vouchers are deposited
at CS.

California, Inyo Co., Cottonwood Meadows at head of Cottonwood Creek, W
4255, Los Angeles Co., Devil’s Backbone, S slope of Mt. San Antonio, W 11891;
E slope of Mt. Williamson, W 11890; 3 mi E of Chilao, W 11880; upper Tujunga
Canyon, 1.5 mi W of Highway 2, W 11879; along road to Mt. Pacifico, 2 mi E of
Mill Creek Summit, W 11878; near summit of Mt. Gleason, 5 mi W of Mill Creek
Summit, W 7946. Madera Co., near Rainbow Falls, Devil’s Postpile N. M., W 8202.
Mono Co., along Highway 120, 6.9 mi E of Highway 295, W 8200. Riverside Co.,
road to Toro Pk., 5.9 mi S of Highway 74, Santa Rosa Mts., W 7896; County Road
R1, 8.6 mi NW of Idyllwild, San Jacinto Mts., W 7844; County Road R1, 1 mi S
of Idyllwild, San Jacinto Mts., W 7858. San Bernardino Co., summit of Mt. San
Gorgonio, W 8703; ridge between Fish Creek and Big Meadows, San Bernardino
Mts., W 8703; Coon Creek, 3 mi SE of Highway 38, San Bernardino Mts., W 3047;
along Highway 38, 2 mi E of Heart Bar Park, San Bernardino Mts., W 3048.

Nevada, Nye Co., SW face of Rainier Mesa, Belted Range, Beatley 9376; top of
Rainier Mesa, Beatley 8836.

50 MADRONO [Vol. 24

TABLE 2. CHARACTERS USED IN THE NUMERICAL ANALYSIS. Actual values were
measured, recorded, and coded. Leaf and pubescence data were taken from the tenth
leaf of the basal rosette. Pubescence density was determined by counting the tri-
chomes within a 1 cm”* square placed over the midrib at the center of the blade.
Bract data were taken from the lowermost bract of the inflorescence. All capitulum
and floral data were taken from the first mature capitulum.

Plant height (cm); Leaf length (mm); Leaf width (mm); Bract length (mm) ;
Bract width (mm) ; Abaxial nonglandular pubescence density (number of trichomes
per cm’) ; Peduncle length (cm) ; Number capitula per plant; Number of phyllaries
per capitulum; Phyllary length (mm); Phyllary width (mm); Number of ray
flowers; Ray corolla length (mm); Number disc flowers; Disc corolla length
(mm); Achene length (mm); Narrow pappus pair length (mm); Broad pappus
pair length (mm); Achene pubescence length (mm).

Te eee PY VY

44 7896
L—- _ -vestita — Le 7858
L___ callicarpha ———X
Poca) cm
I | " 8703 | |
3041 3047 3048 pygmaea 8836 9376
parryi inyoensis —
7946 11878 11879 11880 11890 11891

L__+____- =3§s San Gabriel populations ————

Fic. 1. Basal leaf silhouettes depicting range in size. Silhouettes were drawn
from the tenth leaf of the basal rosette of common garden plants. Each pair repre-
sents the largest and smallest tenth leaf observed within each of the garden popu-
lations. Each number refers to the voucher of the natural population ( = seed
source).

1977] WILKEN: HULSEA a1

and 8202 of ssp. vestita. The lowest level of similarity between the six
problematical OTUs is 0.906, which is the overall similarity between
OTU 11891 and the other five OTUs.

Phenetic relationships of the six San Gabriel OTUs indicate an alli-
ance with ssp. callicarpha. Variation in garden populations of both
groups overlapped with respect to a number of characters, including
plant height, peduncle length, achene length, pappus length, and the
number of phyllaries, ray flowers, and disc flowers. Basal leaf shape and
size in San Gabriel populations are intermediate to those of ssp. vestita
and ssp. parryi, but blade margin approximates that of ssp. parryi or
ssp. invoensis. There is a tendency toward subentire leaves in western
populations in the San Gabriel Mt. (7936, 11878) and crenate to slightly

8703 ] pygmaea
3041

3047 parryi
3048
4255
8200
8202

|
=
|

vestita

inyoensis
ss36 | '"Y

7896
7844
7858
11891
11890
11880] SAN GABRIEL
11879

11878
7946

callicarpha

POPULATIONS

(eS Ee ee Cee eer eee |
64 70 76 82 88 94

Fic. 2. Cluster diagram of 18 populations representing all infraspecific taxa
within Hulsea vestita A. Gray. The phenogram was constructed using single-linkage
clustering. The horizontal axis represents the correlation coefficient used in this

study. Each number refers to the voucher of the natural population ( = seed
source).

OZ MADRONO [Vol. 24

lobed leaves in the eastern populations (11890, 11891; Fig. 1). Plants
of garden populations had reddish-tinged ray and disc corollas. Con-
spicuous presence of floral anthocyanins also characterizes plants of
most natural populations and is consistently found elsewhere only in
ssp. parryi and ssp. pygmaea. The lanceolate to broadly obovate phyllaries
of San Gabriel Mt. populations are most similar to those of ssp. vestita
and ssp. callicarpha but differ by a conspicuous pigmentation, which
is often restricted to phyllary apices.

A total of 410 crosses, representing 101 of 153 possible interpopula-
tional combinations and all 18 intrapopulational combinations, were
attempted. The yield of firm, plump achenes ranged from 21 to 93%
(ave. = 58) of pollinated disc flowers. Correlations between percent
viable achenes and kinds of combinations (e.g., intersubspecific or intra-

F, Pollen Stainability @ 8703
@- 3041
@ 86-100 % @@e 3047
@@@.-. 3048
e@ 71- 85 % @-e@.« « 4255
@@e@e@ee- 8200
e 33 - 70 & @e@@e.:.@.« 8202
@@:2@@@: 9376
x unattempted e664: 66 46 ace
cross , @.«.x« «x « «@@ « 7896
@@:20@000:@*« = 7844
@e@eee:-00:00°° #7858
@:.e@:.exe0e0@@« @e 11891
@@«««@x«x«x x«@«@e 11890
@@e@0@:-@: :-@:-@00@~ 11880
@«.@««@.« «xx xv « @x @* 11879
@:- @@202:@@:2000000*« 11878
e@@e@e-@@-0e*««-@«@* OO 7946
SS SSSSTTAOORHNOOON
Fe SS 2S ooo Reo alors aac

SAN GABRIEL POPS. VESTITA PARRYI
CALLICARPHA INYOENSIS PYGMAEA

Fic. 3. Crossing diagram summarizing data from inter- and intrapopulational
crosses within Hulsea vestita A. Gray. Each number refers to voucher of the
natural population.

1977] WILKEN: HULSEA 53

subspecific) were not statistically significant. Germination ranged from 3
to 100% (ave. = 37), but no significant correlations with kinds of com-
binations were observed. Hybrids of all 18 intrapopulational and 91
interpopulational combinations were grown to flowering maturity and
examined for meiosis and pollen stainability (Fig. 3). Nearly all hy-
brids, regardless of combination, were vigorous under garden and green-
house cultivation. Normal meiosis and high pollen stainabilities charac-
terized most hybrids. The lowest pollen stainabilities were in F, hybrids
involving ssp. pygmaea as one parent. Observations of meiosis in these
hybrids revealed the presence of either 17 bivalents and a chain of 4
chromosomes or 17 bivalents, a chain of 3, and 1 univalent. Most inter-
subspecific F, hybrids resembled the maternal parent with respect to
peduncle length and plant height but were intermediate to both paren-
tal strains with respect to most other morphological characters.

DISCUSSION

Contrary to my earlier assessment that western Transverse Range
populations represent an intergrading complex involving ssp. vestita
and ssp. parryi (Wilken, 1975), the numerical analysis and review of
morphological variation indicate that these populations possess a com-
bination of characters unique within Hulsea vestita. These characters
include subentire to crenate, spatulate leaves with relatively broad peti-
oles, reddish-tinged, lanceolate to broadly obovate phyllaries, reddish-
tinged corollas, and peduncles from 11—47 cm long. Most plants of nat-
ural populations may be distinguished by these characters. Furthermore,
the use of data based on common garden plants suggests the genetic
similarity of these populations. This distinctive morphology, combined
with an allopatric distribution, leads to the conclusion that these popu-
lations represent a discrete taxon. Accordingly, I propose the following
name:

Hulsea vestita ssp. gabrielensis Wilken, ssp. nov.

Herbae perennes scaposae, 17-38 cm altae. Folia basalia spathulata,
ca. 3-6 cm longa, ca. 1.0—-2.5 cm lata, utrinque glandulosa et laxe lanata,
margine integra vel crenulata vel rare crenata. Capitula plerumque nu-
merosa, vel in plantis minoribus tantum 1-3, pedunculis 11-47 cm longis.
Involucrum subcylindricum usque hemisphericum, ca. 2 cm latum, ca. 1
cm altum, phyllariis 3—4-seriatis, interiores oblongis vel anguste lanceo-
latis, exteriores lanceolatis vel late obovatis ex rubreo viridis. Flores radii
16-23, ex rubreo pallide flava. Flores disci 41-91, corolla ca. 6 mm longa,
tubo gracili, lobis ex rubreo flavis, parce glanduloso, faucibus late cylin-
dricis. Achaenia compresso-quandrangulata, ca. 5 mm longa, supra 1 mm

lata, nigra, strigosa. Pappus ex 4 paleis, subequalis. Chromosomatum
humerus: 2 = 19.

54 MADRONO [Vol. 24

Type: California, Los Angeles Co., Angeles Crest Highway, 3 mi E of
Chilao, San Gabriel Mts., 6000 ft., 1 Jun 1973, D. H. Wilken 11880
(Holotype: RSA!; isotype: CS!, others to be distributed).

Distribution: Known from open, gravelly or disturbed sites within or
marginal to the coniferous forest of the western Transverse Ranges of
southern California from Frazier Mt., Ventura Co., east to Mt. San
Antonio, Los Angeles Co.

Representative specimens: CALIFORNIA: Los Angeles Co.: Anderson
132 (WTU), Bacigalupi 4201 (CAS, JEPS, RM, UTC), Bacigalupi &
Alava 6426 (JEPS), Elmer 3700 (DS, GH, MO, POM, US, WTU),
Ewan 7216 (POM), Goodman & Hitchcock 1722 (MO, NY, RM),
Pierson 2440 (JEPS, RSA). Ventura Co.: Hall 6598 (DS, UC, US).

Affinities of ssp. gabrielensis and evolutionary relationships of infra-
specific taxa within Hulsea vestita remain unclear. These taxa are pri-
marily separated by combinations of morphological characters that vary
in a quantitative but discontinuous fashion. Intersubspecific hybrids
were produced with relative ease and, with the exception of hybrids
involving ssp. pygmaea, were characterized by normal meiosis and com-
paratively high pollen fertility. These data suggest that evolution within
the complex has not been accompanied by selection from genetic bar-
riers to cross-compatibility nor by major chromosomal reorganization.
Distribution of the several subspecies is largely allopatric. Although sym-
patric with ssp. vestita in the Sierra Nevada and with ssp. parryi in the
San Bernardino Mts., ssp. pyvgmaea is restricted to alpine or subalpine
sites while the former taxa are associated with coniferous forest at lower
elevations. It is likely, then, that differentiation has primarily been asso-
ciated with geographical and ecological isolation. Maximum isolation of
populations probably was last achieved during the Xerothermic period,
some 8500 to 3000 years ago (Heusser, 1960). During this period, the
distribution of coniferous forest was probably restricted to only the
highest elevations in the mountains of southern California (Axelrod,
1966). Earlier glacial-interglacial cycles also contributed to fluctuations
in elevational distribution of coniferous forest. As suggested by Axelrod,
such conditions favored population instability and local isolation. I sug-
gest that these conditions also contributed to differentiation within Hwl-
sea vestita, as is indicated by the strongly insular patterns of morpho-
logical variations within the complex.

ACKNOWLEDGMENTS
I thank Michael Amling for writing the computer program, Janice
Beatley for collection of ssp. zmyoensis, and Robert P. Adams for advice
concerning the numerical analysis.

LITERATURE CITED
AXELRop, D. I. 1966. The Pleistocene Soboba flora of Southern California. Univ.
Calif. Publ. Geol. Sci. 60:1-79.

1977] TANOWITZ: HEMIZONIA Ss)

Ferris, R. S. 1960. In L. Abrams, Illustrated flora of the Pacific States 4:1-732.
Stanford Univ. Press, Stanford, Calif.

HevusseEr, C. J. 1960. lente Pleistocene environments of North Pacific North
America. Amer. Geogr. Soc. Special Publ. 35:1-308.

Jepson, W. L. 1925. A manual of the flowering plants of California. Associated
Students Store, Berkeley.

Keck, D. D. 1959. In P. A. Munz, A California flora. Univ. Calif. Press, Berkeley.

Muwnz, P. A. 1974. A flora of Southern California. Univ. Calif. Press, Berkeley.

RypsBerc, P. A. 1914. Carduaceae. Helenieae. North Amer. Flora 34(1) :1—80.

SNEATH, P. H. and R. R. Soka. 1973. Numerical taxonomy. W. H. Freeman, San
Francisco.

Wirken, D. H. 1975. A systematic study of the genus Hulsea (Asteraceae). Brit-
tonia 27:228—-244.

AN INTERSECTIONAL HYBRID IN HEMIZONIA
(COMPOSITAE: MADIINAE)

BarRY D. TANOWITZ
Department of Biological Sciences
University of California, Santa Barbara 93106

The tarweed genus Hemizonia consists of 25 annual and six perennial
species that bloom in late spring, summer, and fall. They occur through-
out the Coast Ranges and valleys of California, northern Baja California,
and the adjacent offshore islands. Two of the annual species found in
the Southern Coast Ranges are Hemizonia australis (Keck) Keck (sect.
Centromadia, n = 11) and A. ramosissima Benth. (sect. ‘Deinandron’
sensu. Keck, nm = 12). [Apparently Keck (1935, 1958, 1959) did not
formally establish this section; it will probably be designated as sect.
Hartmannia (Gray) Gray (1874, 1876; Tanowitz, ined.) |. Distribu-
tions of these two species overlap in the low-lying coastal regions of
Santa Barbara County where six intersectional hybrids were discovered.
This hybrid combination is apparently rare.

Intersectional hybrids have been described in a number of genera
such as Ceanothus (Nobs, 1963; Hannan, 1974), Crepis (Babcock and
Stebbins, 1938), Ribes (Keep, 1962), Helianthus (Heiser et al., 1962),
and Perityle (Powell, 1970). A number of intersectional hybrid attempts
in Hemizonia were made by Clausen (1951). All were highly sterile
except for a quite fertile one between H. pungens (H. & A.) T. & G.
(sect. Centromadia, n= 9) and H. ramosissima. Hemizonia pungens,
although well differentiated from other members of its section, clearly

On

6 MADRONO [ Vol. 24

belongs in sect. Centromadia (Kyhos, pers. comm.). Hybrids at this
taxonomic level, however, are rare in nature and can be produced experi-
mentally only with difficulty (Grant, 1963, 1970). It was interesting then
to carry out morphological and cytological studies to aid in determining
possible phylogenetic relationships between the sections.

TABLE 1. MORPHOLOGICAL COMPARISON OF HEMIZONIA AUSTRALIS, H. RAMOSIS-
SIMA, AND PUTATIVE Hyprips.

Character H. australis Hybrids H. ramosissima

Shape of leaf tip
Peduncle length:
maximum (mm)

apiculate acute-obtuse obtuse

9.0(3.8-15.0) 10.7(6.0-16.8) 12.5(6.5—22.0)

minimum (mm) 0.0 2.9(0.5— 6.5) 4.0(0.5-12.5)

max/min 0.0 *5.3(1.4—13.5) 4.0(1.7-16.0)
Number of heads on

ultimate branches 4.9(3-8) 8.7(5-13) 10.4(7-18)
Shape of phyllary tip apiculate acute-obtuse obtuse
Phyllary:

length (mm) 4.3(3.5—-5.5) *4.9(3.5-6.2) 4.2(3.5-5.5)

width (mm) 5.9(5.0-6.5) 4.5(3.0-6.0) 3.1(2.2-4.5)

l/w 0.7(0.5—0.8) 1.1(0.8-1.4) 1.5(0.7-1.9)
Number of ray flowers/head 19( 16-22) 8.9( 8-10) 5
Number of disc flowers/head 40(33-48) 13.6(9-16) 6
Receptacular bract:

length (mm) 4.6(4.2—5.0) *5.3(4.5—5.8) 4.9(4.0-6.2)

width (mm) 1.0(0.9-1.1) *1.6(1.5-2.0) 1.4(1.2-1.8)

l/w 2.2(2.0-2.4) 0.4(0.3-0.5) 0.3 (0.2-0.4)
Ray corolla:

length (mm) 5.1(4.5-6.0) *5.5(4.5-6.0) 5.2(3.0-7.0)

width (mm) 1.6(1.2—2.0) 2.2(1.5-2.8) 3.6(3.0-5.0)

l/w 3.2(3.0-4.0) 2.2(2.0-3.5) 1.5(1.0-1.8)

lobe number 2 2 or 3 3
Ray achene:

length (mm) 1.9(1.5-2.5) *2.2(2.0—2.5) 2.0(1.8-2.5)

width (mm) 1.1(1.0-1.2) 1.0(0.8-1.2) 1.1(1.0-1.5)

l/w 1.6(1.4-1.8) *2.2(1.8-2.7) 1.9(1.2-2.2)
Disc achene length (mm) 1.2(1.5-2.5) 1.4(1.2-1.8) 1.9(1.5—2.5)
Number of pappus paleae/

disc flower 3 (3-4) 6(5-7) 7.6( 7-10)
Length of pappus paleae/

disc flower:

maximum (mm) 2.9(2.2-3.0) 2.4(1.8-3.0) 1.4(1.0-1.8)

minimum (mm) 2.6(2.0-3.2) 1.2(1.0-1.5) 1.1(1.0-1.5)

max/min 1.1(1.0-1.2) *2 .0(1.3-2.7) 1.3(1.0-1.5)

“Numerical values represent the mean of 10 measurements on each of 5 individuals
from 20 populations for each of the parents and 20 measurements on each of the
6 hybrids. Note that for some characters the hybrids exceed either of the parental

types.

1977] TANOWITZ: HEMIZONIA 57

MATERIALS AND METHODS

Comparisons of hybrids and parental taxa for 26 characters are given
in Table 1. These were selected for ease of measurement and represent
some of the most striking differences in character states. Vouchers of
specimens from populations used in this study are listed in Table 2 and
are deposited at UCSB.

Meiotic figures in PMC’s were obtained from flower buds fixed in a
solution of either Newcomer’s fixative or modified Carnoy’s (ethanol:
chloroform:acetic acid, 6:3:2, v/v). Buds were transferred to 70%
ethanol for storage. Cells were stained in iron acetocarmine and slides
were made permanent following Beeks’ method (1955). Pollen was
stained in 1% aniline blue in lactophenol for at least 24 hr. Five hundred
pollen grains for 20 specimens of H. australis and 20 specimens of H.
ramosissima were scored. Twenty flower buds were sampled from each
of the six putative hybrids. Pollen grains that stained evenly were con-
sidered viable; unstained or unevenly stained grains were considered
inviable.

RESULTS

Meiosis in the parents. Hemizonia australis (2n = 22; Venkatesh,
1956; Table 2) exhibits regular meiosis with 11 pairs of morphologically
similar chromosomes. There are normally 2 chiasmata per bivalent, one
in each arm. There is occasionally an interstitial chiasma formed in one
of the arms. This species showed an average of 94.6% stainable pollen,
ranging from 86.0 to 99.8% for all populations. Hemizonia ramosissima
(2n = 24; Johansen, 1936; Table 2) also exhibits regular meiotic divi-
sions with 12 bivalents. These chromosomes are morphologically similar
as well. The chiasma frequency is approximately two per bivalent with
one chiasma per arm. This species showed 98.2% stainable pollen, rang-
ing from 93.6 to 100% for all populations.

Meiosis in the hybrids. Pachytene chromosomes were normally unana-
lyzable. Late diakinesis and metaphase cells were analyzed in four of the
six hybrid plants. Configurations ranged from univalents to hexavalents.
Most cells, however, contained only univalents, bivalents, and trivalents
(Table 3 and Fig. 1). Bivalents were generally rod-shaped, while tri-
valents were predominantly forked (Figs. 1,4). Bivalents and _tri-
valents were generally oriented on the equator; univalents were scat-
tered through the cell (Fig. 2). Diakinesis and metaphase cells were
often difficult to analyze thoroughly due to the large number of chro-
mosomes, the tendency for the chromosomes to be “sticky”, and the
reduced pairing. Numbers of other cells were observed in which the com-
plete complement could not be clearly discerned; many of these cells
seemingly contained univalents, bivalents, and occasionally a trivalent.

58 MADRONO [Vol. 24

TABLE 2. CHROMOSOME NUMBERS OF HEMIZONIA AUSTRALIS, H. RAMOSISSIMA,
AND Hysrips USED FoR ANALYSIS IN THIS Stupy. All collections are from Santa
Barbara County, California. Collection numbers without name refer to Tanowitz
collections.

H. australis (Keck) Keck; n = 11.

Field S of Storke and Hollister Rds, 899-1B, 899-R, 1573-A, 1573-B, 1573-L;

field E of Los Carneros and El Colegio Rds, 1576-A, 1576-B, 1576-C, 1576-D;

lagoon on campus UCSB, Tanowitz & Varney 1583; slough E of Los Carneros

and Hollister Rds, Tanowitz & Zalin 1725—-A.

H. ramosissima Benth; n = 12.

Road cut off Hwy 101 at Gaviota, 465-1C, 1218; San Antonio Creek Trail near

San Marcos Pass Bridge, 747, 1523; Modoc Rd. at entrance to La Cumbre

Country Club, 1525; 04 mi W of Bailard Av. on Carpenteria Av., 1532-A,

1532-H ; along Arroyo Paredo Creek intersecting Hwy. 192, 1534-B; field S of

Storke and Hollister Rds., 1537-A, 1537-B, 1537-C, 1537-D, 1537-M ; El Cielito

Rd. and Stanwood Dr., 0.2 mi from Sheffield Reservoir, Tanowitz & Cowan

1589; W side of campus lagoon, 1507; along Casitas Pass Rd., at intersection

with Livingston Canyon Rd., 2428,

H. australis & H. ramosissima; 2n = 23.

W side of campus lagoon, UCSB 1299; field on UCSB campus near Pardall Rd.,

Tanowits & Smith 1300-E; field E of Storke and Hollister Rds., Tanowitz &

Zalin 1517; W side of campus lagoon UCSB, 1524; field E of Storke and Hol-

lister Rds., near University Village, 1538; field near ocean bluff and San Rafael

Dormitory, UCSB campus, 1568.

TABLE 3. FREQUENCIES AND MEANS OF SYNAPTIC CONFIGURATIONS IN POLLEN
MorTHerR Cetts (PMC) or HEMIZONIA AUSTRALIS X H. RAMOSISSIMA. Data derived
from diakinesis or first metaphase in 155 PMC’s from 4 plants.

Type Frequency/PMC Mean/cell
0) 1 ie 3:04 Secke i. 8.) 9) 10" it

I Pe 1300252 533. 289 222 12 6 63 1 — 1 3.49
II — 2 4 10 28 36 35 25 9 6 — — 5AT
III 12 50 61 25 7 —~ —- — — — — — Asef
IV 74 69 11 l1l-—- -—- eS aOrr ea e e 0.61
V 136 19 — —~ ~ ~—~ ~—~ ~—- ~—~ —- — — O.12
VI 1449 6 — ~ ~—~ ~ ~ ~ ~ ~ ~~ — 0.04

Chiasma frequency is low (approximately 14 per cell) but all chias-
mata appear to terminalize fully. Most bivalents had only one chiasma.
The model frequency of ring bivalents was only two per cell. Only three
bivalents in all the cells analyzed contained three chiasmata. All forked
trivalents had three chiasmata, while chains of three contained only two.
All but four quadrivalents had three chiasmata. Pentavalents had four;
hexavalents, five. Very few cells contained fragments at metaphase.

Forty-eight anaphase cells were analyzed. In every instance laggards
were observed. Occasionally, univalents were left on the metaphase
plate (Fig. 3) or there was precocious separation of univalents (Fig. 4).
There was a very low frequency of bridges or fragments evident at this
stage. Two cells had a large amount of fragmentation, giving them an
appearance ‘of being “shattered”. Such spontaneous breakage has been

1977 | TANOWITZ: HEMIZONIA 59

>

i ¢4 a F< ae

aly
oe a :

ov 3 2

r Q ; ok

Fic. 1-5. Camera lucida drawings of meiotic chromosomes of Hemizonia aus-
tralis x H. ramosissima. 1. Late diakinesis; note forked trivalents and _ hetero-
morphic pair (arrow), 2:-+ 61+ 311; Tanowitz 1568—-B. 2. metaphase I; univa-
lents scattered throughout cell; Tanowitz 1538—A. 3. Telophase I; univalents re-
maining on the metaphase plate; Tanowitz 1568-B. 4. Early anaphase I; note
precocious separation of univalents 4: + 41 + 2m + 1y; Tanowitz 1524-F. 5. Telo-
phase I with two micronuclei; note association of 3 chromosomes (arrow) ; Tano-
witz 1529-B.

described in a number of other plants (Darlington and Upcott, 1941;
Walters, 1956; Burns and Gerstel, 1969). The fragments appeared to
be paired and of more or less equal size. |

First telophase exhibited lagging chromosomes and micronuclei. There
often appeared to be an association of three chromosomes forming a
micronucleus at this stage (Fig. 5). Bridge-fragment formation was as
infrequent in the second division as it was in the first. Sporads of these
hybrids display a great deal of variation. The range was from four to
eight cells, with an average of six cells per sporad. The number of micro-
spores per sporad ranged from one to five, with an average of three per
sporad. Pollen stainability was quite low, averaging 7.2% and ranging
from 0.4 to 17.2%.

60 MADRONO [Vol. 24

DISCUSSION

Generic sections are constructed in order either to reflect the distinc-
tiveness between any two or more species groups or to indicate the affili-
ation of a group of species. Sections are generally circumscribed on the
basis of morphological characters, although cytogenetic, chemical, geo-
graphical, or ecological characters are sometimes used. Sections of Hemi-
zonia are morphologically circumscribed and there are numerous striking
character differences between the sections. Yet earlier investigations
(Clausen, 1951; Venkatesh, 1958) suggested a lower degree of diver-
gence when the genomic make-up of the species belonging to sections
Centromadia and ‘Deinandra was compared.

Although the frequency of hybridization in Hemizonia appears to be
low, it may have been more common in the earlier stages of divergence
of the genus. Venkatesh (1958) suggested that the aneuploid series in
sect. Centromadia (n = 9, 11, 12, 13; Johansen, 1936; Venkatesh, 1958;
Table 2) may have resulted from crosses between species with disparate
chromosome numbers, possibly of different sections. It is also possible
that the series in sect. ‘Deinandra’ (n = 9, 10, 11, 12, 13; Johansen,
1936; Table 2) may have resulted in this manner. Intra- as well as inter-
sectional hybrids appear reasonable sources for speciation. Hybrids of
intra- and intersectional origin were suggested for the origin of some
Clarkia species (Lewis and Roberts, 1956). Nevertheless, intersectional
hybridization in Hemizonia does not appear to be widespread now.

Species may have significant barriers to gene exchange maintained by
both pre- and postmating isolating mechanisms (see Levin, 1971). Ob-
servations in the field suggest that both kinds of barriers occur between
H. australis and H. ramosissima. Weak seasonal isolation is maintained
throughout the respective ranges. Hemizonia ramosissima blooms earlier
than H. australis ; however, the period of overlap is significant. Microhabi-
tat differences are found: H. australis occupies more saline regions than
H. ramosissima and the hybrids are found in the intermediate areas.
Pollinator constancy is quite strong where the species are in close con-
tact (Tanowitz, ined.). Hybrid sterility is certainly evident from the
data. Chromosomal differences are demonstrated clearly by the meiotic
configurations in the hybrids (Figs. 1-5, Table 3) exhibiting univalents
and multivalents, heteromorphic bivalents, and occasional bridges and
fragments. Presumably, these meiotic irregularities contribute greatly
to the sterility of the hybrids and reflect the occurrence of translocations,
paracentric inversions, and perhaps other chromosomal reorganizations.
None of these chromosomal differences per se necessarily preclude cross-
ing between the species and thus many more hybrid individuals might be
expected to occur; however, it is presumed that reproductive barriers,
about which nothing is now known, must occur in combination with
chromosomal disparity, since there is a paucity of hybrids in the field.

The hybrid plants are highly sterile annuals and obligate outcrossers
(Keck, 1959; Tanowitz, ined.; Kyhos, pers. comm.). There is no evi-

1977] TANOWITZ: HEMIZONIA 61

dence of introgression; only putative F,’s have been found in the sym-
patric populations. This suggests that gene flow is non-existent or rare
and that these sharply distinctive species seemingly maintain their integ-
rity completely. The morphological and cytogenetical evidence supports
their assignment to separate sections.

ACKNOWLEDGMENTS
I gratefully acknowledge the help of Dr. Dale M. Smith for his criti-
cism of the manuscript and his support throughout this study. I also
thank Dr. Donald W. Kyhos for helpful comments on sect. Centromadia.
I thank Fern Zalin, Clark Cowan, and Diane Varney for aid in handling
specimens.

LITERATURE CITED

Bascock, E. B. and G. L. STEBBins, Jr. 1938. The American species of Crepis. Their
interrelationships and distribution as affected by polyploidy and apomixis.
Publ. Carnegie Inst. Wash. 504. 199 pp.

BEEks, R. M. 1955. Improvements in the squash technique for plant chromosomes.
Aliso 3:131-133.

Burns, J. A. and D. U. GErsSTEL. 1969. Consequences of spontaneous breakage of
heterochromatic chromosome segments. Genetics 63:427—439.

CLAUSEN, J. 1951. Stages in the evolution of plant species. Cornell Univ. Press,
ee

Dartrncton, C. D. and M. B. Upcotr. 1941. Spontaneous chromosome change. J.
Genet. 41:297-338.

Grant, V. 1963. The origin of adaptations. Columbia Univ. Press, New York.

. 1970. Plant speciation. Columbia Univ. Press, New York.

Gray, A. 1874. Notes on Compositae and characters of certain genera and species.
Proc. Amer. Acad. Arts 9:187-218.

. 1876. Calif. Flora. Cal. Geol. Survey Botany. 2nd Revised ed. '1876- 1880.
1:277- 612.

Hannan, L. 1974. An intersectional hybrid in Ceanothus. Madrofio 22:402.

Hetser, C. B., Wm. C. Martin, and D. M. Smiru. 1962. Species crosses in Helian-
thus: I. Diploid species. Brittonia 14:137—-147.

JoHaAnsEN, D. A. 1936. Cytology of the tribe Madinae, family Compositae. Bot.
Gaz. 95:177-208.

Keck, D. D. 1935. Studies upon the taxonomy of the Madinae. Madrofio 3:4-18.

. 1958. Taxonomic notes on the California flora. Aliso 4:101—114.
. 1959. Hemizonia. In Munz, P.A., A California Floria. Univ. California

Press, Berkeley.

Keep, E. 1962. Interspecific hybridization in Ribes. Genetics 33:1-23.

evan, D. A. 1971. The origin of reproductive isolating mechanisms in flowering
plants. Taxon 20:91-113.

Lewis, H. and M. R. Roberts. 1956. The origin of Clarkia lingulata. Evolution
10:126-138.

Noss, M. A. 1963. Experimental studies on species relationships in Ceanothus.
Publ. Carnegie Inst. Wash. 632. 94 pp.

PowELL, A. M. 1970. Natural intersectional hybridization in Perityle (Compositae) .
Bettonia 22:3-10.

VENKATESH, C. S. 1958. A cyto-genetic and evolutionary study of Hemizonia,
section Centromadia. Amer. J. Bot. 45:77-83.

Watters, J. L. 1956. Spontaneous meiotic chromosome breakage in natural popu-
lations of Paeonia californica. Amer. J. Bot. 43:342-354.

62 MADRONO [Vol. 24

NOTES AND NEWS

CHROMOSOME NUMBERS AND RELATIONSHIPS OF CLAYTONIA SAXOSA AND C. AREN-
ICOLA (PORTULACACEAE) —Claytonia saxosa Brandegee and C. arenicola Henderson
are rather uncommon and poorly understood taxa of western North America. Their
chromosome numbers, reported here for the first time, as well as morphological
observations, provide evidence for suggestions on possible relationships with other
species of Claytonia sect. Limnia. Current studies of the C. perfoliata Donn and
C. spathulata Hook. complexes of sect. Limnia reveal parallel variation in a num-
ber of the vegetative morphological features traditionally used to define these
species (Miller, Syst. Bot. 1:20-34. 1976; Fellows, Madrofio 23: 296-297. 1976).
This work suggests that relationships are better expressed by chromosome base
number and floral features, especially the surface of the seed coat. Our purpose is to
report the chromosome numbers of C. arenicola and C. saxosa and to suggest, from
correlations with seed morphology and other traits, how the relationships shown by
previous studies may be revised.

Two recent reviewers of infrageneric relationships in Claytonia (Swanson, Brit-
tonia 18:299-241. 1966; McNeill, Canad. J. Bot. 53:789-809. 1975) have placed
C. saxosa and C. arenicola in sect. Limnia, together with the other annual species
C. perfoliata, C. spathulata, and C. gypsophiloides F. & M. McNeill put in sect.
Limnia the perennials C. sibirica L. and C. heterophylla (T. & G.) Swanson, whereas
Swanson assigned these taxa to two other sections. Most of the clustering methods
used by McNeill in his numerical taxonomic analysis placed C. sibzrica and C. het-
erophylla as a closely allied pair, adjacent to—but somewhat removed from—the
cluster formed by the five other species mentioned above. In the several dendro-
grams presented by McNeill, C. arenicola links directly to the pair formed by C.
spathulata and C. gypsophiloides. A close morphological tie between these three is
also evident in McNeill’s plot (his fig. 8) of the first two axes in a principal-coordi-
nates analysis of sects. Limnia and Rhizomatosae. On this plot, C. sibirica and
C. heterophylla are in a more distant position, intermediate toward sect. Rhizoma-
tosae. Claytonia saxosa links directly to C. perfoliata in several of McNeill’s den-
drograms but in the plot just cited, it stands alone, about equidistant from C.
perfoliata and C. spathulata.

Our studies show that Claytonia arenicola is diploid, with x = 6 (2n = 12; ID,
Adams Co., Hells Canyon, 13.4 km upriver from Hells Canyon Dam, Miller 496;
OR, Wallowa Co., Hells Canyon, 1.5 km below Hells Canyon Dam, Miller 499;
WA, Asotin Co., Clarkston, 6 km W on S side of the Snake R., Miller 493). This is
the same as the base number found in C. sibirica (Lewis, Bot. Rev. 33:105-115.
1975). The flowers of C. arenicola are virtually indistinguishable from those of
C. sibirica, being protandrous with a showy corolla of “candy-striped” white or
pinkish petals 5-10 mm long. Its breeding system, like that of C. sibirica (Swanson,
Ph.D. Dissertation, Univ. California, Berkeley, p. 59. 1964), appears to be one of
obligate outcrossing, since plants that flowered in an insect-free greenhouse set no
seeds spontaneously. The inflorescence of C. arenicola resembles that of C. stbirica in
having a bract by each pedicel of the raceme. Mature seeds of the species have a
low-tubercled surface similar to that of C. sibirica but distinctly different from the
more prominently tubercled seeds of C. spathulata and C. gypsophiloides. Unlike
the dull-surfaced seeds of C. gypsophiloides and C. spathulata, the seeds of C. si-
birica, C. arenicola, and C. perfoliata show a “shiny highlight” when illuminated.
Therefore, C. arenicola differs significantly from C. spathulata and C. gypsophi-
loides, which have a base chromosome number of x = 8 (Fellows, loc. cit.; Lewis,
Ann. Missouri Bot. Gard. 54:180. 1967; Nilsson, Bot. Not. 119:464-468. 1966), and
in which the racemes have only a single bract at the base. Claytonia perfoliata has
a chromosome base number of x = 6 (Fellows, loc. cit.; Miller, loc. cit.; Swanson,
op. cit.) but it varies from C. arenicola in its consistently small, self-pollinating
flowers and its racemes, which are bracteate only at the base. If special weight is

1977] NOTES AND NEWS 63

given to the above characteristics, therefore, C. arenicola appears to be more closely
related to C. sibirica than to either the C. perfoliata or the C. spathulata-gypso-
philoides complexes.

Claytonia saxosa is diploid with « = 8 (2n = 16; CA, Siskiyou Co., Scott Valley,
mouth of Heartstrand Gulch, Miller 488). Its corollas are showy, with pink petals
6-8 mm long, and the species is putatively outcrossing, as one would also suspect of
the large-flowered, diploid C. gypsophiloides. In the material of C. saxosa we have
examined, the racemes are completely ebracteate (contrary to a statement in the
key by McNeill, op. cit., p. 801). The seed coat of C. saxosa is prominently tubercled
and dull-surfaced as in C. spathulata and C. gypsophiloides, although the shape of
the tubercules is slightly different. Although paired with C. perfoliata in some of the
numerical analyses reported by McNeill (op. cit.), C. saxosa seems to be relatively
more distant from that species than it is from C. gypsophiloides and C. spathulata,
if one assumes that chromosome number and seed coat morphology are conservative
indicators of genetic relationship.

Voucher specimens and permanent microslides for the chromosome counts re-
ported in this study are deposited in OSC. —JoHn M. MILzer and Kenton L.
CHAMBERS, Department of Botany and Plant Pathology, Oregon State University,
Corvallis 97331.

On THE RELATIONSHIPS OF CHENOPODIUM FLABELLIFOLIUM AND C,. INAMOENUM.—
Taxonomists have disagreed on the treatment of C. znamoenum Standley (North
Amer. Flora 21:1-93. 1916; type: Arizona-Mexico border near Douglas, Mearns
2286, US) and C, flabellifolium Standley (op. cit.; type: Baja California, San Mar-
tin Island, 1897, T. S. Brandegee s,.n., UC 116454).

Standley placed C. inamoenum in “group” Leptophylla together with several
other species. One of the species was C. hians Standley (op. cit.; type: near Dulce,
New Mexico, 1911, Standley 8129, US). Another species included in this group was
C. leptophyllum (Nutt. ex Moq.) S. Wats., which was originally described as C.
album var. leptophyllum Nutt. ex Moq. [DeCandolle, Prod. 13(2):71. 1849; type:
Gordon 260, K, with the locality given as LaPlatte (on the Platte River?) ]. Aellen
and Just (Amer. Midl. Naturalist 30:47—76. 1943) considered C. inamoenum to be
the same as C. leptophyllum. Wahl (Bartonia 27: 1-46. 1952-53) commented that
the type of the former “does not agree with any material seen”. Examination of the
type specimen reveals that it is the top of a plant with what appear to be few
primary leaves and with many seeds. The seeds (actually fruits, since the pericarp
is attached) of the type closely resemble those of C. leptophyllum in being 1.0 mm
or less in diameter and with black pericarps. A couple of what I interpret as pri-
mary leaves have two very weakly developed veins. Chenopodium hians typically
has leaves oblong to linear in outline with a midrib and two well developed lateral
veins whereas the leaves of C. leptophyllum bear only a strong midrib and no
discernable lateral veins. While no definitive conclusion can be reached regarding
the type of C. inamoenum, the seed characters are strong evidence for its being
considered as nearer to C. leptophyllum.

Chenopodium flabellifolium was viewed by Standley (op. cit.) as closely related
to the C. neomexicanum complex, since he placed it in his “group” Fremontiana
with several other species having basally lobed leaves. These other taxa included
C. neomexicanum Standley, C. arizonicum Standley, C. palmeri Standley, and
C. parryi Standley, all of which he described in this same paper in 1916. In an
earlier paper (Madrofio 22:185-195. 1973), I considered the types of these names
to be conspecific, and C. lenticulare Aellen, (Feddes Repert. Spec. Nov. Regni Veg.
26:31-64, 119-160. 1929) was likewise considered to belong to the same species.
Whereas Standley considered C. flabellifolium to be related to C. neomexicanum,
Aellen and Just (op. cit.) placed the former in synonymy under C. opulifolium,
which is a sparingly introduced European species (Schrader in Koch and Ziz, Cat.

64 MADRONO [Vol. 24

pl. Palat. 6, 1814, type not seen). Wahl (op. cit.) treated C. flabellifolium in the
same manner as Aellen and Just, but he added “probably” parenthetically after the
name. This I take as an indication of doubt concerning the proper placement of
this species. Wahl, who had a remarkable understanding of Chenopodium in North
America despite limited field work, did not publish further on the relationships of
C. inamoenum and C. flabellifolium. He did, however, annotate a number of speci-
mens in various herbaria and these annotations suggest that in later years he came
to quite a different conclusion on relationships than was expressed in his papers.
Wahl annotated the type of C. flabellifolium as C. inamoenum, and moreover he
indicated that types of both are conspecific with the type of C. arizonicum (type:
Arizona, Santa Rita Forest Reserve, D. Griffiths 5982, US). Thus, if one were to
combine this concept of Wahl with my latest treatment of the C. neomexicanum
complex in which I consider the types of C. arizonicum, C. lenticulare, C. neomexti-
canum, C. palmeri, and C. parryi to be conspecific, then one would have these five
names plus C. flabellifolium and C. inamoenum refer to one species. This is not
tenable, and I shall present my concepts of relationships among these species.

The type of C. flabellifolium is similar to that of C. neomexicanum. The lower
and primary leaves are about as wide and as long and have mostly bipartite basal
lobes. The leaf shape falls easily into the variation encountered in C,. neomexicanum.
The more mature seeds measure 1.1-1.2 mm in diameter and the pericarp is
strongly attached. The type of C. flabellifolium differs from C. opulifolium (as I
understand it) in several respects, the most notable being the more strongly keeled
sepals of the latter. In my opinion, C. flabellifolium is closely related to C. neo-
mexicanum, and indeed they may be conspecific. I have found only two collections
in addition to the type collection that could be referred to C. flabellifolium, and
both were made prior to the present century. The question of whether or not
C. flabellifolium is conspecific with C. neomexicanum cannot be answered with
certainty at present; however, there is no question that the types of C. flabelli-
folium and C. inamoenum are distinct. Chenopodium inamoenum is probably the
same as C. leptophyllum, although the depauperate nature of its type specimen
precludes a confident decision.

Supported by NSF Grants GB-29793X and BMS-21384.—DanieL J. CRAWFORD,
Department of Botany, University of Wyoming, Laramie 82071.

Membership in the California Botanical Society is open to individuals ($12.00 per
year, regular; $8.00 per year, student). Members of the Society receive MaproNo
free. Institutional subscriptions to MApRONO are available ($14.00 per year).

Back issues of MApRONO are available. Volume 1 (1916-1929) and Volume 2
(1930-1934) consist of 17 numbers each and are available complete, unbound, at
$17.00 per volume. Volume 3 (1935-1936) and subsequent volumes consist of eight
numbers each (four per year). Volumes 3 through 20 are $16.00 each. Later volumes
are $24.00 each. (Some volumes lack 1-3 numbers of original printing; Xerox copies
are used to complete these.) Single numbers (original, if available; otherwise,
Xerox copy) are $1.00 each for volumes 1 and 2; $2.00 each for volumes 3 through
20; and $3.00 each for volumes 21-23; and $3.50 each for later volumes.

Applications for membership (including dues), orders for subscriptions, requests
for back issues, changes of address, and undelivered copies of MApDRONO should be
sent to the California Botanical Society, Inc., Department of Botany, University of
California, Berkeley 94720.

INFORMATION FOR CONTRIBUTORS

Manuscripts submitted for publication in MaproNo should be sent to the Editor.
Membership in the California Botanical Society is requisite for publication in
Maprono.

Manuscripts and accompanying illustrative materials must be submitted in dupli-
cate and should follow the format used in recent issues of MApRONO. Original illus-
trations should not be submitted until paper is accepted for publication. All manu-
scripts MUST BE DOUBLE SPACED THROUGHOUT, including title, text, tables,
captions, lists, literature cited, etc. Footnotes (which should be avoided wherever
possible), captions, and tables must be typed on sheets separate from the text.
Presentation of nomenclatural matter (accepted names, synonyms, typification)
should follow the format used for Rhus integrifolia in MADRONO 22:288. 1974. All
measurements should be given in S. I. (metric) units. Where appropriate, scales
should be included on figures rather than in captions as estimates of ratio of repro-
duction such as X %4, X 1620, etc. Institutional abbreviations in specimen citations
should follow Holmgren and Keuken’s list (Index herbariorum, Part 1, The herbaria
of the world. Sixth edition. 1974. Regnum Veg. vol. 92). Abbreviations of names of
journals should be those in Botanico-Periodicum-Huntianum (Lawrence, G. H. M.
et al. 1968. Hunt Botanical Library, Pittsburgh). If the correct abbreviation cannot
be determined, the full title of the journal should be used. Titles of books should be
given in full, together with the place and date of publication, names of publisher,
and an indication of the edition, if other than the first.

Short articles such as range extensions and other brief notes are published in con-
densed form under the heading “Notes and News”. Authors of such articles should
follow the format used in recent issues of MaproNo.

Authors are allowed up to 10 pages per year without page charges; charge for
additional pages is $30.00 per page. Subject to approval by the Editors, articles may
be published ahead of schedule, as additional pages of an issue, provided the author
assumes complete costs of publication.

he

DRONO

VOLUME 24, NUMBER 2 APRIL 1977

Contents

A NEw SPECIES OF DrRYPETES (EUPHORBIACEAE) FROM PANAMA,
Grady L. Webster 65

DaRMERA, THE CoRRECT NAME FOR PELTIPHYLLUM (SAXIFRAGACEAE),
AND A NEw COMBINATION IN PELTOPHYLLUM (TRIURIDACEAE),
Rudolf Schmid and Melvin D. Turner 68

Frost SENSITIVITY AND RESPROUTING BEHAVIOR OF
ANALOGOUS SHRUBS OF CALIFORNIA AND CHILE,

Harold A. Mooney 74
Two New Locat UMBELLIFERAE (APIACEAE) FROM CALIFORNIA,
Mildred E. Mathias and Lincoln Constance 78

VEGETATION ANALYSIS OF A NORTHERN CALIFORNIA COASTAL
PRAIRIE: SEA RANCH, SONOMA CoUNTY, CALIFORNIA,
M.M. Hektner and T.C. Foin 83

OBSERVATIONS ON ANTHOCARP ANATOMY IN THE SUBTRIBE
MIRABILINAE (NYCTAGINACEAE),

James Willson and Richard S pellenberg 104
TAXONOMY OF BEBBIA (COMPOSITAE: HELIANTHEAE),
Molly Whalen 112

NOTES AND NEWS
Tue “GERMINATION FLAP” IN CERTAIN GRAMINEAE,

John R. Reeder 123
REVIEWS

IsABELLA A. ABBoTT and GEorGE J. HOLLENBERG, Marine Algae

of California (John A. West) 124
Micuaet E. Hoare, The Tactless Philosopher. Johann

Reinhold Forster (1729-1798) (Joseph Ewan) 127
James Henrickson and RicHarp M. Straw, A Gazetteer of

the Chihuahuan Desert Region (John L. Strother) 128

A WEST AMERICAN JOURNAL OF BOTANY > SSP

BLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY

MaproNo is published quarterly by the California Botanical Society, Inc., and is
issued from the office of the Society, Herbarium, Life Sciences Building, University
of California, Berkeley. Established 1916. Second-class postage paid at Berkeley.
Return requested.

Editor — BarBaraA D. WEBSTER
Department of Agronomy and Range Science
University of California, Davis 95616

Associate Editor — Gravy L. WEBSTER
Department of Botany, University of California, Davis 95616

Board of Editors
Class of:

1977—WILL1AM Louis CULBERSON, Duke University, Durham, North Carolina

Date M. SuitH, University of California, Santa Barbara
1978—SHERWIN CarLQuist, Claremont Graduate School

LesLiE D. GortTLies, University of California, Davis

DENNIS R. PARNELL, California State University, Hayward
1979—Puuiire W. RUNDEL, University of California, Irvine

ISABELLE TAVARES, University of California, Berkeley
1980—James R. GriFFIn, University of California, Hastings Reservation

Frank A. Lane, Southern Oregon College, Ashland
1981—DaNnIEL J. Crawrorp, University of Wyoming, Laramie

James Henrickson, California State University, Los Angeles
1982—Deran W. Tayvtor, University of California, Davis

RICHARD VoGcL, California State University, Los Angeles

CALIFORNIA BOTANICAL SOCIETY, INC.

OFFICERS FOR 1977

President: Dr. WINSLOW R. Briccs, Carnegie Institution of Washington,
Stanford, California 94305

First Vice President: Dr. Atva Day, Department of Botany, California Academy of
Sciences, Golden Gate Park, San Francisco, California 94118

Second Vice President: Dr. Jos Ku1jt, Department of Biological Sciences,
University of Lethbridge, Lethbridge, Alberta, Canada

Recording Secretary: Dr. CHARLES F. QUIBELL, Department of Biological Sciences,
Sonoma State College, Rohnert Park, California 94928

Corresponding Secretary: Dr. RUDOLF ScHMip, Department of Botany,
University of California, Berkeley, California 94720

Treasurer: Dr. G. Douctas Bare, California Department of Food and Agriculture,
1220 N Street, Sacramento, California 95814

The Council of the California Botanical Society consists of the officers listed above
plus the immediate Past-President, DENNIS R. PARNELL, Department of Biological
Sciences, California State University, Hayward, CA 94542; the Editors of Madrono;
and three elected Council members: L. R. HEcKARD, Jepson Herbarium, Department
of Botany, University of California, Berkeley, CA 94720 (1975-1977); James R.
GriFFIN, Hastings Reservation, Star Route 80, Carmel Valley, CA 93924 (1976-
1978) ; Harry D. Turers, Department of Ecology and Systematic Biology, Califor-
nia State University, San Francisco, CA 94132 (1977-1979).

A NEW SPECIES OF DRYPETES (EUPHORBIACEAE)
FROM PANAMA

Gravy L. WEBSTER
Department of Botany, University of California, Davis 95616

Drypetes, which with about 200 species is one of the larger genera of
Euphorbiaceae (Airy Shaw, 1966), is best represented in the Old World
tropics; only about 20 valid species have been recognized from neotropi-
cal America. The genus is rather isolated in the Euphorbiaceae (Webster,
1967); together with the satellite (and scarcely distinct) genera Neo-
wawrea and Putranjiva, it is referable to the tribe Drypeteae, which has
no very close neighbors in subfamily Phyllanthoideae (Webster, 1975).
The species of Drypetes are rather poorly understood, partly because the
erratic flowering, dioecious flower distribution, and nondescript foliage
make recognition of imperfect specimens difficult.

The elusiveness of Drypetes is indicated by the fact that the species
here described, a tree 15-25 m high, was overlooked by many botanists
on Barro Colorado Island—one of the best-explored localities in the
tropics—until 1971, when it was collected by Dr. Thomas Croat and Dr.
Robin Foster. Recently I have examined two sheets of what appears to
be the same species from southern Venezuela; these collections were
annotated by Dr. Paul C. Standley with a manuscript name. Since both
of the Venezuelan specimens are pistillate, and it is not absolutely certain
that they are conspecific with (he Panamanian plants, it seems best to
describe the species from the Panama specimens and to commemorate
Dr. Standley’s association with the plant by dedicating the specific epi-
thet to him.

Drypetes standieyi Websier, sp. nov. sect. Drypetitis, ad D. vart-
abilem Uitt. accedens sed foliis chartaceis abrupte acuminatis, staminibus
8 vel 9 minoribus, pedicellis 2 brevioribus; a D. amazonica Steyerm.
differt antheris glabriusculis, foliis chartaceis integris.

Tree c. 15-25 m high; trunk c. 0.25—1.25 m diam., buttressed at base;
twigs terete, minutely hispidulous when young (trichomes 0.1 mm long
or less), glabrate and pale in age and prominently lenticellate. Leaves
chartaceous; petioles glabrous or nearly so, flattened adaxially, mostly
5-10 mm long; stipules scarious, very inconspicuous (less than 1 mm
long) ; blades nearly glabrous (slightly strigose on midrib beneath), ellip-
tic to broadly lanceolate, rather abruptly short-acuminate at tip, asym-
metrically cuneate at base, mostly 7-11 cm long, 2.5—6 cm broad, plumb-
eous and somewhat lucent on both sides; midrib salient beneath, main
lateral veins 7-9 on a side, distinctly raised beneath, brochidodromous,
veinlets forming a prominulous reticulum; margins entire. Staminate

Madrofo, Vol. 24, No. 2, pp. 65-128. April 21, 1977.

65

66 MADRONO LVol. 24

Cc

Fic. 1. Flowers of Drypetes standleyi. A, pistillate flower (Foster & Croat
2307). B, staminate flower (Croat 14849).

flowers in axillary clusters subtended by scarious bracteoles; pedicels
7-11 mm long, densely hispidulous; sepals 4, recurved, oblong-spathu-
late, marginally ciliate, appressed-hirsutulous on the back, glabrous or
nearly so within, c. 2.5-3 mm long, 1.7—2.2 mm broad; disk somewhat
fleshy, 1—-1.5 mm across, glabrous; stamens 8 (rarely 9), fllaments 1-2
mm long; anthers linear-oblong, glabrous (sometimes obscurely and
very sparsely pubescent), (1.0—) 1.2-1.6 mm long; pistillode absent.
Pistillate flowers axillary, 1-4 per cluster; pedicels stout (0.6-1 mm
thick), straight, densely hispidulous, 2-6 mm long; sepals 4, deciduous,
broadly obovate, minutely strigose on the back, glabrous within, mar-
ginally ciliate, 2.8-4.4 mm long, 2.5-3.2 mm broad; disk sirigillose-
ciliate, 2.0—2.3 mm across; ovary globose or ellipsoidal, c. 2 mm across,
densely whitish-tomentose, 1-locular; stigma sessile or nearly so, reni-
form, glabrous, c. 1.7—2.5 mm across. Fruits somewhat compressed,
strongly reticulate-wrinkled, appressed-hirsutulous, scarcely beaked, 2—
2.2 cm long, 1.3-1.5 cm across; endocarp lignified, c. 1.5 mm thick with a
pronounced obtuse ventral carina within; ovule attached apically by an
elliptical scar, mature seeds not observed.

Type: Panama, Canal Zone, Barro Colorado Island, E of Armour
Trail, 31 May 1971, R. Foster & T. B. Croat 2307 (pistillate; holotype,
DAV; isotypes DUKE, MO, and to be distributed). Paratype (stamin-
ate flowers): Barro Colorado Island, Armour Trail, 31 May 1971, T. B.
Croat 14849 (DAV, MO) (Fig. 1).

es

1977] WEBSTER: DRYPETES 67

AppITIONAL CoLLEcTions. PANAMA. Canal Zone: Barro Colorado I., 300 m E
of Armour Trail, Croat 14843 (DAV, MO); S of big trees on Armour Trail, Croat
16516 (DAV, MO); S of Zetek 11, Foster 1122 (DAV, DUKE). VENEZUELA.
Apure: Las Piedras, alrededor de Puerto Paez, J. Velez 2635 (VEN; fruits immature
and determination somewhat uncertain). Bolivar: en los rebalsos de Guayape, Bajo
Caura, L. Williams 11986 (VEN).

Drypetes standlevi clearly belongs in sect. Drypetes (sect. Hemt-
cyclia auct.) because of its small stipules, staminate flowers without pis-
tillode, and unilocular ovary (Pax & Hoffmann, 1922; Webster, 1967).
In the treatments of neotropical Drypetes by Monachino (1948) and
Jablonsky (1967), the plants from Barro Colorado would key out to the
South American species D. amazonica Steyrm. or D. variabilis Uitt.,
while in the revision by Pax and Hoffmann (1922) they would be close
to the West Indian species D. dussu Kr. & Urb. or D. glauca Vahl. At
first, it appeared that the Panamanian plants could be accommodated in
D. variabilis, but they clearly differ in the thinner less rigid leaves and in
a number of reproductive characters: smaller staminate sepals, stamens
mostly 8 (occasionally 8 in D. variabilis, but usually 7), shorter fruiting
pedicels, densely tomentose ovary, and wrinkled fruits. The differences
between D. séandleyi and its near congeners may be summarized in the
following key:

Stamens 4, anthers only 0.5 mm long, glabrous; leaves crenate, cuspidate-

acuminate; Guiana . . . . . . . D. fanshawei Sandw.
Stamens 4-12, anthers larger (or ace puneecene
Stamens aOR 4—7; leaves entire or nearly so.

Staminate flowers several per axil; sepals less than 1 mm broad;
anthers less than 1 mm long, pubescent; fruiting pedicels 5—12
mm long; drupes 0.8—1.3 cm long, oblique at tip; leaves char-
taceous; West Indies. . . . . . . . D. alba Poit.

Staminate flowers 1-3 per axil; picalinte sepals more than 1 mm
broad; anthers over 1 mm long, glabrous; fruiting pedicels
15-21 mm long; drupes 2—2.5 cm long, symmetrical; leaves
coriaceous; northern South America . . . D. variabilis Vitt.

Stamens mostly 8-12 (rarely 7).

Drupes 2.5—-3 cm long; staminate sepals at least 3.5-4 mm long,
glabrous except for marginal ciliae; anthers slightly pubescent;
Martinique . . . » 3 « 4 Dodussa Kr. & Urb.

Drupes less than 2.5 cm fone eomninete sepals 2.5—-3.5 mm long,
pubescent on the back.

Anthers glabrous (or very nearly so), 1.2—-1.6 mm long; leaves
chartaceous, entire; Panama and Venezuela. .

_— D. standleyi ences

thers! arsune ely) Retrecen 2. O-—2.4 mm long; leaves coriaceous,

entire tocrenate; Brazil. . . . . D.amazonica Saran.

68 MADRONO [Vol. 24

ACKNOWLEDGMENTS
I wish to thank Dr. Tom Croat and Dr. Robin Foster for providing the material
from Barro Colorado Island, the Missouri Botanical Garden (MO) and the Insti-
tuto Botanico in Caracas (VEN) for loan of specimens, and Ms. Mary Breckon for
making the drawings.

LITERATURE CITED

Arry SHAw, H. K. 1966. A dictionary of the flowering plants and ferns, by J. C.
Willis. 7th ed. Cambridge Univ. Press.

JaBLonsky, E. 1967. Botany of the Guayana highland—part VII: Euphorbiaceae.
Mem. N.Y. Bot. Gard. 17:80—-190.

Monacuino, J. V. 1948. Three new species of Drypetes. Phytologia 3:32-35.

Pax, F., and K. Horrmann. 1922. Euphorbiaceae—Phyllanthoideae—Phyllantheae—
Drypetinae. Das Pflanzenr. 147. XV. (Heft 81) :227-280.

WessterR, G. L. 1967. The genera of Euphorbiaceae in the southeastern United
States. J. Arnold Arb. 48:303-430.

. 1975. Conspectus of a new classification of the Euphorbiaceae. Taxon

24:593-601.

DARMERA, THE CORRECT NAME FOR
PELTIPHYLLUM (SAXIFRAGACEAE), AND A NEW
COMBINATION IN PELTOPHYLLUM (TRIURIDACEAE)

RupotF ScHMip and Metvin D. TURNER!
Department of Botany, University of California, Berkeley 94720

Current work by the senior author on vegetative and floral anatomy
of Burmanniaceae, Petrosaviaceae, and Triuridaceae, which are obscure,
mainly tropical families of achlorophyllous, saprophytic monocotyledons,
has led to two realizations: (1) the long discarded triuridaceous name
Peltophyllum Gardner (1843) must be used in preference to Hexuris
Miers (1850), and (2) of particular relevance to horticulturists and to
western North American botanists, the well-known saxifragaceous genus
Peltiphyllum (Engler) Engler (1891) requires a substitute name.

In 1841 John Miers (see also his 1845 paper) described Tviuris with
a single species, 7. hyalina. Later, George Gardner (1843, 1845) pub-
lished a new genus and species, Peltophyllum luteum. Gardner’s original
description follows (1843: 176; 1845):

PELTOPHYLLUM, Gardner.

Flores dioici. Masc. ignoti. Fam. Perigonium 6-partitum, col-
oratum, patens, persistens; laciniis ovatis, longé acuminatis;
acumine plano. Ovarza plurima, in tori apice sessilia, adpressa,
libera. Styli ad apicem incrassati, oblique truncati. Fructus
ignotus.

1 Present address of MDT: Department of Botany, Duke University, Durham
NC 27706.

1977] SCHMID & TURNER: DARMERA 69

Herba parvula Brasiliensis. Folia a scapo distantia, longé petio-
lata, peltata, valde reticulata. Radix tuberosa, fibrosa. Scapus
subramosus, basi squamosus,; pedunculis basi bracteatis, unt-
floris; floribus lufezs.

Peltophyllum luteum, Gardn. Herb. Bras. n. 3570.

Peltophyllum luteum was based on a single collection (Gardner 3570),
which we have not seen, and which apparently has been lost (Giesen,
1938; Malme, 1896). The Fielding-Druce Herbarium at Oxford Uni-
versity, which has a “‘very full set” (Clokie, 1964: 169) of Gardner’s
specimens, informed us (pers. comm., 1976) that it does not have
Gardner 3570.

Peliophyllum differed from Triuris in having 6 instead of 3 caudate
tepal lobes and also in having remarkable peltate leaves associated with
the flowering scapes. Miers (1850, 1852) pointed out that the peltate
leaves do not belong to the triuridaceous plant, but probably are of
seedlings of Menispermaceae. Rejecting Pelfophyllum, Miers (1850,
1852) created the new name Hexuris gardneri for Gardner’s fertile ma-
terial, and since then, Pelitophyllum Gardner seems not to have been
accepted other than by Schumann (1894) (for later references see
Giesen, 1938, the most recent monograph on the family).

The name Hexuris Miers (1850) is clearly illegitimate. Miers sub-
stituted this name for Peltophyllum Gardner (1843) because ‘‘pelto-
phyllum” is scarcely applicable to the exclusively squamate Triuridaceae.
However, according to the International Code of Botanical Nomencla-
ture (Art. 62), a legitimate name cannot be rejected merely because it
is inappropriate or disagreeable. Therefore, the name Pelfophyllum,
however descriptively inappropriate, must be retained since it is clearly
possible from Gardner’s description (1843, 1845) to select the fertile
material as the type (which is here designated the lectotype of Pelto-
phyllum luteum Gardner). It should be emphasized that apart from the
accounts of the peltate leaves and attached tuberous roots, which appar-
ently belong to Menispermaceae (Miers, 1850, 1852), Gardner (1843,
1845) adequately described the new triuridaceous plant, although fruits
and male flowers were not discovered until much later by Malme (1896).
Gardner in 1845 also accurately figured a female plant complete with
its roots and scale leaves.

The consequence of the above is that the well-known name Hexuris
Miers must be replaced with the more obscure Peltophvllum Gardner.
Schumann (1894) and the staff workers of Index Nominum Genericorum
had come to the same conclusion. Following Giesen’s (1938) circum-
scription of Hexuris Miers, Peltophyllum then includes 2 species: P.
luteum Gardner (1843) and P. caudatum (Poulsen) Schmid & Turner.

The consequence to western North American botanists of the imbroglio
over Triuridaceae comes with the recognition that the saxifragaceous

70 MADRONO [Vol. 24

name Peltiphyllum (Engler) Engler (1891)? is illegitimate because it is
a mere orthographic variant (Art. 75) of the earlier, legitimate Pelto-
phyllum Gardner (1843, Triuridaceae) and thus is a later homonym
(Art. 64). Therefore, a different generic name must be used for the single
species of Peltiphyllum (Engler) Engler, the familiar “umbrella plant”
(or “Indian rhubarb”’) of California and southern Oregon, P. peltatum
(Torrey in Bentham) Engler, a name that has been used in all manuals
of our region since the Englerian genus was established (see also Engler,
1930, and Wagner, 1907).

In 1899 the horticulturist Andreas Voss |1857—1924—-see obituary in
Gartenwelt 28:238-240 (1924)| published the name Darmera?® with
Peltiphyllum (Engler) Engler (1891) [the latter name corrected to
Peltophyllum Engler in Post and Kuntze (1903)]| in synonymy. The
reason for Voss’s (1899) erection of Darmera was his recognition of the
prior publication of the form genus Peltophvllum Massalongo (1854,
1859a), which was applied to a fossil leaf and fruit attributed to Nymph-
aeaceae (see Summary). Peltophvllum Massalongo, however, is also an
illegitimate later homonym of Peltophyllum Gardner, a fact that Voss
(1899) overlooked.

Darmera Voss (1899), not Peltiphyllum (Engler) Engler (1891) ,4
therefore is the correct name for the saxifrage. Voss clearly made the
combination Darmera peltata (Torrey in Bentham) Voss in 1899 in a
highly obscure horticultural journal, Gdrtnerisches Zentral-Blatt (full
citation in Just’s botanischer Jahresbericht 27 (Abt. 2):194 (1899),
which ceased publication after only a single volume, issued in 1899.

2 The genus Peltiphyllum is based on Saxifraga Linnaeus section Peltiphyllum
Engler (1872), with the single species Saxtfraga peltata Torrey in Bentham.

3° Darmera commemorates Karl Darmer of Berlin, the founder and first president
of the Allgemeinen Deutschen Gartner-Verein, and an indefatigable promoter of
horticulture in Germany (Voss, 1899).

* Because it lacks a Latin description, Peltiphyllaceae Krach (1976: 23) is not
validly published (Art. 36); however, were this name validly published, it would
be illegitimate since Peltiphyllum (Engler) Engler is illegitimate (Art. 18). The
proper designation for a familial segregation of Darmera Voss (syn.: Peltiphyllum)
from Saxifragaceae would, of course, be one based on the generic name Darmera.
Although this might be supported by the growing body of organographic, cytologi-
cal, chemical, anatomical, and especially embryological evidence (Krach, 1976;
also Bensel and Palser, 1975a, b; Saxena, 1973; Spongberg, 1972), familial recogni-
tion (sensu Krach, 1976) at this time seems premature. Perhaps segregation at the
tribal or subtribal level is more appropriate. On the other hand, the distinctness of
Darmera from other Saxifragaceae (e.g., Darmera differs from all other saxifrages in
its embryo development and unitegmic testa) precludes submerging the genus in
Saxifraga, a return to the original Englerian (1872) concept (see footnote 2). As
noted by Takhtajan (1959, 1973), Darmera is certainly worthy of further detailed
investigation.

1977] SCHMID & TURNER: DARMERA fal

Consequently, Darmera Voss (1899) has been overlooked by subse-
quent botanists (including /ndex Kewensis), the name appearing only in
a 1903 publication by Voss, but not in his later books; in Post and
Kuntze (1903), to which Voss had contributed; in Pilger (1906-08) ;
Engler (1930); Lemée (1930); and in /ndex Nominum Genericorum.
Curiously, the last 4 sources all relegate Darmera Voss to the synonymy
of Peltiphyllum (Engler) Engler (1891) despite the fact that the latter
is doubly bastardly in view of the prior legitimate Peltophyllum Gardner
(1843) and the prior illegitimate Peltophyllum Massalongo (1854).
None of the preceding sources, incidentally, correctly cite the original
and valid publication of Darmera and Darmera peltata, namely, Voss
(1899).

The lamentable conclusion is that two admirably appropriate descrip-
tive names, Peltiphyllum (Engler) Engler and Hexuris Miers, are ille-
gitimate. It is ironic that the legitimate but descriptively inappropriate
Peltophyllum Gardner must be used for plants bearing only minute scale
leaves. It is also regrettable that Peltiphvllum peltatum (Torrey) En-
gler, a familiar and striking Californian/Oregonian plant that is often
cultivated, must yield its generic name to insignificant South American
saprophytes and assume the obscure name Darmera Voss. The only
alternative to these required changes is to conserve Peltiphvllum (Engler )
Engler against Peltophyllum Gardner.

SUMMARY
(1) Triuridaceae:

PELTOPHYLLUM Gardner, Proc. Linn. Soc. 1:176 (1843), Trans.
Linn. Soc. 19:157, pl. 15 (1845), non Massalongo (1854), non
Engler (1891). Type: Peltophyllum luteum Gardner.

Hexuris Miers, Proc. Linn. Soc. 2:72 (1850), Trans. Linn. Soc.
21:44 (1852), nom. superfl., illeg.
(a) PELTOPHYLLUM LUTEUM Gardner, op. cit. (1843, 1845).
Hexuris gardneri Miers, op. cit. (1850, 1852), nom. superfl,
illeg.
Triuris lutea (Gardner) Bentham & J. D. Hooker, Gen. PI.
3:1002 (1883).
(b) Peltophyllum caudatum (Poulsen) Schmid & Turner,
comb. nov.
Sciaphila caudata Poulsen, Vidensk. Meddel. Naturhist.
Foren. Kjgbenhavn 36-38:165 (1886).
Hexuris. caudata (Poulsen) Giesen, Pflanzenreich Heft
104:75 (1938).
(2) Saxifragaceae:
Darmera A. Voss, Gart. Zentral-Bl. 1:645 (1899).

72 MADRONO [Vol. 24

Saxifraga L. sect. Peltiphyllum Engler, Monograph. Gatt. Saxi-
fraga, 108 (1872). Type: Saxifraga peltata Torrey in Bentham.

Peltiphyllum (Engler) Engler in Engler & Prantl, Nat. Pfl. 3(2a):
61 (1891), Nachtrag 3:141 (1906), Nachtrag 4:109 (1914),
nom. illeg., later homonym of Peltophvllum Gardner (1843).
DARMERA PELTATA (Torrey in Bentham) Voss, op. cit. 646

(1899).

Saxifraga peltata Torrey in Bentham, Pl. Hartweg., 311
(1849).

Peltiphyllum peliatum (Torrey in Bentham) Engler, op. cit.
(1891).

(3) ?Nymphaeaceae, fossil leaf and fruit:

PELTOPHYLLUM Massalongo, Monograf. Dombeyacee Foss., 22
(1854), Spec. Photo. Anim. Quor. Pl. Foss., 75, pl. 28 (1859a),
nom. illeg., non Gardner (1843), non Engler (1891—‘‘Peltiphyl-
lum’). Type: Peltophyllum nelumbioides Massalongo.

PELTOPHYLLUM NELUMBIOIDES Massalongo, op. cit. (1854,
1859a).

A check of the Compendium Index of Paleobotany (see
Schmid and Schmid, 1973) and of the paleobotanical literature
revealed acceptance of Massalongo’s illegitimate genus only by
Schimper (1874: 603), Meschinelli and Squinabol (1893: 326),
and, of course, Massalongo himself (1854, 1858, 1859a, b).
Other than the listing of the genus by Andrews (1970) in his
index of generic names, there has been no recent discussion or
acceptance of Peltophyllum Massalongo. The taxonomic affini-
ties of Massalongo’s fossil leaves from the Eocene of north-
eastern Italy are not known. Massalongo’s assignment of the
fossils to Nymphaeaceae (‘“‘Cabombee” or ‘‘Cabombeae”—
Massalongo, 1858, 1859a, b, but initially to Dombeyaceae in
Massalongo, 1854) was based on strictly superficial resem-
blances (Schmid and Schmid, 1973, 1974), as was typical of
paleobotanical work of this era. After examining the original
photograph in Massalongo (1859a), Hickey had “serious doubts
that the type is really nymphaeaceous” (pers. comm., 1976). In
view of the above, we are not perpetuating Peltophyllum Massa-
longo with a substitute generic name.

ACKNOWLEDGMENTS
We thank T. O. Duncan, J. E. Eckenwalder, L. R. Heckard, M. Mouse, R. B.
Phillips, A. R. Smith, E. G. Voss, and specially D.-E. Johnson, P. C. Silva, and J. L.
Strother for helpful comments and nomenclature advice, and also Dr. Paul Hiepko
(Berlin) for supplying the photocopy of Voss’s 1899 paper.

1977] SCHMID & TURNER: DARMERA

~I
Ww

LITERATURE CITED

Anprews, H. N., Jr. 1970. Index of generic names of fossil plants, 1820-1965. U'S.
Geol. Surv. Bull. 1300.

BENSEL, C. R., and B. F. PAtser. 1975a. Floral anatomy in the Saxifragaceae sensu
lato. II. Saxifragoideae and Iteoideae. Amer. J. Bot. 62:661-675.

, and — —. 1975b. Idem. IV. Baueroideae and conclusions. Amer. J.
Bot. 62:688-694.

Croxir, H. N. 1964. An account of the herbaria of the Department of Botany in
the University of Oxford. Oxford: University Press.

ENGLER, A. 1872. Monographie der Gattung Saxifraga L. mit besonderer Beriick-
sichtigung der geographischen Verhaltnisse. Breslau: J. U. Kern’s Verlag (Max
Miller).

. 1891. Saxifragaceae. In A. Engler and K. Prantl [eds.], Die natiirlichen
Pflanzenfamilien 3,2a:41—93. Leipzig: Wilhelm Engelmann. [Pp. 41-48, 1890,
49-93, 1891.] [Also Nachtrage 1:180—-182, 348 (1897), 2:29 (1900), 3:141-142
(1906), 226 (1907), 4:108-111 (1914).]

. 1930. Saxifragaceae. In A. Engler and K. Prantl’s Die natirlichen Pflan-
zenfamilien. 2. Aufl. by A. Engler [ed.]. 18a:74-226 [ed. by H. Harms]. Leip-
zig: Wilhelm Engelmann.

GARDNER, G. 1843. Description of Peltophyllum, a new genus of plants allied to
Triuris of Miers, with remarks on their affinities. Proc. Linnean Soc. London
1:176-177. [Reprinted in Ann. Mag. Nat. Hist., I, 13:217. 1844].

. 1845. Description of Peltophyllum, a new genus of plants allied to Triuris
of Miers, with remarks on their affinities. Trans. Linnean Soc. London 19:
155-160.

GresEN, H. 1938. Triuridaceae. Jn A. Engler and L. Diels [eds.], Das Pflanzenreich
IV/18(Heft 104) :1-84. Leipzig: Wilhelm Engelmann.

Kracu, J. E. 1976. Samenanatomie der Rosifloren I. Die Samen der Saxifragaceae.
Bot. Jahrb. Syst. 97:1-60.

LemMEE, A. 1930. Dictionnaire descriptif et synonymique des genres de plantes
phanérogames. T. 2. Brest: Imprimerie Commerciale et Administrative.

Matme, G. O. A:N. 1896. Ueber Triuris lutea (Gardn.) Benth. et Hook. Bih. Kongl.
Svenska Vetensk.-Akad. Handl., Afd. 3, 21(14) :1-16.

Massatonco, A. B. 1854. Monografia delle Dombeyacee fossili fino ad ora cono-
sciute. Verona: G. Antonelli.

. 1858. Palaeophyta rariora formationis tertiariae agri Veneti. Atti R. Ist.
Veneto Sci., III, 3:729-793.

————. 1859a. Specimen photographicum animalium quorumdam plantarumque
fossilium Agri Veronensis. Verona: Vicentini-Franchini.

. 1859b. Syllabus plantarum fossilium hucusque in formationibus tertiariis
agri Veneti detectarum. Verona: A. Merlo.

MEscHINELL, A., and X. Sgurnasor. 1893. Flora tertiaria Italica. Patavi: Sumpti-
bus Auctorum Typis Seminarii.

Miers, J. 1841. Description of a new genus of plants from Brazil. Proc. Linnean
Soc. London 1:96. [Reprinted zn Ann. Mag. Nat. Hist., I, 7:222. 1841.]

————.. 1845. Description of a new genus of plants from Brazil. Trans. Linnean
Soc. London 19:77-80.

. 1850. On the family of Triuriaceae. Proc. Linnean Soc. London 2:71-77.
[Reprinted in Ann. Mag. Nat. Hist., II, 7:323-327. 1851.]

————. 1852. On the family of Triuriaceae. Trans. Linnean Soc. London 21:43-59.

Pitcer, R. [ed.]. 1906-08. Nachtrige III to A. Engler and K. Prantl [eds.], Die
naturlichen Pflanzenfamilien. Leipzig: Wilhelm Engelmann. [P. 141 on Darmera
issued 1906. ]

74 MADRONO [Vol. 24

Post, T. von, and O. Kunrze. 1903. Lexicon generum phanerogamarum.
Stuttgart: Deutsche Verlags-Anstalt. [Issued 1903, but dated 1904—+teste F. A.
Stafleu, 1967, Taxonomic Literature, p. 364. ]

SAXENA, N. P. 1973. Studies in the family Saxifragaceae. IX. Anatomy of the flower
of some members of Saxifraginae. J. Indian Bot. Soc. 52: 251-266.

SCHIMPER, W. P. 1874. Traité de paléontologie végétale ... T. 3. Paris: J. B.
Bailliere et Fils.

ScHMiD, R., and M. J. Scumip. 1973. Fossils attributed to the Orchidaceae. Amer.
Orchid Soc. Bull. 42:17-27.

, and —————. 1974. On Massalongo’s fossils: Protorchis and Palaeorchis.
Amer. Orchid Soc. Bull. 43:213—216.

SCHUMANN, C. 1894. Triuridaceae. In C. F. P. de Martius, A. W. Eichler and I.
Urban [eds.], Flora Brasiliensis 3(3):645-668. Lipsiae: Frid. Fleischer in
Comm. [Entire volume issued 1890-94. ].

SPONGBERG, S. A. 1972. The genera of Saxifragaceae in the southeastern United
States. J. Arnold Arb. 53:409-498.

TAKHTAJAN, A. 1959. Die Evolution der Angiospermen. Jena: Gustav Fischer Verlag.

. 1973. Evolution und Ausbreitung der Blutenpflanzen. Ed. R. Fritsch, A.
Muller, and H. Ohle. Stuttgart: Gustav Fischer Verlag.

Voss, A. 1899. Darmera Voss. Gart. Zentral-Bl. 1:645-646.

. 1903. Salomon’s Worterbuch der deutschen Pflanzennamen ... . 2. Aufl.
Stuttgart: Verlagsbuchhandlung Eugen Ulmer.

Wacner, R. 1907. Zur Morphologie des Peltiphyllum Peltatum (Torr.) Engl.
Sitzungsber. Kaiserl. Akad. Wiss., Math.-Naturwiss. K]., Abt. 1, 116:1089-1107.

FROST SENSITIVITY AND RESPROUTING BEHAVIOR OF
ANALOGOUS SHRUBS OF CALIFORNIA AND CHILE

Harotp A. MOoNEY
Depariment of Biological Sciences
Stanford University, Stanford, California 94305

Detailed comparisons of the structure and function of sclerophyll vege-
tations centered in the matched climatic regions of central Chile and
coastal California have shown a large degree of similarity in spite of
divergent evolutionary histories of the floras of these 2 areas (Mooney
et al., 1970; Mooney et al., 1977; Parsons, 1976). Principal differences
are related to features responsive to land-use treatment which has dif-
fered substantially between regions, particularly in the past century
(Mooney ez al., 1972; Aschmann and Bahre, 1977).

Additional functional differences between these vegetations, particu-
larly in phenological patterns of the woody plants, have been ascribed
in part to the small climatic dissimilarities that exist between regions
(Mooney et al., 1977). The broad aspects of the climate of the 2 regions
are, however, quite s’milar. Monthly rainfall, drought duration, and
mean temperatures can be matched station for station in coastal Cali-
fornia and Chile (di Castri, 1973). Furthermore, the direction and mag-
nitude of climatic changes that have occurred from the Pleistocene to
the present have been quite comparable (Miller et al., 1977).

1977 | MOONEY: FROST & RESPROUTING 7

GL

The principal climatic difference between regions is that Chile has a
more equable thermal environment, evidently because of the more pro-
nounced maritime influence. Frosts are uncommon in coastal Chile but
frequent at equal latitude California stations (Miller e¢ al., 1977). This
could explain the more prolonged flowering and fruiting period of shrubs
in Chile than in California at sites that are otherwise closely matched
(Mooney et al., 1977).

If indeed differences in climatic equability between regions have had
an important evolutionary impact on the vegetation, this might be evi-
dent in their differing frost sensitivities. A spell of unusually cold weather
in California led to a test of this possibility.

During December 1972 there was a record freeze in California of a
duration that had not occurred for the previous 40 yrs. In the San Fran-
cisco Bay region, at Palo Alto, temperatures dipped to — 5 C on 2 sepa-
rate days (U.S. Weather Bureau, 1972). Widespread damage occurred
to exotic plantings in this area.

In 1968 a garden was established at the Carnegie Institution of Wash-
ington, at Stanford University, consisting of a large number of species
of evergreen trees and shrubs characteristic of the Californian chaparral
and of the Chilean matorral. Additionally, a number of subligneous
species, generally drought deciduous, from both regions were planted.
Plants were grown from seed, mostly collected from the same latitude in
California and Chile (33°), thus from comparative generalized mediter-
ranean-type climates (di Castri, 1973). The Carnegie garden is located
at a higher latitude (37°N) than the origin of the study populations;
however, many species extend naturally to these higher latitudes. At the
time of the 1972 freeze these plants had grown to full-sized shrubs.

Considerable damage was noted on many plants subsequent to the
freeze. Observations were made on the extent of their recovery the fol-
lowing summer. For most species 10 individuals were observed.

RESULTS

Only 2 of the 9 California evergreen species were damaged by the
freeze; both were Rhus species (Table 1). Both species are restricted to
latitudes lower than the other Californian evergreens. Rhus laurina, nor-
mally found only at the lowest elevations within the chaparral or within
the coastal sage drought-deciduous community, had the greatest mor-
tality. Both species, however, had individuals that base-sprouted. Re-
sprouting from stems of Rkus ovata was also noted.

Three Californian drought-deciduous species, Artemisia californica,
Encelia californica, and Ptelea aptera, had individuals that were frost
killed, as well as those that recovered by base sprouting. The other 5
Californian drought-deciduous species were undamaged.

Thus, less than one-third of the Californian species were frost dam-
aged. Of these, all had individuals that recovered by base-sprouting.

76 MADRONO [ Vol. 24

TABLE 1. CONDITION OF PLANTS Srx MONTHS SUBSEQUENT TO FREEZE.

CALIFORNIA ORIGIN % No damage % Dead % Resprouting

Evergreen species
Rhus laurina 0) 60 40
Rhus ovata 0 20 80
Adenostoma fasciculatum 100 0 0
Cercocarpus betuloides 100 0 0
Heteromeles arbutifolia 100 0 0
Prunus ilicifolia 100 0 0
Quercus agrifolia 100 0 0
Rhamunus ilicifolia 100 0 0
Umbellularia californica 100 @ 0

Drought Deciduous species
Artemisia californica @ 60 40
Encelta californica @) 40 60
Ptelea aptera 0 40 60
Adolphia californica 100 0 0
Fraxinus trifoliata 100 0 0
Salvia apiana 100 0 0
Salvia leucophylla 100 0 0
Salvia mellifera 100 0 0

CHILEAN ORIGIN

Evergreen species
Azara celastrina 0 0 100
Baccharis paniculata 0 0 100
Beilschmiedia miersti 0 0 100
Colliguaya odorifera 0 0 100
Cryptocarya alba 0 20 80
Escallonia pulverulenta 0 0 100
Lithraea caustica 0 66 33
Schinus latifolius 0 40 60

—
[2)
(@)

Escallonia illinita 0
Quillaja saponaria 100 0 0
Maytenus boaria 0

_
(2)
©

Drought Deciduous species

Flourensia thurifera 0 60 40
Lepechinia salviae 0 75 25
Lobelia chilensis 0 20 80
Lobelia poly phylla 0 60 40
Podanthus mitiqui @) 0 100

0 33 66

Trevoa trinervis

In contrast, over 80% of the Chilean species were frost damaged,
including all the drought-deciduous elements. As with the Californian
species, all the damaged species had individuals that recovered by re-
sprouting. There was a somewhat higher mortality of drought-deciduous
than of evergreen plants, particularly among the Chilean species.

1977 | MOONEY: FROST & RESPROUTING 77

DiIscUSSION AND SUMMARY

There was a considerable difference in tolerance of Chilean and Cali-
fornian species to a substantial freeze. As a group the Chilean species
were considerably more sensitive, as was predicted. It can be inferred
from these results, which are supported by modern climatic data, that
the Chilean species have not been subjected to as cold temperatures as
have those from California, in their native habitat. Thus, although the
general climatic pattern of summer drought and cool, wet winter is com-
parable in the 2 regions, the climates differ in a biologically important
parameter, i.e., frequency of cold temperatures.

This evidently has resulted in the selection of vegetations that, al-
though similar appearing (Mooney e¢ al., 1970; Parsons, 1976) and simi-
larly adapted physiologically (Mooney and Dunn, 1970), differ in certain
significant physiological features.

All Californian and Chilean frost-damaged species recovered by base
or “stump” sprouting. Stump sprouting is often cited as an evolutionary
response to a fire climate. However, as Axelrod (1973) has noted, most
species that are associated with California chaparral today have been in
evistence for over 12 to 18 million yrs.—long before the origin of a
regional summer drought; hence a pronounced fire climate.

It is thus quite likely that resprouting behavior has evolved as a more
general adaptive response to such environmental stresses as cold and
drought, both of which increased in intensity coincident with evolution
of the sclerophyll vegetation (Axelrod, 1973). This survival feature
would thus be a “pre-adaptation” to fire, which would be reinforced as
the mediterranean climate developed fully.

ACKNOWLEDGMENTS
This study was supported by NSF Grant GB27151. D. Axelrod and F. Kruger
provided helpful comments.

LITERATURE CITED

AsSCHMANN, H. and C. Banre. 1977. Man’s impact on the wild vegetation. In H.
Mooney, [ed.], A study of convergent evolution: scrub ecosystems of California
and Chile (in press).

AXELROD, D. I. 1973. History of the mediterranean ecosystem in California. Jn F. di
Castri and H. Mooney, [eds.], Mediterranean Type Ecosystems. Springer
Verlag, New York.

Di Castri, F. 1973. Climatographical comparisons between Chile and the western
coast cf North America. Jn F. di Castri and H. Mooney, [eds.], Mediterranean
Type Ecosystems. Springer Verlag, New York.

Miter, P., D. BrapBury, E. Hayex, V. La Marcue, and N. THRower. 1977.
Macroenvironment. Jn H. Mooney, [ed.], A study of convergent evolution:
scrub ecosystems of California and Chile (in press).

Mooney, H. and E. Dunn. 1970. Convergent evolution of mediterranean climate
evergreen sclerophyll shrubs. Evolution 24:292-303.

————.,, E. Dunn, F. SHropsuire, and L. Sonc. 1972. Vegetation comparisons be-
tween the mediterranean climatic areas of California and Chile. Flora 159:
480-496.

78 MADRONO [Vol. 24

, J. KumMeErow, A. JoHnson, D. Parsons, S. KEELEY, A. HorrMan, R.
Hays, J. Girrperto and C. Cuu. 1977. The producers—their resources and
adaptive responses. Jn H. Mooney, [ed.], A study of convergent evolution:
scrub ecosystems of California and Chile (in press).

, E. Dunn, F. SHROPSHIRE, and L. Sonc. 197-. Land use history of Cali-
fornia and Chile as related to the structure of the sclerophyll scrub vegeta-
tions. Madrono 21:305-319.

Parsons, D. J. 1976. Vegetation structure in the mediterranean scrub communities
of California and Chile. J. Ecology 435-447.

U.S. Dept. CoMMERCE. 1972. Climatological data, California, 76: December, 1972.

TWO NEW LOCAL UMBELLIFERAE (APIACEAE)
FROM CALIFORNIA

Miuprep E. MATHIAS
Department of Biology, University of California, Los Angeles 90024

LINCOLN CONSTANCE
Department of Botany, University of California, Berkeley 94720

Current interest in the possible extirpation of increasing numbers of
restricted populations of native plants in California has focussed atten-
tion on narrow endemics of all sorts, since they are especially vulnerable.
The names of two such examples have been entered on one or more of
the lists of rare or endangered species without having been properly
legitimized. The purpose of this contribution is to describe and illustrate
them, since it seems unwise to delay their debut any longer.

Angelica callii Mathias & Constance, sp. nov. Plantae caulescentes
crassae 1—2 m altae, foliis glabris minute scaberulisve, inflorescentia sca-
berula hirsutulave; folia ovata ternato—1—2—pinnata, divisionibus princi-
palibus interdum reflexis rhacidibus geniculatis, foliolis lanceolatis ovato-
lanceolatisve 3-13 cm longis, 1.5—4 cm latis, acutis vel obtusis, acute ser-
ratis; petioli 0.5—3 dm longi basi anguste vaginantes; folia cauline sursum
reducta plerumque pinnata, foliis summis dilatatis sine lamina; pedunculi
paulo graciles 1-2 dm longi; involucrum nullum; involucellum pler-
umque nullum; radii 25-50 subaequales patenti-adscendentes basi con-
spicue connati 2.5—7(—10) cm longi; pedicelli plures inaequales patenti-
adscendentes basi conspicue connati 5-15 mm longi; flores albi vel sub-
rosei, petalis ovalibus obovatisve basi extra paulo hirsutulis, stylonodio
conico quam stylis gracilibus breviore, ovariis hirsutulis; carpophorum
bipartitum; fructus ovalis obovatusve 3.5—5 mm longus, 2.5—4 mm latus,
hirsutulus glabratusve, costis dorsalibus demissis rotundatis coarctatis
suberosis quam intervallis multo latioribus, costis lateralibus quam eis
dorsalibus multo latioribus suberosis, quam corpore fructus plerumque
angustioribus; vittae ad valleculas solitariae magnae ad pericarpium
adherentes; chromosomatum numerus ” = 11. Fig. 1.

79

MATHIAS & CONSTANCE: UMBELLIFERAE

1977]

g. d, transection of fruit.

seedlin

basal leaf. c,

)

Angelica callit. a, habit. b

Fic. 1,
e, dorsal view of entire fruit. f, fruiting umbellet. All from Call & Call 2459, 2544,

and 2550.

80 MADRONO [Vol. 24

Plants stout, 1-2 m tall, the foliage glabrous to minutely scaberulous,
strongly scented, the inflorescence scaberulous to hirsutulous; leaves
ovate, 1-4 dm long, 1-3 dm broad, ternate-pinnately divided, the main
divisions sometimes reflexed and the rachis geniculate, the leaflets lanceo-
late to ovate-lanceolate, 3-13 cm long, 1.5—4 cm broad, acute to obtuse,
the larger petiolulate and with one or 2 narrow lobes or leaflets at base,
the others sessile, sharply serrate; petioles stout, 0.5-3 dm long, nar-
rowly sheathing at base; cauline leaves reduced upward, mostly pin-
nate, with moderately dilated sheaths, the uppermost sheaths bladeless;
peduncles rather slender, 1-2 dm long; involucre wanting; rays 25—50,
2.5—7(—10) cm long, spreading-ascending, subequal, conspiculously web-
bed; involucel of a few inconspicuous filiform bractlets, or lacking;
pedicels 5-15 mm long, spreading-ascending, unequal, conspicuously
webbed; flowers white or pinkish, the petals oval to obovate, a little
hirsutulous at base dorsally; styles slender, much longer than the conical
stylopodium; ovaries hirsutulous; fruit oval to obovate, 3.5—5 mm long,
2.5-4 mm broad, hirsutulous to glabrate, the dorsal ribs low, rounded,
crowded, much broader than the intervals, the lateral ribs broader than
the dorsal, narrower than to about equalling the body, all corky-thick-
ened; vittae large and solitary under the intervals, apparently with
smaller ones under the ribs, making them continuous about the seed
and adhering to it when and if it becomes loose in the pericarp; chromo-
some number 2 = 11.

Type: California, Tulare Co., on stream banks 2 mi E of Lookout
Guard Station near Sequoia National Park, 4600 ft altitude, 18 Oct
1965, Tracey & Viola Call 2459 (Holotype: UC).

DISTRIBUTION: Stream banks between altitudes of 3800 and 6500 ft,
W slope of Sierra Nevada in Tulare County and adjacent Kern County,
California.

SPECIMENS EXAMINED: Tulare Co.: banks of Bear Creek near Coy Flat E of
Springville, 5000 ft, 6 Nov 1965, Call & Call 2461; banks 0.5 mi E of California Hot
Springs, 3300 ft, 15 Sep 1966, Call & Call 2549; steep E-facing rocky stream bed
above Redwood Meadow ca 15 mi NE of California Hot Springs, 6500 ft, 15 Oct
1966, Call & Call 2550 (garden-grown progeny and chromosome voucher, Constance
693) ; shaded banks of Bear Creek above Coy Flat near Camp Nelson, 3800 ft, 15
Oct 1966, Call & Call 2551. Kern Co.: small stream on N-facing slope 1 mi N of
Greenhorn Summit, 5800 ft, 7 Sep 1966, Call & Call 2544.

This appears to be closest to Angelica wheeleri S. Wats. of Utah, but
is distinguishable by (1) subequal fruiting rays, (2) much less inflated
and hence narrow and inconspicuous upper cauline leaf sheaths, and (3)
lower and broader dorsal fruit ribs. In pubescence, leaf serration, web-

bing of rays and pedicels, and in the shape and pubescence of the ovary |

and fruit, the two are strikingly similar.
This interesting new Angelica was discovered a decade ago by Dr. and
Mrs. Call of California Polytechnic State University, San Luis Obispo,

ee

1977] MATHIAS & CONSTANCE: UMBELLIFERAE 81

who are discerning students and discriminating collectors of Umbelli-
ferae. It has not, so far as we are aware, been secured by anyone else.
The chromosome count was made by Dr. and Mrs. Tsan-Iang Chuang.

Lilaeopsis masonii Mathias & Constance, sp. nov. Plantae perennes
horizontaliter rhizomatosae caespitosae glabrae; folia teretia lineari-
filiformia 1.5—-7 cm longa diametro usque 1 mm, septiis paucis obscur-
isque; pedunculi 7-15 mm longi debiles quam folia breviores; umbellae
3—8—flores; pedicelli adscendentes vel patentesque vel reflexi 2-6 mm
longi; fructus ovoideus 1.5-1.8 mm longus, 1.25—1.5 mm latus, costis
dorsalibus acutis obscuris, eis lateralibus prominentibus latis suberoso-
incrassatis; chromosomatum numerus ” = 22. Fig. 2.

Plants perennial rhizotamous, forming a low turf, glabrous; leaves
quill-shaped, terete, linear-filiform, 1.5-7 cm long, less than 1 mm in
diameter, the septae few and obscure; peduncles 7-15 mm long, weak,
shorter than leaves; umbels 3—8—flowered; pedicels ascending to spread-
ing or reflexed, 2-6 mm long; fruit ovoid, 1.5-1.8 mm long, 1.25-1.5 mm
broad, corky-thickened; chromosome number = 22.

Type: California, Sacramento Co., moist sandy soil with Scirpus and
Equisetum, Twitchell Island, margin of Sacramento River 6.5 mi S of
RioVista, at sea level, 14 Jul 1955, L. Constance & H. L. Mason 3611
(Holotype: UC).

DISTRIBUTION: Low-lying shores of San Francisco Bay at the deltaic
mouths of its tributaries. Mason has recently reported verbally that the
plant occurs on some of the islets elsewhere in the delta, but we have
seen no material.

SPECIMENS EXAMINED: Napa Co.: am Ufers des Napaflusses siidlich von Napa, 24
Jul 1913, W. N. Suksdorf 630 (K, UC). Solano Co.: wet soil at edge of slough with
Triglochin and Juncus, Suisun Marsh ca 1 mi S of Suisun, 19 May 1957, J. M.
Tucker 3332 (CAS, DS, UC). Sacramento Co.: moist sand and mud with Lzmo-
sella, Hydrocotyle, Helenium, Arundinaria, Scirpus and Salix, Sherman Island,
margin of San Joaquin River 0.5 mi E of N end of Antioch Bridge, 14 Jul 1955,
Constance & Mason 3610 (UC, including garden-grown progeny and chromosome
voucher) ; N end of Antioch Bridge, Apr 1954, Mason s. n. (UC). Contra Costa
Co.: lower San Jcaquin River, Antioch, Oct 1942, Mason 12,556 (UC); along
lower river just above Antioch, 12 Jun 1955, P. H. Raven 8292 (CAS).

In his “Flora of the Marshes of California,” Mason (1957) stated:
“On the basis of vegetative characters there appear to be two forms of
this species: (1) the coastal form, extending southward to Marin County,
California, from British Columbia, which has somewhat broad, often
flattened, and conspicuously septate phyllodes; and (2) the San Fran-
cisco Bay and river-mouth form, which has very fine, terete, and only
obscurely septate phyllodes. Additional collections and further study
may show that these merit taxonomic recognition” (p. 631). Also Hill,
in his classic monograph of the genus Lilaeopsis (1927), in citing the
Suksdorf specimen listed above as his only California representative,

82 MADRONO [Vol. 24

Fic. 2. Lilaeopsis masonii. a, habit. b, individual node showing foliage and
inflorescence. c, flower. d, transection of fruit. e, lateral view of entire fruit. All
from Constance & Mason 3610.

1977] HEKTNER & FOIN: COASTAL PRAIRIE 83

remarked its narrower and more slender leaves, but referred it to L.
occidentalis Coult. & Rose because of the agreement in fruit structure
between the California plant and those from further north.

The most critical distinguishing feature between the coastal and delta
plants is that the leaves of the latter are not only narrower, more slen-
der, and usually shorter, but that they are truly terete, have proportion-
ately fewer septae, particularly toward the apex, and that these septae
are so obscure that they are likely to remain unobserved. The fruit
characters, upon which Hill relied so heavily to distinguish species, are
essentially identical. Even chromosome number is of no taxonomic assis-
tance since L. masonii like L. occidentalis, has a complement of m = 22;
both are presumably tetraploid. The taxon referred to as Lilacopsts sp.
in Constance, Chuang & Bell (1976, No. 481, p. 619) is L. mason,

LITERATURE CITED
Constance, L., T.-I. CHuanc and C. R. Bety. 1976. Chromosome numbers in
Umbelliferae. V. Amer. J. Bot. 63:608-625.
Hitt, A. W. 1927. The genus Lilaeopsis: a study in geographical distribution. J.
Linn. Soc. 47:525-551.
Mason, H. L. 1957. A flora of the marshes of California. University of California
Press, Berkeley.

VEGETATION ANALYSIS OF A
NORTHERN CALIFORNIA COASTAL PRAIRIE:
SEA RANCH, SONOMA COUNTY, CALIFORNIA

M. M. HEKTNER and T. C. ForIn
Division of Environmental Studies and Institute of Ecology
University of California, Davis 95616

The northern coastal prairies of California are distributed along parts
of the coastal zone from the California-Oregon border south to Monterey
Bay. Previous authors have outlined the natural history and the distri-
bution of species of the coastal grassland ecosystem (Beetle, 1947;
Burcham, 1957; Munz, 1973; Crampton, 1974; Ornduff, 1974) and a
number of floristic surveys have been completed (Davy, 1902; Penalosa,
1963; Barbour, 1970, 1972; Howell, 1970; Hardham and True, 1972).
Ecological analysis of the coastal grassland, however, has been limited
(Huffaker and Kennett, 1959; Batzli and Pitelka, 1970; Barbour et al.,
1973; Elliott and Wehausen, 1974; Davidson, 1975); Heady et al.
(1977) also reached this conclusion.

In 1974, we began an analysis of the coastal perennial grassland com-
munity at Sea Ranch, Sonoma County, California. The two major goals
of this program are first to document the structure of a coastal grassland
that has not been grazed by livestock for approximately 10 years, and
secondly to develop hypotheses about dominance and diversity relation-
ships suitable for experimental tests.

84 MADRONO LVol. 24

For many years the coastal grasslands have been strongly influenced
by persistent livestock grazing. With increased coastal development for
residential purposes and with the proliferation of park reserves large
sections of the coastal grassland are no longer grazed and are undergoing
changes in species composition and standing crop. We set out to docu-
ment the changes as one such area recovers from grazing. Over the long
term we hope to understand the regulation of dominance, species diver-
sity, and patchiness of vegetation distribution by studying population
parameters of selected species using methods described elsewhere (Foin
and Jain, 1976).

This paper presents the results of our first survey of the Sea Ranch
grasslands in 1974. An annotated species list for the Sea Ranch costal
terraces is being published elsewhere (Hektner and Foin, 1977).

THE STUDY SITE

Sea Ranch (38° 40’ N, 123° 24’ W) is a low density subdivision
approximately 180 km north of San Francisco (Fig. 1). It is situated
along 16 km of the northern Sonoma County coastline and extends up
to 2 km inland. The 729 ha of terrace grassland under study are bounded
by California State Highway 1 on the east, Gualala Point Sonoma
County Park on the north, the Sea Ranch southern boundary line, and
the Pacific Ocean. Approximately 95% of the coastal terrace area has
been reserved as permanent open space by the Sea Ranch Homeowner’s
Association and allowed to develop with minimum disturbance. Hence,
these open areas permit long term studies of the dynamics of the coastal
prairie.

The area that is now Sea Ranch was included in an 8,100 ha Spanish
land grant, known as Rancho German, given to Ernst Rufus in 1846.
Lumbering soon became an important industry and a mill was established
at the mouth of the Gualala River. The grant was eventually broken
up and one of the parcels sold was Black Point (now part of the Sea
Ranch), from which lumber and cattle were shipped to San Francisco
(Morgan and Morgan, 1974). Horses, cattle and sheep grazed the Sea
Ranch area (then known as Del Mar Ranch) continuously until the
mid-1960’s. Parts of the terrace were occasionally plowed for planting
peas, potatoes, and even artichokes. Unspecified species of clover were
also sown by the former owner to provide additional forage for sheep
(Ohlson, pers. comm.). Development began in 1963 in the southern
portion, and pastures were abandoned as it proceeded northward. The
last sheep were removed from the northern section in 1968, but cattle
grazed the north end from 1967 to 1969.

Like most of northern California, Sea Ranch receives most of its pre-
cipitation during late fall, winter, and spring. Records kept by local
residents on the terrace (elevation 22 m) and published in the /ndepen-
dent Coast Observer (a local newspaper) show a 5-year average seasonal

1977] HEKTNER & FOIN: COASTAL PRAIRIE

Ty
\ GUALALA
\RIVER /

“e* Up

/

TMi
my ‘ 4

i. HWY a
« ( ty
| \ Y)

° \ mn YPr <

3 . 2»

7
PAGHP AG c AN \\
WM kc vv anos
ANNUAL

[| PERENNIAL
WY MIXED ANNUAL AND PERENNIAL

y] help
ESEESEE CALAMAGROSTIS

IMM wooDLAND

NO VEGETATION OR CULTIVATED

Fic. 1.

dots represent hedgerows of Cupressus macrocar pa.

Distribution of major vegetation types at Sea Ranch, California. Rows of

85

86 MADRONO [Vol. 24

rainfall of 842 mm (range, 458-1308 mm), with 92% falling from Octo-
ber through March. For the same period (July 1970—June 1975), at the
top of German Ridge (elevation 274 m) 6.4 km to the north, seasonal
rainfall averaged 1609 mm (range, 932—2475 mm).

Fog associated with periods of slack winds during summer contributes
an unknown amount of moisture. Wind direction varies seasonally,
coming mainly from the north and northwest during summer, and from
the south and southeast during winter. Checks of wind velocities during
planning studies for the development of Sea Ranch (Lawrence Halprin
and Associates, unpubl.) during 1954 through 1961 showed that in
general winds of less than 5 koh seldom occurred more than 10% of the
time in any month. Winds in the 6—24 kph class persisted for 50-65% of
every month and 25-30% of any month had winds in excess of 26 koh.

There are no continuous temperature records for Sea Ranch, but mean
monthly temperature records are available for the past 43 years for Fort
Ross, 31 km south. Annual mean temperature is above 11C, with monthly
means ranging from 6.5C in the winter to 15C in the summer (Davidson,
O75

Due to the proximity of Fort Ross to Sea Ranch, the long-term
climatic records for Fort Ross were used to construct a Thornthwaite
climatic diagram (Fig. 2) as an approximation of the Sea Ranch climate.
Climatic data representing 86 years of recorded precipitation and 11
years of temperature (U.S. Environmental Data Service, 1964) were
used to determine potential evapotranspiration and thus calculate the
water balance throughout the year. We used an unpublished computer
program written by Randall and Major of the U.C. Davis Botany De-
partment and calculations follow Black (1966) and Thornthwaite et al.
(1957). These calculations assume 100 mm of water available from soil
storage and that the rate of water removal by plants from the soil is
pronortional to the amount remaining in the soil.

As in most areas having Mediterranean climates, growth is slow dur-
ing winter, peaks in spring and fall and declines sharply during summer.

Most of the terrace area on the coastal side of Highway 1 is nearly
level with slopes of less than 10%. The coastal terrace consists of two
wave-cut benches, both formed during the Quaternary, but subsistence
and erosion have greatly obliterated their boundaries (Moore, pers.
comm. ).

Soils of the terrace areas are predominantly of two types: Baywood
loamy sand and Rohnerville loam (U.S.D.A., 1972). The Baywood series,
generally directly adjacent to the ocean, consists of very well drained
loamy sand formed in wind-modified sandy coastal plain sediments and
soft sandstone. The Rohnerville soils, formed in material weathered from
soft sandstone, are moderately well drained, with a subsoil mainly of
sandy clay. At one point where the terrace is very narrow a small area of
Kneeland loam extends down to the ocean bluff at 15-30% slopes. Be-

1977]

HEKTNER & FOIN: COASTAL PRAIRIE

FT ROSS. CALIFORNIA
38°31N 123°24 W

ELE 354M

200

180

1604

100

80

60

40+

20

Oa at oe
OOo)

MONTHS
sett WATER DEFICIT Ads kA Kk A

olefeletet AVERAGE TEMPERATURE
III] water surpcus

ee ee eee}

Co
PRECIPITATION

4

Y, SOIL MOISTURE UTILIZATION oa 6 O
ZZ POTENTIAL EVAPOTRANSPIRATION
“

WN SOIL MOISTURE RECHARGE

NN

SCALE Ea OF WATER PER MONTH

Fic. 2. Thornthwaite climatic diagram for Fort Ross, California.

(ora)

“I

88 MADRONO [Vol. 24

cause of their thick dark color, granular structure, and generally high
base saturation, these soils are considered typical prairie soils similar to
those of the Midwest (Ornduff, 1974; Burcham, 1957; Barshad, 1946).
In addition there are two small areas of dune sand, both stabilized by
beachgrass (Ammo phila arenaria ).

METHODS AND MATERIALS

The first step in the vegetation analysis was the production of a map
to establish the location of major vegetation types and to estimate the
contribution of each type to vegetative cover (Fig. 1). This map was
constructed from infrared aerial photographs taken by the National
Aeronautics and Space Administration for the U.S. Army Corps of Engi-
neers using an RC-86” focal length camera at an altitude of 6,096 m in
1972, and at 914 m in 1974. Slide projections were superimposed on a
base map and differences in infrared color were used to establish tenta-
tive vegetation boundaries. Boundaries were later verified by ground
survey.

Vegetation units selected for sampling included headlands, shrub (lu-
pine)-dominated, mixed grassland, perennial grassland, and Calama-
grostis-dominated. The criteria used to distinguish each unit were as
follows: headlands—areas along the ocean bluffs with very low vegetative
height, usually 20 cm or less, with abundant forbs; lupine areas—where
lupine cover (Lupinus arboreus ) extended through more than 2-3 con-
tiguous adult individuals (the patch of bushes defined the area); grass-
lands—mixed, when neither annual nor perennial grass relative cover
exceeded 50%, and perennial, those having > 50% relative cover of
perennial grass species; Calamagrostis areas-where those in which the
relative cover of Calamagrostis nutkaensis was > 50%. These five types
were selected because they were quantitatively* important (as deter-
mined from the aerial map) or were an essential element in the succes-
sional sequence, or both. In particular, annual grasslands were not
sampled because they were quantitatively unimportant and because they
appear to be restricted to sites where disturbance from construction or
horse grazing continues to be heavy.

Sampling of the vegetation was conducted during August and Septem-
ber, 1974, at 5 sites ranging from 4,500-16,000 m?. At each site, sample
quadrats were placed randomly within a 10 m distance along parallel
transects themselves placed randomly within a 10 m band, giving one
sample per 100 m?. For each quadrat percentage bare ground and cover
of each species in the quadrat were recorded. Each species was coded for
analysis using mnemonics in Reed et al. (1963) and tabulated using cross
tabulation and Chi? homogeniety tests from the SPSS/FASTABS pro-
gram (Nie et al., 1975). Taxonomy and nomenclature follow Munz
(1973) except for Deschampsia holciformis, which follows Crampton
(1974). Voucher specimens have been deposited in the Botany Denpart-
ment Herbarium of the University of California, Davis.

1977] HEKTNER & FOIN: COASTAL PRAIRIE 89

Quadrat size varied, depending on the mean diameter of the species.
Lupinus arboreus and Calamagrostis nutkaensis were sampled in quad-
rats of 50.25 m® area; bunchgrass, 12.26 m*; and all others, 0.25 m°.
Circular quadrats were used with the differing sized quadrats placed con-
centrically for sampling of all species.

Cover esimates were made using a Domin Index (modified by Major,
after Evans and Dahl, 1955). In this study, the scale was defined as
follows: 1 = 1 or 2 individuals, cover less than 0.01% ; 2 = few individ-
uals, cover less than 0.1%; 3 = several individuals, cover less than 1%;

4 = numerous individuals, cover less than 4%; 5 = cover 5-10% of
the total area; 6 = cover 11-20% of the total area; 7 = cover 21-33%
of the total area; 8 = cover 34-50% of the total area; 9 = cover

51-75% of the total area; 10 = cover 76-90% of the total area; and
11 = cover 91-100% of the total area.

Note that values 1-3 are measures of density; those from 4—11 are
true cover estimates. Total cover for a species was estimated as the per-
centage of the quadrat occupied by a vertical projection onto the ground
surface of all individuals of that species. The Domin Index was also
used to estimate bare ground. For tabulation purposes, Domin indices
were converted to the midpoint value of their corresponding percentage
range; for example, a Domin value of 11 = 91-100% is equivalent to
95.5%.

RESULTS

As might be expected, analysis of the vegetation map (Fig. 1) reveals
that major map units (headlands, perennial grasslands (including Cala-
magrostis areas), mixed grasslands, annual grasslands, and woodland)
are not uniformly distributed along the coastline-inland gradient. To
show this, 61 transect lines perpendicular to the coast were randomly
placed on the map and 30 chosen randomly for sampling of vegetation
type at each of 6 stations 0, 30, 122, 244, 488, and 975 m from the bluff
edge. Chi’ tests for uniform distribution of each type along the inland
gradient revealed that 3 types (perennial grasses, headlands, and wood-
lands) are significantly localized p < 0.025) with the modal frequency
from coastline inland in the order headlands, perennial, and woodland.
Annual grasslands were too infrequently sampled to test, and mixed
grasslands were not significantly localized (p > 0.10), although the
modal frequency is closer to the coast (122 m) than that for the peren-
nial type (244-488 m). More recent surveys suggest that succession
toward greater dominance by perennial grasses is progressing. This anal-
ysis suggests that large scale vegetation unit patchiness is related to
environmental gradients from the coastline inland.

We digitized the map and used the computer to estimate the area
occupied by each vegetative unit. For the entire area of Sea Ranch,
headlands cover approximately 3%, perennial grasslands 33%, mixed
grasslands 11%, annual grasslands 2%, Calamagrostis 2%, woodland

90 MADRONO [Vol. 24

47°, and 2% has no vegetation or is cultivated. For the coastal terrace
only, headlands occupy 10%, perennial grasslands 38%, mixed grasslands
33% , annual grasslands 8%, Calamagrostis 5%, woodland 1%, and 5%
has no vegetation or is cultivated.

The dominants of the 5 vegetation types sampled are given in Table 1.
Dominance in this table is defined on the basis of frequency (> 5% oc-
currence in all samples taken in a particular vegetation type) and rela-
tive cover (> 5%; relative cover is defined as the summation of cover
values of a species, normalized as a percentage of all cover values for all
species). Despite this rather generous interpretation of dominance, no
vegetation unit has more than 4 or less than 3 dominants even though
the least rich type has 24 species. In each unit the small number of domi-
nant species accounts for 62 to 92% of the total relative cover. Domi-
nance is spread over a number of species, with only 3 (Anthoxanthum
odoratum, Holcus lanatus, and Rubus spp.) dominant in more than one

TABLE 1. DOMINANT SPECIES OF FIVE VEGETATION TYPES AT SEA RANCH. Domi-
nants have a frequency > 5 samples and relative cover > 5%. I = introduced
species, N = native, A = annual, P = perennial, G = grass, F = forb, S = shrub
or vine.

Life Relative
Vegetation history Frequency cover
type Taxon pattern (%) (%)
HEADLANDS Aira praecox IAG 100 34.87
Hypochoeris radicata NPF 100 26.47
Lupinus variicolor NPS 90 18.39
Lasthenia chrysostoma NAF 50 12.34
Total 92.07
LUPINUS Lupinus arboreus NPS 96 28.27
Anthoxanthum odoratum IPG 63 16.90
Holcus lanatus IPG 86 16.33
Total 61.50
MIXED Plantago lanceolata NPF 68 23.71
Cynosurus echinatus IAG 76 5.27
Anthoxanthum odoratum IPG 50 13.24
Danthonia pilosa IPG 69 11.64
Total 63.86
PERENNIAL Deschampsia holciformis NPG 65 28.21
Anthoxanthum odoratum IPG 63 26.88
Holcus lanatus IPG 61 13.17
Rubus spp. NPS 36 5.11
Total isso
CALAMAGROSTIS Calamagrostis nutkaensis NPG 100 55,15
Rubus spp. NPS 100 11.03
Veratrum fimbriata NPF 72 10.41

Total 76.59

1977] HEKTNER & FOIN: COASTAL PRAIRIE 91

vegetation type. Among the dominants, perennials are predominant over
annuals (t = 4.01, p < 0.005), with dominance among grasses, forbs,
and shrubs differing from unit to unit (F = 5.11, p < 0.025). Annual
forbs and grasses, perennial forbs, and prostrate shrubs (primarily Lu-
pinus variicolor) are more important near the ocean than inland, where
perennial grasses dominate to the near exclusion of everything else.

There is no significant difference in representation between native and
introduced dominants (t = 0.476, p > 0.50).

Tables 2—6 present a quantitative list of all species in each of the vege-
tation types. For each taxon, we have tabulated the origin (introduced
or native), the frequency of occurrence in samples, the percentage of
relative cover, the mean absolute cover, and the standard error of mean
absolute cover. A factor for converting relative cover to absolute cover
is given in the legend for each table. Absolute cover is defined as the
percentage of total sampled area actually covered by a species, and mean
absolute cover is the sum of all cover values for that species divided by
the number of plots in which it occurs.

By definition, the dominant species discussed above have high fre-
quency and relative cover values, so it is expected that they have high
mean cover values as well. However, there are many more species in
each vegetation type of intermediate or low importance value. In Tables
2—6, these fall into 3 main groups: 1) species with high frequency and
low cover values, i.e., species that are well dispersed; 2) species that are
infrequent but which have high mean cover values; and 3) species that
are both infrequent and low in cover value. The last category includes
rare and unsuccessful species of little relative importance in the grass-
land, while the second is more complex because it includes infrequent but
large species, and highly overdispersed species in the sense of Hairston
(1959). Myrica californica is an example of the first, while Cardionema
ramosissimum, Juncus effusus var. brunneus, and Lasthenia chrvsostoma
are examples of the second.

The patchiness observed between vegetation units (Fig. 1) may also
be seen in the distribution of species among these units. Only 3 species
(Deschampsia holciformis, Hypochoeris radicata, and Horkelia califor-
nica) were found in all 5 vegetation types, compared to 13 species in 4
of the 5 types, 20 in three, 30 in two, 26 in only one.

Within a type there is also considerable heterogeneity. The estimated
standard errors, when converted into coefficients of variation, are lowest
for dominants within each type. There are no evident trends, however,
when nondominant species are considered. Some species are highly vari-
able in mean cover (Anagallis arvensis is the best example); others are
not (Cynosurus echinatus and Danthonia pilosa, Table 4). For those
species occurring in more than one type, the coefficients of variation may
differ widely (e.g., Deschampsia holciformis, Table 2, = 3.627: Table

92 MADRONO [Vol. 24

5, = 0.169). Presumably the size of the coefficient is a function of coloni-
zation and species growth form as well as species interactions, but there
is no clear trend.

TABLE 2. Cover VALUES OF SPECIES OF THE SEA RANCH HEADLANDS VEGETATION
Type. I/N = introduced (I) or native (N); F = frequency of occurrence (percent
of total samples) ; RC = percent relative cover; MC = mean actual percent cover;
SE = standard error of mean actual percent cover; t = species present, relative
cover < 0.01%. When a species only occurs in one plot, mean actual percent cover
and standard error are not applicable and therefore indicated by a dash (—). Ten
plots were sampled at one site. For conversion to absolute cover of each species
multiply relative cover value by 0.79.

Taxon I/N F RC MC SE
ANNUAL GRASSES

Aira praecox ii 100 34.87 27.40 4.76

Bromus mollis I 10 0.25 — —

Festuca dertonensis I 70 0.20 0.23 0.11

Bromus diandrus I 10 te o- -—
ANNUAL Forss

Lasthenia chrysostoma N 50 12.34 19.40 9.22

Plantago hookeriana var.

californica N 40 0.20 0.39 0.11

Orthocarpus pusillus N 30 0.06 0.34 0.17

Clarkia davyi N 10 0.06 — —

Daucus pusillus N 10 0.01 — —

Anagallis arvensis I 10 t — —_

Silene gallica if 10 t — —
BIENNIAL GRASSES

Bromus carinatus N 20 0.01 0.05 0.00
BIENNIAL ForRBS

Gnaphalium purpureum N 10 0.95 — =
PERENNIAL GRASSES

Deschampsia holciformis N 90 O77 0.68 0.78

Hordeum californicum N 90 0.65 O57 0.19
PERENNIAL ForBs

Hypochoeris radicata I 100 26.47 20.80 3.13

Plantago lanceolata I 30 1.02 2.68 2.41

Convolvulus occidentalis var.

saxicola N 30 O07 £50 0.50

Cirsium quercetorum N 10 O25 — —

Eschscholzia californica N 10 025 = ==

Horkelia californica N 10 0.06 — —

Oxalis corniculata I 20 0.07 O32 0.19
Woopy VINES, SHRUBS, AND

SMALL TREES

Lupinus variucolor N 90 18.39 16.06 4.56
LONGEVITY UNKNOWN

Trifolium sp. — 10 t — ==

BarE GROUND — mo 2.44 = oes,

1977] HEKTNER & FOIN: COASTAL PRAIRIE 93

TABLE 3. Cover VALUE OF SPECIES OF THE SEA RANCH LUPINE VEGETATION
Type. I/N = introduced (I) or native (N); F = frequency of occurrence (percent
of total samples) ; RC = percent relative cover; MC = mean actual percent cover;
SE = standard error of mean actual percent cover; t = species present, relative
cover < 0.01%. When a species only occurs in one plot, mean actual percent cover
and standard error are not applicable and therefore indicated by a dash (—). 71
plots were sampled at one site. For conversion to absolute cover of each species

Taxcn I/N F RC MC SE
ANNUAL GRASSES
Bromus diandrus I 14.1 135 1057 9.44
Bromus mollis I 14.1 0.62 4.86 4.14
Aira caryvophyllea if 15.4 0.35 250 1.47
Gastridium ventricosum I 1.4 0.20 — --
Festuca dertonensts I 15.4 On1S 0.93 0.71
Lagurus ovatus I 4.2 0.03 0.08 0.06
Cynosurus echinatus I 8.5 0.03 On 0.11
Hordeum leporinum i 2.8 0.01 0.03 0.21
Briza minor I 1.4 t — —
Lolium multiflorum I 1.4 t _ —
ANNUAL SEDGES AND RUSHES
Juncus bufonius N 2.8 0.12 4.75 DS
ANNUAL Forss
Lotus angustissimus I 7.0 3,02 47.30 PST.
Carduus pycnoce phalus I 1.4 0.10 — —
Madia capitata N 7.0 0.15 2031 3.06
Geranium dissectum I 2.8 t 0.10 0.00
Silene gallica I 2.8 t 0.01 0.00
Lotus micranthus N 1.4 — —
ANNUAL OR BIENNIAL FORBS
Cirstum brevistylum N 1.4 0.10 - —
Cirsium vulgare I 4.2 0.20 5.20 Sub5
PERENNIAL GRASSES
Anthoxanthum odoratum iL 63.4 16.90 29.45 4.92
Holcus lanatus I 85.9 16.33 21.00 3.38
Deschampsia holciformis N 18.3 0.33 ZO 0.88
Lolium perenne I 7.0 0.03 0.40 0.40
Danthonia pilosa I 2.8 t 0.03 0.21
Elymus glaucus N 1.4 t —- —
PERENNIAL SEDGES AND RUSHES
Juncus effusus var. brunneus N 7.0 221 47.21 14.85
Cyperus eragrostis N 1.4 0.62 _ —
Juncus effusus var. pacificus N 3.9 0.42 Sio0 1.29
Carex obnupta N 1.4 0.20 — —
Carex spp. — 2.8 0.03 1.00 1.00
PERENNIAL Fors
Plantago lanceolata I 1575 0.94 6.68 2:20
Rumex acetosella I 14.1 0.82 6.40 2.48
Horkelia californica N 1.4 0.80 _ —
Hypochoeris radicata I 19057 0.61 3.40 1.30
Oenanthe sarmentosa N 2.8 0.54 12-25 Salis;
Cardionema ramosissimum N

Stachys rigida N 1.4 0.10

94 MADRONO [Vol. 24

TABLE 3 (CONT.)

Taxon I/N F RC MC SE
Abronia latifolia N 1.4 0.10 — ~~
Lotus corniculatus I 2.8 0.03 2.00 0.00
Rumex sp. — 1.4 t -— —
Lythrum hysso pifolia N 2.8 t S274 22.60
Mimulus guttatus N 1.4 t -—

Veronica scutellata N 1.4 f a —

Woopy VINES, SHRUBS, AND

SMALL TREES
Lupinus arboreus N 95.8 28.37 2 2 3.87
Rubus ursinus—R. vitifolius N 9.9 2,13 23.86 7.67
Salix spp. N 2e8 O22 8.75 6.75

FERNS
Pteridium aquilinum var.

pubescens N 2.8 0.83 32250 30.50

LoncEvity UNKNOWN
Galium spp. o 2.6 0.20 7.75 (itis
Trifolium sp. — 1.4 0.26 — —
Lotus spp. a 1.4 t — —

BARE GROUND — 21.48 — —

Species cover also varies depending on the time of year. Sampling was
conducted in August and September when annuals had already dropped
seed. Had an earlier sample been taken, comparison would probably
show an increase in cover values for the annual species. A number of
perennial species such as Brodiaea laxa, Calochortus tolmiet and Ranun-
culus californicus had been observed in the study sites during the spring
but by August were not evident. The extent of cover increase of these
species given an earlier sampling date is unknown, but we believe that
their cover importance relative to all other species would remain minimal.
Even at peak activity, these species are not dominants in the vegetation
and therefore might be expected to change relative covers by no more
than 5-10% at most.

Finally, the tables do not adequately reflect the small scale patchi-
ness in each vegetation type. Within the perennial type, for example,
some areas are almost exclusively Deschampsia, while in others Holcus
is dominant, and in others, Anthoxanthum and Danthonia pilosa. Fur-
thermore, there appears to be variation in patch size, from several hun-
dred m’ to less than one, with and without clear boundaries. By combin-
ing samples we have constructed an overall view of species composition
in the grassland.

Wn

1977 | HEKTNER & FOIN: COASTAL PRAIRIE 9

TaARLE 4. Cover VALUES OF SPECIES OF THE SEA RANCH MIXED GRASSLAND VEGE-
TATION Type. I/N = introduced (1) or native (N); F = frequency of occurrence
(percent of total samples) ; RC = percent relative cover; MC = mean actual per-
cent cover; SE = standard error of mean actual percent cover; t = species present,
relative cover < 0.01%. When a species only occurs in one plot mean actual cover
and standard error are not applicable and therefore indicated by a dash (—). 144
plots were sampled from four sites. For conversion to absolute cover of each species
multiply relative cover value by 0.78.

Ten I/N F RC MC SE

ANNUAL GRASSES

Cynosurus echinatus I (ee 15.27, 15.74 1.96
Aira caryophyllea I 67.4 3.45 3.99 O:73
Aira praecox I 4.2 0.55 O25 3.91
Bromus mollis I 62.5 1.48 1.85 0.63
Festuca dertonensis I Wes) 0.29 0.60 0.20
Briza minor J 211 O27 0.40 0.14
Lagurus ovatus I 1.4 0.26 14.50 12.50
Bromus diandrus i 12:5 0.24 1.39 0.81
Avena barbata I 4.2 0.16 2492 2.54
Hordeum leporinum I 0.7 0.02 —~ —
ANNUAL SEDGES AND RUSHES
Juncus bufontus N 1.4 0.02 125 0.75
ANNUAL Forss
Lasthenia chrysostoma N 1.4 0.24 13.50 13:50
Lotus angustissimus I 0.7 0.24 oe a
Sherardia arvensis i 20.1 0.15 0.60 O27
Hemizonia multicaulis N Deak 0.14 5.17 8.95
Silene gallica I 26.4 Ou? 0:35 0.20
Clarkia davyi N 1.4 0.07 SoS 5.30
Anagallis arvensis I S13 0.05 0.13 0.34
Geranium dissectum I 23 0.02 0.50 0.50
Pogogyne serpylloides N 0.7 0.02 — —
Trifolium tridentatum N 0.7 0.02 — —-
Plantago hookeriana var.
californica N ZA t 0.02 0.02
Madia capitata N 1.4 t 0.01 0.00
Navarretia squarrosa N 0.7 t oe —
Trifolium dubium i 0.7 t - —
Vicia benghalensis I 0:7 t -- —
BIENNIAL GRASSES
Bromus carinatus N ioe 023 137 0.63
ANNUAL OR BIENNIAL Fors
Linum bienne I 47.2 0.48 0.79 0.34
PERENNIAL GRASSES
Anthoxanthum odoratum N 50.7 13.24 20:37 2.70
Danthonia pilosa I 69.4 11.64 13.08 1.99
Holcus lanatus I 20.1 0.95 Sal 153
Stipa pulchra N 22.9 0.40 1.38 0.42
Elymus glaucus N 3133 0.25 0.65 0.19
Deschampsia holciformis N 7.6 0.13 137 0.66
Hordeum californicum N 4.2 0.08 1.42 1,22
Lolium perenne I 19.4 0.04 0.16 0.08

Danthonia californica N 0.7 t — —

96 MADRONO [Vol. 24
TABLE 4 (Conr.)
Taxon I/N F RC MC SE
PERENNIAL SEDGES AND RUSHES
Juncus effusus var. brunneus N 2-1 0.31 11.50 8.05
Juncus effusus var. pacificus N 0.7 0.07 — —
Carex spp. N 0.7 t -= —
Eleochoeris palustris N 0.7 t -- —
PERENNIAL FOrRBS
Plantago lanceolata I 68.5 23.11 21.48 esi
Iris douglasiana N 9.7 3.74 20.04 5.61
Hy pochoeris radicata I 34.7 1.24 2.79 0.65
Horkelia californica N 7.6 1.01 11.31 2.65
Corethyrogyne californica var.
obovata N 2.8 0.66 -= —
Achillea borealis ssp. arenicola N 5.6 0.58 8.19 4.99
Convolvulus occidentalis var.
saxicola N 7.6 0.52 3,39 res
Cardionema ramosissimum N 1.4 0.39 22.00 2.00
Rumex acetosella I 17.4 0.31 1.42 0:55
Stachys rigida N 1.4 0.14 8.00 7.50
Aster chiloensis N 1.4 0.08 4.75 3.89
Plantago hirtella var.
galeottiana N OW 0.07 —~ —
Fragaria chiloensis N 1.4 O02 Os 0.98
Oxalis corniculata I 1.4 0.02 1.00 1.00
Sisyrinchium bellum N 1.4 t 0.05 0.00
Lythrum hysso pifolia N Z.1 t 0.02 0.02
Brodiaea coronaria N 1.4 t 0.03 0.02
Epilobium watsonii var.
franciscanum N O7 t — —
Veronica scutellata N 0.7 ft — —
Woopy VINES, SHRUBS, AND
SMALL TREES
Rubus ursinus—R. vitifolius N 278 4.67 13513 1.56
Salix lasiole pis N 0.7 0.85 — =
Lupinus arboreus N 2.1 0.04 1.50 0.50
FERNS
Pteridum aquilinum var.
pubescens N 10.4 1.56 11.70 2.47
Loncevity UNKNOWN—FORBS
Trifolium spp. —- 1.4 t 0.03 0.02
Lotus spp. “= ZA t 0.01 0.00
Galium spp. == Oy. t — a

BARE GROUND

1977] HEKTNER & FOIN: COASTAL PRAIRIE 97

TABLE 5. COVER VALUES OF SPECIES OF THE SEA RANCH PERENNIAL GRASSLAND
VEGETATION Type. I/N = introduced (I) or native (N); F = frequency of occur-
rence (percent of total samples) ; RC = percent relative cover; MC = mean actual
percent cover; SE = standard error of mean actual percent cover; t = species
present, relative cover < 0.01%. When a species occurs in only one plot mean
actual percent cover and standard error are not applicable and therefore indicated
by a dash (—). 181 plots were sampled from 3 sites. For conversion to absolute
cover of each species multiply relative cover value by 0.96.

Taxon I/N F RC MC SE
ANNUAL GRASSES
Aira caryophyllea ii 33.7 1.24 3255 £03
Bromus mollis I 28.2 0.83 3.60 1.34
Aira praecox I (2 0.39 S27 2:30
Lagurus ovatus {i 6.6 0.31 4.50 3.46
Festuca dertonensis I Z221 0.29 1,30 O55
Briza minor I 23.22 0.12 5.19 O52
Cynosurus echinatus I 122 0.06 0.50 O17
Avena barbata I TEA 0.01 0.84 0.09
Bromus diandrus I 3:3 t 0.01 0.00
ANNUAL SEDGES AND RUSHES
Juncus bufonius N OFS) t om --
ANNUAL ForsBs
Sherardia arvensis i 4.4 0.02 0.30 0.11
Trifolium dubium I 17 0.01 0.80 0.58
Silene gallica I hit t 0.10 0.03
Plantago hookeriana var.
californica N 1.7 t 0.20 O.17
Anagallis arvensis I 5.0 t 0.06 0.06
Pogogyne serpylloides N 0.5 t — —
Hemizonia multicaulis N 1.1 t a —
Lotus angustissimus I i £ 0.01 0.00
Galium aparine I i t 0.03 0.02
Geranium dissectum I Oo t 0.10 0.08
Orthocar pus castillejoides N 1.1 t 0.03 0.02
Daucus pusillus N 1.1 t 0.01 0.00
Clarkia davyi N 0.5 t — —
Lotus micranthus N 1.1 t 0.01 0.00
Madia capitata N 1.4 t 0.01 0.00
Trifolium spp. — 0.5 t — —
Vicia benghalensis I 0:5 t — —
BIENNIAL GRASSES
Bromus carinatus N Le t On7 ofa ly)
BIENNIAL Fores
Erechtites prenanthoides if ie 0.09 7.80 7.85
Gnaphalium pur pureum N 3.3 0.01 0.18 0.07
Erechtites arguta I 0.5 t — —
ANNUAL OR BIENNIAL FORBS
Linum bienne I 10.5 0.01 0.33 0.12
PERENNIAL GRASSES
Deschampsia holciformis N O52 28.21 41.61 2.22

Anthoxanthum odoratum Bi 63.0 26.91 32.48 2.48

98 MADRONO [Vol. 24

TABLE 5 (CoONT.)

PERENNIAL Grasses (continued)

Holcus lanatus I 61.3 13.18 20.65 2.43
Calamagrostis nutkaensis N 0.5 O55 — —
Hordeum californicum N (| 0.09 4.00 2.00
Danthonia pilosa I 122 0.07 0.58 0.35
Elymus glaucus N S25 t 0.10 0.01
Danthonia californica N 23 t 0.14 0.01
Lolium perenne I 22 t 0.01 0.00
Festuca arundinacea I 0.5 t —- —
PERENNIAL SEDGES AND RUSHES
Juncus effusus var. brunneus N 6.6 1.59 15.34 9.16
Carex spp. N dee 0.48 6.39 2.34
Carex obnupta N 22 0.29 12.63 5.69
Juncus effusus var. pacificus N 6.1 0.10 15a 0.66
PERENNIAL FORBS
Plantago lanceolata i 50.2 4.92 10.51 1.05
Hypochoeris radicata I 37.0 1.89 5.86 0.94
Tris douglasiana N 20.4 1.59 7.47 1.28
Rumex acetosella I 14.4 0.28 1.85 On.
Fragaria chiloensis N 4.4 0.24 5.13 2.43
Aster chilensis N 0.5 0.09 — —
Cirsium quercetorum N 0:5 0.09 — —
Eryngium armatum N 3.9 0.08 1.86 0.98
Acaena californica N 0.5 0.04 — —
Oxalis corniculata I 2.8 0.01 0.42 0.39
Sisyrinchium bellum N 7 0.01 0:35 0.15
Convolvulus occidentalis var.
saxicola N 1.1 t 0.28 0.02
Horkelia californica N 1.1 t 0.28 0.02
Lotus corniculatus I 1.1 t 0.01 0.00
Cardionema ramosissimum N 0.5 t — —
Scrophularia californica N 0:5 ig —- —
Weopy VINES, SHRUBS, AND
SMALL TREES
Rubus ursinus—R. vitifolius N 35.9 Sch 13.68 1.85
Lupinus arboreus N 15.5 215 13.36 2555
M yrica californica N 0.5 0:55 — a
Lupinus variicolor N 1.1 0.02 2.00 0.00
FERNS
Pteridium aquilinum var.
pubescens N 19.9 2.48 11.98 224
UnKNowWN LONGEVITY—FORBS
Galium spp. a 0.5 0.09 — =
Lotus spp. — 1.1 t 0.01 0.00

BarE GROUND — 5.93 — —

1977] HEKTNER & FOIN: COASTAL PRAIRIE 99

TABLE 6. COVER VALUE OF SPECIES OF THE SEA RANCH CALAMAGROSTIS VEGE-
TATION Type. I/N = introduced (I) or native (N); F = frequency of occurrence
(percent of total samples) ; RC = percent relative cover; MC = mean actual per-
cent cover; SE = standard error of mean actual percent cover; t = species present,
relative cover < 0.01%. When a species only occurs in one plot mean actual per-
cent cover and standard error are not applicable and therefore indicated by a dash
(—). 25 plots were sampled at one site. For conversion to absolute cover of each
species multiply relative cover value by 1.10.

Taxon I/N F RC MC SE
ANNUAL FORBS
Sonchus asper i 8 t 0.03 3.35
BIENNIAL ForBS
Erechtites prenanthoides I 36 22 6.73 3.07
Erechtites arguta I 8 0.02 0.28 O23
ANNUAL OR BIENNIAL FOoRBS
Cirsium vulgare I 4 t -— —
PERENNIAL GRASSES
Calamagrostis nutkaensis N 100 Sous 61.50 Ress)
Elymus glaucus N 24 3.49 15.94 15.91
Holcus lanatus I 100 3.14 3.49 0.88
Anthoxanthum odoratum I 28 0.08 0.32 0.29
Deschampsia holciformis N 20 t 0.03 0.01
PERENNIAL SEDGES AND RUSHES
Carex obnupta N 60 Sai 10.54 3.06
Juncus effusus var. brunneus N 52 0.48 1.01 0.58
Carex spp. N 8 t 0.01 0.00
Juncus effusus var. pacificus N 4 t a —
PERENNIAL FORBS
Veratrum fimbriata N 72 10.41 16.88 1.59
Oenanthe sarmentosa N 60 2413 3.97 0.78
Tris douglasiana N 68 2.13 3.42 1.23
Vicia gigantea N 64 0.74 12. 0.47
Galium trifidum var.
subbiflorum N 48 0.63 1.44 0.83
Stachys rigida N 48 O51 1.18 0.61
Veronica scutellata N 52 0.34 0.71 0.23
Mimulus moschatus N 32 0.29 1,01 0.30
Campanula californica N 64 0.23 0.39 0.17
Sidalcea malvae flora N 4 0.07 — —
Achillea borealis ssp.
arenicola N 4 t — —
Horkelia californica N 4 t — —
Hypochoeris radicata I 4 t — —
Smilacina stellata var.
sessilifolia N 4 t _- —
Woopy VINES, SHRUBS, AND
SMALL TREES
Rubus ursinus—R. vitifolius N 100 L103 £2206 1.79
M yrica californica N 4 O27 — ——

100 MADRONO {Vol. 24

TABLE 6 (CoNnr.)

Taxon I/N F RC MC SE
FERNS

Pteridium aquilinum var.

pubescens N 8 0.28 efits) S78

LoncEviITy UNKNOWN

Trifolium spp. -- 4 t — -~
BARE GROUND — 0.46 — —

DISCUSSION

Floristic comparisons between the Sea Ranch grasslands and other
areas of coastal prairie show a consistent set of characteristic species. In
a 1902 description of the north coast prairie, Davy stated that the pre-
vailing grasses were Danthonia californica, Festuca rubra, Calama-
grostis aleutica (now C. nutkaensis), and Deschampsia caespitosa. These
native perennial grasses have now been joined by a number of introduced
erasses that are an equally important component of today’s grasslands:
Holcus lanatus, Anthoxanthum odoratum, Agrostis tenuis, Festuca arun-
dinacea, and Danthonia pilosa (Beetle, 1947; Penalosa, 1963; Howell,
1970; Batzli and Pitelka, 1970; Crampton, 1974; Elliott and Wehausen,
1974; Heady et al., 1977).

Some of the non-grass species also considered to be indicators of the
coastal prairie are /ris douglasiana, Carex tumicola, Carex obnupta, Ca-
massia quamash var. linearis, Spiranthes romanzoffiana, Viola adunca,
and Juncus effusus. Most of these are more or less restricted to the
coastal area, but various floral descriptions include an even greater num-
ber of species that are also common to the annual grasslands inland.
Some of the more prominent species include Azra caryophyllea, Briza
minor, Avena barbata, Avena fatua, Bromus mollis, Bromus diandrus,
Lolium multiflorum, Brodiaea pulchella, Sisyrinchium bellum, Lasthenta
chrvsostoma, Eschscholzia californica, and Plantago lanceolata.

Note that most of these species were found at Sea Ranch. However,
the data emphasize patchiness over uniformity: patchiness in the distri-
bution of vegetation units, patchiness of species and dominant distribu-
tions from area to area, and patchiness within any one area. Clearly, the
Sea Ranch grasslands are not uniform entities and part of the patchiness
must result from variation in the physical environment. For examole,
Barbour et al. (1973) have shown that soil salinity decreases inland. At
Bodega Head they showed that the higher salinity near the bluffs and
the physical effects of onshore winds affect the distribution of certain
species (Lupinus arboreus, in particular). In drier areas Festuca idaho-
ensis and Danthonia californica may be more important dominants than

1977] HEKTNER & FOIN: COASTAL PRAIRIE 101

Deschampsia holciformis or Calamagrostis nutkaensis (Huffaker and
Kennett, 1959; Crampton, 1974). Other than the restriction of the head-
lands type to exposed bluffs and the Calamagrostis type to very wet-
claypan areas, we have not observed obvious correlations between pre-
sumed soil gradients and grassland vegetation type. In particular, the
distributions of major soil types (U.S.D.A., 1972) and vegetation types
seem unrelated, although Baywood soils do tend to be immediately adja-
cent to the ocean bluffs and the Rohnerville soils further inland.

Soil moisture is probably a factor in the distribution of at least some
species. Since the area is made up of a series of terraces, the inland soils
are older and deeper, and this, together with the fact that the ridge
behind the terrace receives twice the amount of rainfall, suggests that
the terrace areas at the base of slopes receiving the greatest amount of
runoff might be particularly favorable for perennial development. The
southern portion of Sea Ranch, with its narrow terrace, apparently per-
mits enough seepage and runoff to support the stands of Calamagrostis
there. In this case, the Calamagrostis vegetation type indicates the abun-
dance of water.

In addition to the effects of physical environment, we feel that dis-
turbance (grazing and construction) has a large influence on species
composition. This finds some support in the literature. Burcham (1957)
suggests that perennials disappeared from California’s Central Valley
under grazing pressures. Similarly, Clements and Shelford (1939) stated
that three-fourths of the land south of Mt. Shasta, and from the coast
to the foothills of the Sierra Nevada in Northern California, was origin-
ally perennial climax grassland and that replacement by annuals was
largelv caused by overgrazing. Hormay and Fausett (1942) estimated
that °0-100% of the forage available on heavily grazed rangeland con-
sisted of annuals. With perennials remaining green throughout the dry
summer season, they are subject to heavy use, which reduces plant vigor
and leaves space for the increase of less palatable species characteristic
of earlier seral stages (Burcham, 1957).

Huffaker and Kennett (1959) documented an example in Humboldt
County where prior range practices had changed a once-perennial grass-
land dominated by Danthonia californica to one of mostly annuals. By
withholding grazing until seed shattered and by rotating grazing, peren-
nials again increased. More recently, Elliott and Wehausen (1974)
showed that the coastal grassland at Pt. Reyes was highly responsive to
grazing. With increased grazing pressure, there was an increase of exotic
annual species and a decrease in the native, predominantly perennial
vegetation. In particular, perennial species dominant at the Sea Ranch
were prominent when protected from grazing at Pt. Reyes (Descham psia
holciformis, e.g.).

With grazing having ended less than 10 years earlier, we have shown
that the grassland at the Sea Ranch is dominated by perennials, although

102 MADRONO [Vol. 24

no one species has yet established anything approaching uniform domi-
nance. This analysis enables us to make some predictions to be confirmed
by future sampling of the vegetation: we expect the dominant perennial
grasses to continue to spread and increase their cover values, and to
this extent to clarify the picture of succession within the grassland from
annuals to perennials. We expect the restricted areas of annual grasses
to become more scarce and more restricted to areas of disturbance, and
the mixed grasslands to become perennial grasslands within a few years.

ACKNOWLEDGMENTS

Support was provided by grants to the Division of Environmental Studies, U.C.
Davis, from the Rockefeller Foundation, from the University of California Council
for Advanced Studies of the Environment, and from the North Coast Institute. We
thank Cathy Hendel Bowen for her invaluable field assistance, and B. Crampton
and J. McCaskill for aid in species identification. We are grateful to the U.S. Army
Corps of Engineers, especially Mr. Michael Murphy, for use of their NASA aerial
photographs. Special thanks go to the members of the Sea Ranch Association, par-
ticularly G. Wickstead, W. Crooks, R. Kirkwood, C. Kuhn, J. Wingate, and I. Mor-
gan, as well as the North Coast Institute, for providing the research site and general
support. We are indebted to S. K. Jain, J. Major, and G. Webster, for helpful criti-
cism of an earlier draft of this manuscript.

LITERATURE CITED

Barzour, M. G. 1970. The flora and plant communities of Bodega Head, California.
Madronio 20:239-313.

. 1972. Additions and corrections to the flora of Bodega Head, California.
Madrono 21:446-448.

, R. B. Craic, F. R. Dryspare, and M. T. GHIsELIN. 1973. Coastal ecology.
Univ. Calif. Press, Berkeley.

BarsHap, I. 1946. A pedologic study of California prairie soils. Soil Science 61:
423-442.

Barzu1, G. O. and F. A. PiTeLkKA. 1970. Influence of meadow mouse populations on
California grassland. Ecology 51: 1027-1039.

BEETLE, A. A. 1947. Distribution of the native grasses of California. Hilgardia
17:309-357.

Brack, P. E. 1966. Thornthwaite’s mean annual water balance. Silvi-culture Gen.
Utility Libr. Program GU-101:1-20. State University College of Forestry.
Syracuse, N.Y.

Breckon, G. J. and M. G. Barzpour. 1974. Review of North American Pacific coast
beach vegetation. Madrono 22:333-360.

BurcuaM, L. T. 1957. California Range Land. Resource Agency, Div. of Forestry,
State of California, Sacramento.

CLEMENTS, F. E. and V. E. SHELForD. 1939. Bio-ecology. John Wiley and Sons.

CraMPTON, B. 1974. Grasses in California. Univ. Calif. Press, Berkeley.

Davinson, E. D. 1975. Demography of Lupinus arboreus at Bodega Head, California.
Ph.D. dissertation, Univ. California, Davis.

1977] HEKTNER & FOIN: COASTAL PRAIRIE 103

Davy, J. B. 1902. Steck ranges of Northwestern California: Notes on the grasses and
forage plants and range conditions. U.S.D.A. Bur. Plant Ind., Bull. 12:1-81.
Extiott, H. W. and J. D. WeEHAUSEN. 1974. Vegetational succession on coastal

rangeland of Point Reyes Peninsula. Madrono 22:231-238.

Evans, F. and E. Dant. 1955. The vegetational structure of an abandoned field in
southeastern Michigan and its relation to environmental factors. Ecology
36:685—706.

Forn, T. C. and S. K. Jan. 1976. Ecosystems analysis and population biology:
lessons for the development of community ecology. Bioscience, in press.

HarpHaM, C. B. and G. H. True. 1972. A floristic study of Point Arena, Mendocino
County, California. Madrono 21:499-504.

Hairston, N. G. 1959. Species abundance and community organization. Ecology
40:404-416.

Heapy, H. F., T. C. Forn, M. M. HEKTNeEr, M. G. BarsBour, D. W. Taytor, and W.
J. Barry. 1977. Coastal prairie and northern coastal scrub. Terrestrial vegeta-
tion of California (J. Major and M. G. Barbour, eds.). Wiley Interscience, New
York. In press.

HeEKTNER, M. M. and T. C. Forn. 1977. A flora of the coastal terraces of Sea Ranch,
Sonoma County, California. Wasmann J. Biol., in press.

Hormay, A. L. and A. FAusett. 1942. Standards for judging the degree of forage
utilization on California annual-type ranges. Calif. Forest and Range Exp. Sta.
Tech. Note 21:1-13.

HoweEt1, J. T. 1970. Marin flora. Second ed. Univ. Calif. Press, Berkeley.

Hurraker, C. B. and C. E. Kennett. 1959. A ten-year study of vegetational
changes associated with biological control of Klamath weed. J. Range Managm.
12:69-82.

LAWRENCE HaLprin and AssociaTEs. The Sea Ranch: ecology and landscape. Law-
rence Halprin and Associates. 1620 Montgomery Street, San Francisco (un-
published).

Morcan, J. and N. Morcan. 1974. Island on the coast. Oceanic California, Inc., San
Francisco.

Moorer, M. Bainbridge, Behrens and Moore, P.O. Box 3041, Monterey, Calif. 93940.

Muwnz, P. A. 1973. A California flora and supplement. Univ. Calif. Press, Berkeley.

Nir, N., C. H. Hutt, J. G. Jenkins, K. STEINBRENNER, and D. H. Bent. 1975. Sta-
tistical package for the social sciences. McGraw-Hill, Inc. N.J.

Outson, E. Former owner of the Sea Ranch property, then known as the Del Mar
Ranch. General Delivery, Stewart’s Point, Calif. 95480.

Ornpurr, R. 1974. An introduction to California plant life. Univ. Calif. Press,
Berkeley.

PENALosA, J. 1963. A flora of the Tiburon Peninsula, Marin County, California.
Wasmann J. Biol. 21:1-74.

Reep, M. J., W. R. Powett, and B. S. Bat. 1963. Electronic data processing codes
for California wildland plants. U.S. For. Serv. Res. Note PSW-N20.

THORNTHWAITE, C. W., J. MaTuHer, and D. B. Carter. 1957. Instructions and tables
for computing potential evapotranspiration and the water balance. Drexel Inst.
Technol. Publ. in Climatology 10(3) :183-311.

U.S.D.A. 1972. Soil survey of Sonoma County, California. Supt. of Doc., Washing-
ton, D.C.

U.S. ENVIRONMENTAL Data SERVICE. 1964. Climatic summary of the United States,
Climatic summary for 1951 through 1960: California. Washington, D.C., 1964:
Corr. repr. 1972. (Climatography of the United States No. 86-4).

OBSERVATIONS ON ANTHOCARP ANATOMY IN THE
SUBTRIBE MIRABILINAE (NYCTAGINACEAE)

JAMES WILLSON and RICHARD SPELLENBERG
Department of Biology, New Mexico State University, Las Cruces 88003

Flowers of Nyctaginaceae have a uniseriate perianth. The calyx is
petaloid and of short duration throughout the upper portion and mor-
phologically simple compared to the lower portion, which forms a persis-
tent, fleshy to hard and leathery, glabrous or pubescent, often 5(-—10)
ribbed to winged accessory fruit, the anthocarp. Anthocarp is used here
to designate “the accresent perianth base comprising the accessory fruit
enclosing the mature ovary, which is itself an achene or utricle” (Bogle,
1974). The terms anthocarp and diclesium (Lawrence, 1963) are used
synonymously in reference to Nyctaginaceae fruit. We have chosen to
use the former which is traditional in American botanical literature.

Within Nyctaginaceae anthocarp morphology has long been useful in
distinguishing taxa (Standley, 1918; Heimerl, 1934). Recent delimita-
tion of taxa at specific (Smith, 1975) and generic (Galloway, 1975)
levels has also been partially based on anthocarp morphology. Anthocarp
morphology and anatomy have been considered in ecological and taxo-
nomic studies within subtribe Abroniinae (Wilson, 1972, 1974, 1975,
1976; Galloway, 1971, 1975) and in Boerhavia, (Bhargava, 1932; Ma-
heshwari, 1929), but the scope of these investigations does not include
comparative examination of anthocarp anatomy to determine the phylo-
genetic information these data may hold.

This paper reports results of preliminary investigations to, determine
the feasibility of utilizing anthocarp anatomy to elucidate phylogenetic
and taxonomic relationships in subtribe Mirabilinae, tribe Mirabileae.
The subtribe Mirabilinae was selected for preliminary investigation be-
cause we believe it is a natural group and because within this subtribe
anthocarp structure is more diverse than in other subtribes. Morpho-
logical diversity of the anthocarp in subtribe Mirabilinae is expressed
primarily by the formation of ridges or wings from a portion of the
anthocarp wall. Anthocarps of subtribe Mirabilinae may be morphologi-
cally placed into 3 general types: 1) those with smooth walls or slightly
rounded or angled ridges; 2) those with narrow wings acute in cross
section; and 3) those with broad laminar wings (Fig. 1). The anatomical
basis of this morphological diversity among species serves as the focal
point of this study. We felt it was necessary in an anatomically based
preliminary phylogenetic study of the subtribe Mirabilinae to include
not only congeneric species which exhibit one or more anthocarp
types, but species from different genera which exhibit the same general
anthocarp type. The following species were selected: Mirabilis oblongi-
folia (Gray) Heimerl, M. viscosa Cav. and Boerhavia coccinea Mill.

104

1977] WILLSON & SPELLENBERG: ANTHOCARP ANATOMY 105

Fic. 1. Diagrammatic side views and cross sections of 3 general diclesium types.
A, broad membranous wings. B, narrow wings. C, ridges.

exhibit rounded ridges; B. intermedia M. E. Jones and B. spicata Choisy
have angled ridges; B. alata S. Wats. has narrow wings; Selinocar pus
lanceolatus Gray and Ammocodon chenopodiodes Standl. have broad
laminar wings.

METHODS AND MATERIALS

Anthocarps in late stages of maturation were killed and fixed in the
field using FAA (Sass, 1966), then transferred to tetrahydrofuran (THF)
for dehydration (Leuty, 1964). Standard methods of paraffin infiltration,
embedding, and sectioning at 12 »m were employed (Sass, 1966). Sec-
tions were stained in toluidine blue (Feder and O’Brien, 1968), mounted
in Permount, and drawn using a micro-projector. Several anthocarps
from each plant or population were studied. Where advisable, confirma-
tion of anatomical observations was made by examining anthocarps from
geographically distant conspecific populations.

RESULTS

Anthocarp walls and accompanying wings or angles of each species
are constructed from 5 common elements: epidermis, sclerenchyma, poly-
hedral parenchyma, vascular strands, and columnar parenchyma cells.
Raphide bundles are common in all anthocarps but there is no trend in
number or distribution. In Sclinocarpus and Ammocodon no scleren-
chyma is present in the walls between the wings, whereas in species of
Mirabilis sclerenchyma is present in the wall between the angles in a
band discontinuous from sclerenchyma within the angles. In species of
Boert-avia sclerenchyma forms a continuous band, the sclerenchyma in
the an’hocarp wall contiguous with that of the wing or angle bases.

Graphic representation of transverse sections for each species is pre-
sented in Figs. 2-4. These illustrations are meant to replace complete
and routine anatomical descriptions. Instead, a brief description of antho-
carp morphology and citation for voucher specimens deposited at NMC

106 MADRONO [Vol. 24

Fic. 2. Transverse sections of diclesium wings. A, Ammocodon chenopodiodes.
B, Selinocarpus lanceolatus. Cross hatched areas represent intercelllar spaces within
aerenchyma. CC = columnar cells, VT = vascular trace, S = sclerenchyma cylinder.

are followed by description of anatomical features which conspicuously
differ between taxa.

Ammocodon chenopodiodes (Fig. 2A): (USA, New Mexico, Dona
Ana Co., Las Cruces, R. Spellenberg 2183). Anthocarp ca 5 mm long,
wings ca 2 mm broad, glabrate, body sulcate between wings, sparsely
puberulent. Anatomy: A single vascular strand is embedded in each
sulcus region. Adaxial one-third of enlarged wing base filled with aeren-
chyma. Sclerenchyma cylinder in wing base with 3 vascular strands
spaced equidistantly around abaxial edge of cylinder. Columnar paren-
chyma cells extend into each wing section filling wing lamina.

Selinocarpus lanceolatus (Fig. 2B): (USA, New Mexico, Dona Ana
Co., NE tip of county, R. Spellenberg and T. K. Todsen 2640). Antho-
carp ca 6-7 mm long, wings ca 2-3 mm broad, glabrate, body finely
costate, truncate’ at both ends. Anatomy: A single vascular strand is
embedded in each region between the wings.’ Adaxial half of enlarged
hastate wing base filled with aerenchyma. Sclerenchyma cylinder in wing
base with 3 equidistantly spaced vascular strands positioned around ab-
axial edge. Columnar parenchyma cells extend into each wing section
filling wing lamina.

Mirabilis oblongifolia (Fig. 3A): (USA, New Mexico, Lincoln Co.,
3 mi W of Alto, R. Spellenberg and D. Jackson 2657). Anthocarp obo-
void, ca 3-5 mm long, minutely pilose, 5—angled, angles broad and
usually tuberculate, sides coarsely or finely tuberculate. Anatomy: Scler-
enchyma cylinder bordered laterally by sclerified parenchyma, and with
a vascular strand at abaxial edge of cylinder. Columnar parenchyma cells
extend to each ridge tip.

1977] WILLSON & SPELLENBERG: ANTHOCARP ANATOMY 107

—F

wricge?

PA

ic
“3 PUN

ie
Seeere.

c

Sa /
Se eeeusee!
SK
SO '&

ry

Fic. 3. Diclesium ridge anatomy. A, Mirabilis oblongifolia. Ridge anatomy with
accompanying tubercule. B, Mirabilis viscosa. CC = columnar cells, VT = vascular
trace, S = sclerenchyma cylinder.

Mirabilis viscosa (Fig. 3B): (Mexico, Guanajuato, just E of Silas, S
of Cd. Guanajuato, R. Spellenberg 2969). Anthocarp obovoid, ca 5 mm
long, angular, glabrous, densely covered with large coarse tubercules.
Anatomy: Sclerenchyma band in basal region of ridge and contiguous
with a vascular strand along abaxial edge of band. Columnar parenchyma
cells extend into ridge and tubercule.

Boerhavia coccinea (Fig. 4A): (USA, New Mexico, Dona Ana Co.,
NMSU Campus, R. Spellenberg and J. Willson 3735). Anthocarp nar-
rowly obovoid, ca 2.5—4 mm long, rounded at the apex, densely glandu-
lar-puberulent or glandular-pilose, 5—sulcate angles and sulci smooth.
Anatomy: Sclerenchyma band thickened in ridge base, extending acutely
into angle, and contiguous with a vascular strand at abaxial tip. Colum-
nar parenchyma cells extend into angle.

Boerhavia spicata (Fig. 4B): (USA, Arizona, Maricopa Co., 5 mi SE
of Morristown, R. and M. Spellenberg 2646). Anthocarp narrowly obo-
void, ca 2.5 mm long, rounded at the apex, acute to acuminate at the
base, stramineous, 5—angulate, angles from thick to thin, acute, sulci
broad, open and rugulose. Anatomy: Sclerenchyma band broadened in
angle base, projecting obtusely into angle. Four or 5 vascular strands
present about the abaxial surface of the sclerenchyma cylinder. Columnar
parenchyma cells extend into the angle.

Boerhavia intermedia (Fig. 4C): (USA, Arizona, Pima Co., 21.6 mi
SE of Why, R. Spellenberg and J. Willson 3607). Anthocarp narrowly
obpyramidal, ca 2-3 mm long, glabrous, truncate at apex, 5—angulate,

108 MADRONO [Vol. 24

yr)

ag aaa

SS

\)
A
‘N CS
ow e=-
SS S|
iS
ww! SN

K

LE

ANNI

ui

4 all) <vN
aC (S

3g mass (RETA /
W017 20g fats:
yee" LGYGAY EG
MR IS ORY
atta G4
¢ wa ACR, s e’ 7 . So}

Fic. 4. Diclesium ridge anatomy. A, Boerhavia
C coccinea. B, Boerhavia spicata. C, Boerhavia inter-
media. Diclesium wing anatomy. D, Boerhavia alata.
CC = columnar cells. VI = vascular trace, S =
sclerenchyma cylinder.

1977] WILLSON & SPELLENBERG: ANTHOCARP ANATOMY 109

angles smooth, obtuse, sulci narrow, transverse-rugulose. Anatomy: Ob-
tuse tip of sclerenchyma band barely projecting into angle where it is
contiguous with a single vascular strand. Irregular columnar and isodia-
meiric parenchyma cells fill angle.

Boerkhavia alata (Fig. 4D): (Mexico, Sonora, Guaymas, SE section of
city, R. Spellenberg, D. Jackson and D. Martin 2696). Anthocarp ca
4 mm long, sharply and broadly thin-winged, wings ca 1-1.5 mm wide,
sulci narrow and transversely rugose. Anatomy: Sclerenchyma band
thickened and forming a point in wing base, the point contiguous with 5
or 6 vascular strands about it abaxial periphery. Columnar parenchyma
cells extend into the wing filling the interior wing lamina. Peripheral and
laterally bordering the interior columnar cells is a second set of short
columnar parenchyma cells.

DISCUSSION

The anatomical pattern within the wing or angle and corresponding
anthocarp wall for each species suggests that wings or angles are modi-
fications of the basic anthocarp wall structure. They are constructed
from cells with similar position and morphology. The sclerenchymatous
parenchyma of Mirabilis oblongifolia, the peculiar columnar parenchyma
of Bocrkavia alata, Selinocarpus, and Ammocodon, and the aerenchyma
of Selinocarfus and Ammocodon are exceptions and are unique to wings
or angles in those taxa.

Gray (1853) reported of Pentacrophys wrightu A. Gray ( = Acleisan-
thes), subtribe Mirabilinae, that ‘“. . . the superficial tissue of ribs abound
in tubular cells, containing a spirally coiled thread, which is usually dis-
engaged upon the application of moisture...” and that ‘. . . this struc-
ture was found in most Nyctaginaceae .. .” We did not observe this
coiled thread in any species, but we did observe a discharge of mucilage
or mucilage-like material when Boerhavia and Mirabilis anthocarps were
placed in water. In Selinocarpus and Ammocodon the laminar portion of
the wing exfol'ates on contact with water, exposing the columnar cells,
which separate like the bristles of a brush. Phaeoptilum spinosum Radlk.
(subtribe Phaeoptilinae, tribe Mirabileae) possesses a broadly winged
anthocarp somewhat similar to Selinocarpus and Ammocodon but these
neither excrete mucilage nor exfoliate in water. The exact functional and
ecological significance of the mucilage or mucilage-like material in Bovr-
havia and Mirabilis and the exposure of the columnar cells in Selino-
carpus and Ammocodon is not known. However, mucilage excreted by
wetted fruits or seeds may be involved with epizoochory and may be
associated with germination (Kozlowski, 1972) or “carnivory” (Barber
and Page, 1976). Tle rapid separation of the long tubular cells of the
wings of the anthocarps of S¢linocarpus and Ammocodon suggests (in
addition to the importance of wings in seed dispersal) that these cells
function in capillary retention of water near the seed. The absence of

110 MADRONO [Vol. 24

both mucilage excretion and exfoliation in P. spinosum anthocarps sug-
gests that wings in this species are associated primarily with wind
disversal.

Toluidine blue typically stains parenchyma walls pinkish-purple due
to the binding of the stain with carboxylated polysaccharides and poly-
uronides. Walls containing polyphenolic compounds, such as lignified
secondary walls, typically stain green, greenish-blue, or bright blue
(O’Brien and McCully, 1969). The columnar cells in B. coccinea, B.
intermedia, B. spicata, and Mirabilis and the exterior columnar cells of
B. alata stain purple. The columnar cells of Ammocodon, Selinocar pus
and the interior columnar cells of B. alata, however, stain blue. The par-
enchyma tissue adaxially adjacent to the columnar cells of Ammocodon
and Scelinocarpus and the short columnar cells of B. alata also stain
bright blue rather than the purple of the polyhedral parenchyma within
the wing or angle. These reactions indicate that in addition to similarity
in morphology and position, interior columnar cells of B. alata and par-
enchymatous columnar cells of Selinocarpus and Ammocodon have simi-
lar wall composition. Also, the short columnar cells in B. alata and the
columnar cells in Mirabilis, B. coccinea, B. intermedia, and B. spicata
stain similarly and probably contain similar compounds in their walls
or cytoplasm. Species that emit mucilage have purple-staining columnar
cells and those that do not have blue-staining columnar cells. Boerhavia
alata contains both types of columnar cells but those nearest the exterior
surface and initially exposed to water are the purple-staining peripheral
columnar cells; they enclose blue staining columnar cells. Since B. alata,
Selinocar pus, and Ammocodon are in the same tribe (although the first
seems fairly distantly related to the others), the presence of the similar
blue staining columnar cells in Boerhavia may be a result of parallelism.
Blue staining columnar cells in Boerhavia occur only in B. alata. It is not
possible to ascertain whether wings of Selinocarpus and Ammocodon
arose from the purple staining, mucilage-excreting columnar cells through
development of a secondary set of elongate cells (as in B. alata) with
subsequent loss of the mucilage-excreting cells, or in some other manner.
Future studies of the ontogeny and ecology of Mirabilinae anthocarps
may support one of these sequences and thus aid in elucidating phylo-
genetic relationships in these taxa.

The southwest African Phaeoptilum spinosum, the sole member of the
subtribe Phaeoptilinae, has an anthocarp outwardly similar to Selino-
carpus and Ammocodon. Limited material and difficulties in sectioning
precluded a complete anatomical study of the mature anthocarp. How-
ever, it was noted that the laminar portion of the wing is filled with lone,
narrow, columnar cells that stain bright blue as in Selinocarpus and
Ammocodon, these cells arise from polyhedral parenchyma cells, which
also stain bright blue. A single, continuous band of sclerenchyma fibers
is present in the anthocarp wall between the wings and forms a cylinder
in the wing base, as in Boerhavia. Morphological and antomical similari-

1977] WILLSON & SPELLENBERG: ANTHOCARP ANATOMY 111

ties between the anthocarp of P. spinosum and the species with winged
or ridged anthocarps of the Mirabilinae are probably also the result of
parallel evolution. We concur with Nowicke (1970) who suggested that
only a remote connection for Pkaeoptilum and the Mirabilinae is possible
when pollen type, endemism to southwest Africa, and the unisexual
flowers of P. spinosum are considered.

ACKNOWLEDGMENTS
We thank Chief Scientist Richard Duncan for permission to collect on White
Sands Missile Range, Thomas K. Todsen for his time spent accompanying us there,
and the herbaria at Kew and University of California, Berkeley, for supplying
fruits of Phaeoptilum. Partial support of field work was received from N.M.S.U.,
College of Arts & Sciences, Mini-Grant 3103-121.

LITERATURE CITED

Barber, J. T. and C. R. Pace. 1976. Carnivorous seeds? Bull. Ecol. Soc. Amer.
AIBS Meeting Abstr. 57:24.

Buarcava, H. R. 1933. Contributions to the morphology of Boerhavia repanda. J.
Indian Bot. Soc. 11:303-326.

Bocte, A. Linn. 1974. The genera of Nyctaginaceae in the southeastern United
States. J. Arnold Arb. 55:37 pp.

Fever, N. and T. P. O’Brien. 1968. Plant microtechnique: some principles and
new methods. Amer. J. Bot. 55:123-142.

GaLtoway, L. A. 1971. Systematics of the North American desert species of Abronia
Jussieu and Tripierocalyx (Torrey) Hooker (Nyctaginaceae). Ph.D. diss. Texas
Tech Univ., Lubbock.

————. 1975. Systematics of the North American desert species of Abronza and
Tripierocalyx (Nyctaginaceae). Brittonia 27:328-347.

Gray, A. 1853. New genera and species of Nyctaginaceae, principally collected in
Texas and New Mexico. Auer. J. Sci. II, 15:259-263.

Hermert, A. 1934. Nyctaginaceae. In Engler and K. Prantl, Die nattirlichen Pflan-
zenfamilien, ed. 2, 16(c) :86-134.

KoztowskI, T. T. 1972. Seed biology, Vol. I. Importance, development, and germi-
nation. Academic Press, New York.

Lawrence, G. H. M. 1963. Taxcnomy of vascular plants. MacMillan Company,
New York.

Lreuty, S. J. 1964. Rapid dehydration of plant tissue for paraffin embedding;
tetrahydrofuran vs. t-butanol. Stain Technol. 44:103-104.

Mauesuwart, P. 1929. Contributicns to the morphology of Bocrhavia diffusa. J.
Indian Bot. Soc. 8:219-234.

NowIcKE, JOAN W. 1970. Pollen morpholegy in the Nyctaginaceae I. Nyctagineae
(Mirabileae). Grana 10:79-88.

O’Brien, T. P. and M. E. McCutty. 1969. Plant structure and development. Mac-
Millan Ccmpany, London.

Sass, J. E. 1966. Botanical microtechnique, 3rd ed. Iowa State Univ. Press, Ames.

SmiTH, J. M. P. 1975. Taxonomic study of Acleisanthes (Nyctaginaceae). M.A.
Thesis, Univ. Texas, Austin.

Witson, RurtH C. 1972. Abronia: I. Distribution, ecology, and habit of nine species
of Abronia found in California. Aliso 7:421-437.

——. 1974. Abronia: II. Anthocarp polymorphism and anatomy for nine species
of Abronia found in California. Aliso 8:113-128.

—. 1975. Abronia: III. Pericarp and seed coat anatomy ‘and its ecological

implications for nine species of Abronia. Aliso 8:289-299.

. 1976. Abronia: IV. Anthocarp dispersibility and its ecological implications

for nine species of Abronia. Aliso 8:493-506.

TAXONOMY OF BEBBIA (COMPOSITAE: HELIANTHEAE)

Mo.Ltity WHALEN
Department of Botany, The University of Texas, Austin 78712

Bebbia is a genus of Mexico and the southwestern United States with
its center of diversity in Baja California, where all 3 of its taxa occur. It
is named for the botanist Michael S. Bebb, who worked on Salix. Bebbia
is a small genus of little economic importance. Since it was proposed by
Greene in 1885, it has been largely ignored, except for some controversy
regarding its tribal position. A taxonomic treatment is provided here
with keys and distribution maps.

GENERIC RELATIONS

The generic position of what has come to be known as B2bbia has
been questionable since it was first described as Carphephorus sect.
Kuhnioides (Eupatorieae) by Asa Gray (1873). Greene renamed the
taxon as a distinct genus, Bebbia, in the Heliantheae. It has certain
superficial characteristics of the Eupatorieae, particularly a long plumose
pappus and striate involucral bracts. However, these characters are also
found, although less frequently, in the Heliantheae. Other characters of
Bebbia clearly unite it with the Heliantheae: the persistent chaff sub-
tending each floret, the Helianthoid achenes, the style branches, and the
deen yellow corollas.

King and Robinson (1970) note that the Eupatorieae can be easily
separated from the Helenieae—Heliantheae complex on the basis of micro-
characters. They describe anther appendages of Helenieae-Heliantheae
as keeled with the abaxial surface concave and with a thickened central
region. They also state that members of this group often have glands
(not vis*ble with a dissecting microscope) on the abaxia! surface of the
anther anpendages. They report clusters ‘of such glands in Bebbia juncea.
Although I do not know how these characters hold up throughout the
Heliantheae, King and Robinson’s description of the anther appendages
in Bebbia is correct. I examined a number of specimens of each of the 3
taxa in Bebbia and usually found a cluster of glands (rarely 1 or 2) on
the abaxial side along the thickened central region of the anther appen-
dage. To the extent that these characters are valid in distinguishing the
Eupatorieae from the Heliantheae, they support the placement of Bebbia
in the Heliantheae.

Naturally occurring acetylenic compounds are of taxonomic signifi-
cance within the Compositae (Bohlmann et al., 1973). Recent examina-
tion of B. juncea by Bohlmann (pers. comm.) has established the pres-
ence of one acetylenic compound in the roots:

a Te ae (CH:);—-CH=CH..

112

1977] WHALEN: BEBBIA 113

This compound is the most common thus far detected in the Heliantheae,
occurring in 11 species of Helianthus, 3 of Tridax, 4 of Galinsoga
(Bohlmann et al., 1973), and in one species of Schistocarpha (Bohl-
mann et al., 1976). However, it was not found in any of the 35 species
of Eupatorieae examined (Bohlmann et al., 1973). Thus, there is chemi-
cal as well as morphological support for the placement of Bebdza in the
Heliantheae.

Various suggestions have been made as to the position of Bebbia
within the Heliantheae. It has been considered to belong to the Verbe-
sininae or Madiinae (Greene, 1885) or Galinsoginae (Hoffmann, 1894,
who followed Bentham, 1873; Torres, 1968). Stuessy (1977) places it in
his Neurolaeninae, which he considers related to the Galinsoginae, the
former distinguished from the latter in having alternate leaves, phyllaries
of unequal length (increasing inward), lanceolate paleae, and setose pap-
pus. Phyllaries of unequal length, lanceolate paleae, and seiose pappus
(characteristic of Bebbia) are found in members of Stuessy’s Galinsog-
inae as well as his Neurolaeninae. Furthermore, Bebbia does not have
alternate leaves. (The presence of alternate leaves is not diagnostic for
the Neurolaeninae anyway since of the 2 core genera, Neurolaena has
alternate and Schistocar pha has opposite leaves.) Thus, I believe Bebbia
fits well within the Galinsoginae even as constituted by Stuessy. Bebbia
appears close to Tridax (Galinsoginae) and not particularly close to
Schistocarpha or Neurolaena, which differ from Bebdza in floral charac-
ters, 1.e., in their truncate style branches, non-plumose pappus, glabrous
achenes without distinct faces, and more chartaceous involucre. Bebbia
is also related to Dyscritothamnus, Clappia, Pseudoclappia, and Varilla,
which are treated as a distinct subtribe of the Heliantheae, the Varillinae,
by Turner and Powell (1977) but are included in the Neurolaeninae by
Stuessy (1977). If these genera, along with Schistocarpha and Neuro-
laena, constitute a natural group within the Heliantheae as Stuessy
suggests, then Bebbia would appear to be transitional between the 2 sub-
tribes, and it might prove taxonomically more expedient to recognize but
a single large subtribe, the Galinsoginae, to include both Stuessy’s Neuro-
laeninae and Turner and Powell’s Varillinae.

The genera to which Bebdia probably is most closely related are Dys-
critothamnus (Neurolaeninae) and Tridax (Galinsoginae), both of which
may have radiate or eradiate heads. Bebbia and Dyscritothamnus are
similar in habit and in many aspects of floral morptology (Table 1).
Turner and Powell (1977) consider Dyscritothamnus, Varilla, Pseudo-
clappia, and Clabpia to form a tightly knit group, alike in their succu-
lent habit, distribution, and ecology, and having similar overall leaf and
floral morphology. However, Dyscritothamnus seems to be more similar
to Bebbia than the other genera in having a more pronounced plumose
pappus, slender and more recurved style branches not appendaged at the

114 MADRONO [Vol. 24
TABLE 1. COMPARISON OF Dyscritothamnus AND Tridax witH Bebbia.
Bebbia Dyscritothamnus Tridax

Bareniial shrubs

Pappus setose, uniseriate,
plumose with lateral
processes 0.1-0.3 mm
long

Inner phyllaries wholly
chartaceous, straw-col-
ored with orange striae

Chaff with orange striae

Outer phyllaries
herbaceous

Lobes of disk corolla
pubescent, tube densely
glandular

Leaves opposite through--
out or opposite below
and alternate above

Perennial shrubs

Pappus setose, uniseriate,
plumose with lateral
processes 0.6-1.0 mm
long

Inner phyllaries wholly
chartaceous, stramineous
with orange striae

Chaff with orange striae

Outer phyllaries
subscarious

Lobes and tube of disk

corolla glabrous

Leaves alternate

Annual or weak-stemmed,

perennial herbs

Pappus uniseriate, usually
plumose, scales or bristles

Inner phyllaries partially
herbaceous, light green,
often with purple mar-
gins, or wholly purple

Chaff usually with yellow-
green striae

Outer phyllaries
herbaceous

Lobes of disk corolla
usually pubescent, tube
pubescent or glabrous

Leaves opposite, or rarely
alternate above and

opposite below

top, more distinctly 3—angled, obpyramidal achenes, and similar chaff
and phyllaries (Table 1). Bebbia also appears to be close to Tridax
(Galinsoginae) and to approach 7ridax in many of the characters that
distinguish it from Dyscritothamnus (Table 1). Also, Tridax often has
glandular-tipped trichomes on the peduncles and outer phyllaries, as are
found in Bebdia atriplicifolia. These are not present in Dyscritothamnus
and related genera discussed above. The achenes and floral morphology
are similar in Bebbia and Tridax, although there is considerable vari-
ability within Tridax itself. Bebbia can be readily distinguished from
both Tridax and Dyscritothamnus by the several characters listed in
Table 1.

A chromosome number of 2” = 9 II has been consistently reported
for the 3 taxa in Bebbia: B. atriflicifolia (Turner et al., 1973), B. juncea
var. juncea (Turner et al., 1973), and B. juncea var. aspera (Powell
and Turner, 1963; Solbrig et al., 1972; Sonora, Guaymas, M. Whalen
195, TEX). Chromosome numbers also support an affinity of Bebbia
with the Galinsoginae and Neurolaeninae, genera of which are primarily
based on x = 8 and 9 (Stuessy, 1977). Chromosome counts for Dyscri-
tothamnus have not been reported. Tvidax has base chromosome num-
bers of x = 9 and 10 (Powell, 1965).

1977] WHALEN: BEBBIA £15

TAXONOMIC ‘TREATMENT
Bebbia was originally proposed by Greene (1885) to house 2 species:
B. atriplicifolia and B. juncea, the latter containing 2 varieties. I. M.
Johnston (1924) subsequently reduced B. atriplicifolia to a variety of
B. juncea. For the reasons discussed below, I believe Greene’s original
treatment delimits better the natural taxa within Bebdza.

BEBBIA Greene, Bull. Calif. Acad. Sci. 4:179. 1885.—Car phe phorus, sect.
Kuhnioides Gray, Proc. Amer. Acad. Arts 8:632. 1873, LECTOTYPE
(chosen here): Bebbia juncea (Benth.) Greene = Carphe phorus
junceus Benth.

Perennial, strongly-scented, often leafless shrubs, low and spreading
with upright annual branches or forming dense, rounded masses; the
root thick, fusiform, woody, 1—5 cm in diameter (the basal portions with
gray fissured bark); stems ribbed, glabrous or pubescent, leaves sparsely
to moderately pubescent, sessile or petiolate, opposite throughout or op-
posite below and alternate above, blades linear to triangular and lobate,
the margins entire to laciniately dentate; hairs of stems, leaves, and
involucre simple, white, tuberculate—based, antrorsely-curved, 0.1—0.6
mm long (sometimes interspersed on the peduncle with glandular tri-
chomes); heads discoid, 25—70 florets, campanulate, solitary or form-
ing loose corymbose capitulescences; receptacle convex, chaffy; chaff
5.0-8.5 mm long, scarious, stramineous with red-orange striations and
sometimes red-tipped, persistent, lanceolate, partly enclosing the achene;
phyllaries 3-5 seriate, graduate, striate, unequal in length; outer series
pubescent, herbaceous with chartaceous margins, inner series longer, be-
coming more glabrous and chartaceous; disk florets regular, the corolla
6.5-10.0 mm long, having a short glandular tube 1-2 mm long, a nar-
rowly funnelform limb, and 5 broadly ovate, pubescent lobes; style
branches ca 2 mm long, slender, exserted, recurved, with stigmatic lines
on the inner surface running from base to tip; anthers linear-oblanceolate,
tapered to the base and with ovate-acute apical appendages ca 0.5 mm
long; achenes 2.0-3.5 mm long, compressed, 3—angled, obliquely clavate
with evident epigynous disk, black when mature, pubescent with ascend-
ing white hairs 0.2-0.6 mm long; pappus uniseriate, of 15-30 slender,
white, subequal, plumose bristles as long as or longer than the corolla;
base chromosome number, « = 9. (Figs. 1, 2).

Key to taxa of Bebbia
Leaves withot distinct petioles, the blades narrow, linear to linear-
oblanceolate, entire or with one to a few lobes; peduncles and outer
phyllaries without glandular hairs; perennial shrubs forming glo-
bose masses; heads solitary or few-clustered in loose corymbs borne
onlongbranches. . . . ..... . . . . LB. juncea.

116

B.
B

atriplicifolia
juncea

O var. juncea

® var. aspera

@ intermediates

MADRONO

Fic, 1,

[Vol. 24

Distribution of Bebbia.

1977] WHALEN: BEBBIA by,

Fic. 2. Representative floral sketches of Bebbie (from B. juncea var. aspera,
Clokey 8192, TEX). a, disc achene. b, disc corolla. c, style branches. d, stamens.
e, chaff.

Involucral bracts imbricate and obtuse, rounded or mucronate, not
acute; outer phyllaries ovate, 1-3 mm long, 1-4 mm wide, usu-
ally as wide as or wider than long; heads usually solitary, some-
times borne in a small corymb; central and northern Baja Cali-
FOIMIas &.) . . . . la. B. juncea var. juncea.

involucral bracts Aeure atl not markedly imbricate; outer phyllaries
narrow, lanceolate, longer than wide, !—3 mm long, 0.7—-1.5 mm
wide; heads solitary or more often in open corymbs of a few
heads; northern Baja California, Sinaloa, Sonora, and south-
western United States . . . . . . Ib. B. juncea var. aspera.

Leaves with distinct petioles, blades hastate to laciniate-triangular with
coarsely dentate margins; peduncles and outer phyllaries with glan-
dular hairs; plants clambering, flat-topped, often scandent or semi-
scandent on shrubs, trees, or rocks; capitulescence usually of many
heads clustered in a compact corymb on short peduncles ca 1.5 cm
in length; southern Baja California . . . . . 2. B.atriplicifolia

1. BEBBIA JUNCEA (Benth.) Greene, Bull. Calif. Acad. Sci. 4:180.
1885.—Carphephorus junceus Benth., Bot. Voy. Sulph. 21. 1884.—
Type: Mexico, Baja California, Magdelena Bay, 1844, Hinds s.n.
(Holotype: K, not seen; photograph of holotype, GH! ).

Bebbia filifolia Jones, Contr. West. Bot. 18:80. 1933.—Typr: Mexico,
Baja California, Loreto, Cayuca Ranch, 23 Oct 1930, Jones 27783
(Holotype: CAS!).

118 MADRONO [Vol. 24

Perennial shrub with a dense crown of intricately branched, nearly
leafless stems forming a rounded bush, 0.5-3.0 m high; upper stems
lithe, woody, glabrous to hispidulous-pubescent, divergent at an angle
of (10)30-45(55) degrees; leaves sessile, opposite below, alternate and
often reduced to subulate bracts above, the blades narrow, linear to
linear-oblanceolate, entire or with 1—5 deltoid to linear lobes, 0.5—6.0 cm
long, 1-6 mm wide, moderately pubescent to canescent; peduncles 1.5—
10.0 cm long, pubescent with simple, white trichomes; heads solitary or
in small corymbs (2-5 heads) borne on the upper branches which are
5-8 cm long; outer phyllaries canescent; corollas yellow.

DISTRIBUTION (Fig. 1): Mostly in the Sonoran desert of northwestern
Mexico and adjacent southwestern United States.

Blake (1945) took B. filifolia to be synonymous with B. juncea, i.e.,
he noted Jones’ type specimen to be “B. juncea approaching var. aspera’.
The type of B. filifolia, however, appears to belong to B. juncea var.
juncea since it has the more imbricate involucre and mucronate or
rounded phyllaries characteristic of this taxon and was collected south
of Comondu in the southern part of the range of var. juncea.

la. BEBBIA JUNCEA (Benth.) Greene var. JUNCEA.

Bush 1-3 m high; stems smooth, glabrous to lightly pubescent; heads
usually solitary, 1.0-1.7 cm long, 1.1-2.0 cm wide; phyllaries strongly
imbricate in 3—5 series, obtuse, rounded or mucronate; outer bracts ovate,
farinose, 1-3 mm long, 1-4 mm wide, usually as wide as or wider than
long; inner bracts longer, narrower, 1-3 mm wide, 1.5—6.0 mm long, usu-
ally longer than wide (Figs. 3, 4a, 4b).

DISTRIBUTION (Fig. 1): Baja California from El Pilar (24°28’N) to
Cedros Island (28°23’N), both on the peninsula and on islands offshore
Common on rocky or sandy soil between 5 and 1000 m, usually in washes
and arroyos but also in the oven desert. Flowering year round.

Plants of var. juncea collected south of Muleie, in the area of the type
locality, have fewer (15-20) and broader phyllaries than those of plants
to the north, which have 20-30 bracts per head. The leaves also tend to
be larger in the southern forms. Heads of plants collected from basaltic
cliffs of the Sierra Giganta above Pt. Escondido are particularly large,
when pressed 2.5 cm wide and 1.3 cm long, with 50-60 florets per head.

lb. BEBBIA JUNCEA var. ASPERA Greene, Proc. Calif. Acad. Sci.4:180.
1885.—Bebbia aspera (Greene) A. Nels., Bot. Gaz. 37:273. 1904.—
Type: Greene cited “‘southeastern borders of California, and adjacent
Arizona”. Neotype (chosen here): United States, Arizona, Greenlee
Co., mountains near Clifton, 3 Sep 1880, FE. L. Greene s.n. (NY!).
Bush 0.5—1.5 m high; stems smooth and lightly pubescent or more
pubescent and hispidulous; peduncles 1.5-6.0 cm long; beads 0.4—1.7
cm long, 0.4-2.0 cm wide, solitary or clustered in loose corymbs (2-5
heads) borne on long upper stems; involucral bracts in 2-4 series,
weakly imbricate and acute; outer bracts narrow, lanceolate, canescent,

1977] WHALEN: BEBBIA 119

i

Pilati

Fic. 3. Habit sketch of Bebbia juncea var. juncea (T.S. Brandegee s.n., UC).

1-3 mm long, 0.7-1.5 mm wide, usually longer than wide; inner bracts
narrow, linear, longer, usually 3—7 mm long, 0.7-2.0 mm wide (Figs. 2,
Ac).

DISTRIBUTION (Fig. 1): Northern Baja California, Sonora, and Sina-
loa; southern California, Nevada, Arizona, southwestern Utah, New
Mexico, and far western Texas where it is rare in the Franklin Moun-
tains near El Paso. Common on rocky or sandy soil between 15 and
1450 m; in washes, canyons, and rocky stream beds in the desert, and on
dry hillsides and gravel slopes. Flowering most of the year.

120 MADRONO LVol. 24

I cm.

Fic. 4. Heads of Bebbia juncea. a, B. juncea var. juncea from the northern part
of its range (A. L. Haines s. n., UT). b, B. juncea var. juncea from south-central
Baja (Mason 1895, US). c, B. juncea var. aspera (Clokey 5948, UT).

Nelson (1904) elevated Bebbia juncea var. aspera to specific rank for
the following reasons: ‘I believe that no one who will take the trouble
to compare the description of B. juncea from Cedros Island, which fur-
nished the type, with the full description of the inland forms will ques-
tion their distinctness.’’ Geographical intergrades between vars. juncea
and aspera link the two taxa. Therefore, I agree with Greene that this
taxon should be treated as a variety and not as a distinct species. Mor-
phological intermediates are designated on the distribution map (Fig. 1).

Imbrication and shape of the involucral bracts, and not stem pubes-
cence or size and number of heads, which Greene used to delimit the
two varieties, are the most reliable characters in distinguishing these
two taxa. Roughness of the stems is particularly variable. Plants of both
varieties in Mexico have smooth, glabrous or almost glabrous stems,
while in the United States plants may be smooth or rough-stemmed.
There is a general trend for plants of var. aspera in the northern part of
its range to be more hispid and to have smaller, more numerous heads,
which are usually not solitary, but rather are borne in few-headed cor-
ymbs. Plants from Sonora and Sinaloa are closer to var. aspera, but occa-
sionally approach var. juncea, especially in characters of the involucre.

2. BEBBIA ATRIPLICIFOLIA (Gray) Greene, Bull. Calif. Acad. Sci. 4:181.
1885.—Carphephorus atriplicifolius Gray, Proc. Amer. Acad. Arts
5:159. 1859.—Bebbia juncea var. atriplicifolia (Gray) I. M. Johnst.,
Proc. Calif. Acad. Sci., IV. 12:1197. 1924-—Typr: Mexico, Baia
Calif., Cape San Lucas, Aug 1859—Jan 1860, L. J. Xanthus 47 (Holo-
type: GH! Isotype: US!).

Plants with a stout woody base, forming dense, intricate, sprawling,
flat-topped masses, 0.5—3.0 m wide, self-supporting or supported by other
shrubs; stems brittle, pubescent, upper stems divergent at an angle of
40-85 degrees; leaves opposite, with distinct petioles, 3-17 mm long,
blades hastate to triangular, 2.5 cm long, white tomentose, with coarsely
dentate margins; capitulescence projecting above the plant, corymbose

1977] WHALEN: BEBBIA 121

wn wi
aie \
s 4 ive xs ' x a
ANN
ae ss
RR ‘ :
Sy
\ = SS ay
: gt
ea ey?
V4 7
‘yy

: .

SS _—
=> Re oon 4
Arr,

ASS

y I y)
eH :

Fic. 5. Habit sketch of Bebbia atriplicifolia (Carter 2261, UC).

with a clearcut branching system of usually 5-25 heads clustered on
short peduncles; peduncles 0.3—4.0 cm long, on the average 1.5 cm in
length, with stalked, glandular hairs and simple trichomes; heads 0.9—2.0
cm long, 1.0—2.5 cm wide with involucral bracts broadly acute, in 2—4
series, not markedly imbricate; the outer rows herbaceous, 1.3—5.0(—7.0)
mm long, moderately pubescent with both simple and glandular hairs;
inner rows more chartaceous and glabrous with the innermost bracts
4—9 mm long; involucral bracts and chaff often red-tipped; corolla yel-
low-orange to orange-red (Fig. 5).

122 MADRONO [Vol. 24

DISTRIBUTION (Fig. 1): Common in gravelly or sandy soils between
5 and 1450 m from the southern tip of Baja California at Cape San
Lucas north to Comondu (26°03’N); in washes and on dunes and gra-
nitic bluffs near the ocean, also frequent in arroyos and on rocky hill-
sides. Flowering year round.

Plants of this species are larger-headed to the north. At the southern
margins of its range the heads of B. atriplicifolia have from 25-30 florets
with average head size 1.3 cm long by 1.2 cm wide. At the northern
extreme of its distribution, large-headed forms occur with from 35-60
florets per head and a usual head size of 2.5 cm long and 1.2 cm wide.

I. M. Johnston reduced B.. atriplicifolia to varietal rank because ‘‘Al-
though the two forms |B. atriplicifolia and B. juncea| seemed distinct
in the field, a study of the material in the Brandegee Herbarium has
seemed to substantiate Mr. Brandegee’s statements (Proc. Calif. Acad.
Sci. IT, 2:180. 1889, and Zoé 1:271. 1890) that the forms approach each
other too closely” (Johnston, 1924). These plants would appear more
distinct in the field because they differ in habit. Furthermore, based on
my study of herbarium material, including that from the Brandegee
Herbarium, the two taxa appear to be good species readily distinguish-
able by their habit, characters of the leaves and inflorescence, and the
presence or absence of glandular hairs on the peduncles. The range of
B. atriplicifolia overlaps in central Baja California with that of the more
northerly B. juncea var. juncea, and the two have been collected in close
proximity at four different localities. No sign of morphological intergra-
dation can be inferred from herbarium material from these sites.

Brandegee (1889) reports finding connecting forms between B. juncea
and B. atriplicifolia at Comondu and San Gregorio in Baja California.
He distinguishes between the two taxa solely on the basis of ovate versus
acuminate involucral bracts and lanceolate versus triangular leaves. The
shape of the bracts will not differentiate between B. juncea as a whole
and B. atriplicifolia. However, the characters listed in the present key
easily separate the two taxa.

ACKNOWLEDGMENTS
I am grateful to the curators and staffs of the following herbaria for loans of
specimens. The number of specimens borrowed from each institution is given in
parentheses: CAS (65), GH (110), MO (67), LL (32), ND (15), NY (120), TEX
(26), UC (161), and US (124). I am indebted to B. L. Turner for guidance in the
completion of this study, and along with Roger Sanders and M. D. Whalen, for
critically reading the manuscript.

LITERATURE CITED
BENTHAM, G. 1873. Notes on the classification, history, and geographic distribution
of Compositae. J. Linn. Soc. 13:335-577.

Brake, S. F. 1945. Asteraceae described from Mexico and the southwestern United
States by Marcus E. Jones, 1908-1935. Contr. U.S. Natl. Herb. 29:117-137.
BoHLMANN, F., T. BuRKHARDT, and C. ZpERo. 1973. Naturally occurring acetylenes.

Academic Press, London.

1977 | NOTES AND NEWS 123

————, C. ZprEro, and M. Grenz. 1976. Imhaltsstoffe einiger Gattungen der Tri-
bus Helenieae und Senecioneae. Phytochemistry 15:1309-1310.

BRANDEGEE, T. S. 1889. A collection of plants from Baja California. Proc. Amer.
Acad. Arts 2:117-216.

GREENE, E. L. 1885. Studies in the botany of California and parts adjacent. Bull.
Calif. Acad. Sci. 4:179.

HorrMann, O. 1894. Compositae. Jn Engler and Prantl, Naturl. Pflanzenfam. IV.
5:1-706.

Jounston, I. M. 1924. Expedition of the California Academy of Sciences to the
Gulf of California in 1921. The botany. Proc. Calif. Acad. Sci. IV. 12:951-1218.

Krnc, R. M. and H. Rosrnson. 1970. The new synantherology. Taxon 19:6-11.

Netson, A. 1904. Contributions from the Rocky Mountain Herbarium, V. Bot. Gaz.
37:260-279.

Powe LL, A. M. 1965. Taxonomy of Tridax (Compositae). Brittonia 17:47-96.

, and B. L. Turner. 1963. Chromosome numbers in the Compositae. VII.
Additional species from the southwestern United States and Mexico. Madrono
17:128—140.

Sotpric, O. T., D. W. KyHos, A. M. Powe tt, and P. RAvEN. 1972. Chromosome
numbers in the Compositae. VIII: Heliantheae. Amer. J. Bot. 59:869-878.

Stuessy, T. F. 1977. A revised subtribal classification of the Heliantheae. Jn Har-
borne, J.. V. Heywood, and B. L. Turner, [eds.], Chemistry and systematics
of the Compositae. Academic Press, London.

Torres, A. M. 1968. Revision of Jaegeria (Compesitae-Heliantheae). Brittonia
20:52-73.

Turner, B. L., and A. M. Powe. 1977. Disposition of genera in the dismantled
tribe Helenieae (Asteraceae). In Harborne, J., V. Heywood, and B. L. Turner,
leds.], Chemistry and systematics of the Compositae. Academic Press, London.

i , and T. J. Watson, Jr. 1973. Chromosome numbers in Mexican

Asteraceae. Amer. J. Bot. 60:592-596.

NOTES AND NEWS

THE “GERMINATION FLAP” IN CERTAIN GRAMINEAE.—In a recent paper (Madrofio
23: 68-72. 1975), T. L. Rost & A. D. Simper report on “the germination lid” and
its occurrence on lemmas of grasses. The statement is made that this structure was
first observed in Setaria in 1949 by Keys (Trans. Kansas Acad. Sci. 52:474-477).
This is quite incorrect. Hitchcock & Chase as long ago as 1910 recognized this feature
and commented on it in their monograph of Panicum (Contr. U.S. Nat. Herb. 15:
XIV + 396 pp). On page 18 they state, “fertile lemma chartaceous-indurated, typi-
cally obtuse, the nerves obsolete, the margins inrolled over an inclosed palea of the
same texture, a lunate line of thinner texture at the back just above the base, the
radicle protruding through this at germination,” (italics mine). That this “germina-
tion lid” was recognized as being characteristic of all genera of Paniceae by Hitch-
cock & Chase is indicated by the statement under the Tribe Paniceae, on page 26 of
Hitchcock’s Manual (U.S. Dept. of Agric. Misc. Publ. 200. 1935): “fertile lemma
and palea indurate or at least firmer than the glumes and sterile lemma, a lunate
line of thinner texture at the back just above the base, the rootlet protruding
through this at germination.”

Rost & Simper mention also that there is some question regarding the systematic
position of Anthephora and Olyra. Although this might have been true several
years ago, it is certainly not so today. Fifteen years ago, I published a detailed
study of Anthephora (Reeder, J. R. in Trans. Amer. Microscop. Soc. 79:2i1-218.
1960) which left little doubt that this genus is a member of the Tribe Paniceae,

124 MADRONO LVol. 24

most closely related to Cenchrus. With respect to Olyra, the evidence is over-
whelming that this genus is bambusoid. In 1947, Virginia Page (Bull. Torrey Bot.
Club 74:232-239) demonstrated convincingly that the leaf anatomy of this genus
is of the bamboo type. My investigations of the embryo (Amer. Jour. Bot. 44:
756-768. 1957) indicate that the affinities of this genus are bambusoid. The embryo
in the Paniceae is very different. Most students of the Gramineae agree that the
embryo characters give perhaps the most powerful evidence regarding affinities of
genera in this group. Another indication that Olyra is bambusoid comes from cytol-
ogy. Chromosome counts of O. latifolia by Reeder et al. (Taxon 18:441,442. 1969)
and by Pohl & Davidse (Brittonia 23:293-324. 1971) gave 2n = 22; O. yucatana was
found to have this same number by Tateoka (Bull. Torrey Bot. Club 89:77-82.
1962); and 2%~= 22 was also reported in O. loretensis by Gould & Soderstrom
(Canadian Jour. Bot. 48:1633-1639. 1970). This chromosome information supports
the contention that Olyra has bambusoid affinities. Basic numbers of x = 12 or 11
are usual among bamboos and their kin, whereas x = 9 or 10 is characteristic of
the Paniceae.

Another character of considerable importance in grass taxonomy is the hilum. In
all members of the Paniceae, this structure is punctate; in Bambuseae it is linear.
Examination of “seeds” of Olyra shows a hilum which is distinctly linear, clear
evidence that this genus was wrongly placed in the Paniceae by Hitchcock & Chase.
According to R. W. Pohl (pers. comm.) another bambusoid grass, Streptochaeta,
has a distinct “germination flap” on the fertile lemma.

To suggest that Olyra belongs in the Paniceae merely because it has a “germina-
tion lid,’ and to ignore the other overwhelming contrary evidence is certainly
naive. This sort of reasoning led agrostrologists of the past to group Eragrostis with
Poa. A more satisfactory explanation for the eccurrence of a “germination lid” in
panicoid genera and in Olyra would seem to be not that this indicates close phyletic
relationship, but rather that it is a case of parallelism. If the caryopsis is tightly
enclosed within an indurated lemma and palea, there must be a means for the
rootlet of the embryo to emerge at germination if the seedling is to become estab-
lished. One would expect natural selection to favor those plants which developed a
weak spot (“germination flap”) directly above the position of the rootlet. The
enhanced germination would permit an increased number of offspring—Joun R.
ReEEDER, Department of Botany, University of Arizona, Tucson 85721.

REVIEW
Marine Algae of California. By IsABELLA A. ABBotT and GEORGE J. HOLLENBERG.
827, 701 figs. Stanford University Press, Stanford, California 94305. 1976. $22.50.

The publication of this massive contribution has been awaited eagerly by marine
phycclogists in California and elsewhere. The authors are noted for their major
contributions to the systematics and morphology of Pacific marine algae. In some
respects the bock’s format is similar to that of Smith’s Marine Algae of the Monterey
Peninsula published in 1944 which it now supplants. MAC, however, contains a
number of very important features not present in Smith.

The extensive introduction has information on classification, form and physiclogy
of marine algae, a brief geographical treatment of the California coast with simpli-
fied maps indicating major coastal landmarks and suggestions about techniques for
collecting and preparing specimens. Following the introduction is one of the book’s
highlights—a 25 page account by G. F. Papenfuss entitled Landmarks in Pacific
North American Marine Phycology, which emphasizes the early expeditions and
botanists who provided our first knowledge of the remarkable flora. on this coast.
It includes photographs of several of the more noted phycologists associated with
the flora, memorable anecdotal information about many, as well as mention of algal
taxa named in their honor.

1977 | REVIEW 125

The main text of the book is divided into four sections treating the genus Vau-
cheria (Chrysophyta), and the phyla Chlorophyta, Phaeophyta and Rhodophyta.
Keys and descriptions for the orders, families, genera and species are included.
Illustrations of each species are on the page adjacent to its description. The main
text is followed by a master dichotomous key to the genera of the three major phyla,
a glossary, ilterature cited, and a taxonomic index.

In the preface the authors explain their rationale for the book’s format and con-
tent. It is intended to serve as a manual for identification by students with minimal
training in lower plant biology. Conservation of space was essential to keep the num-
ber of pages and price to a minimum. Consequently, descriptions are “concise”,
generaliy having somewhat less information than those in Smith. The authors chose
1972 as the cutoff date for changes and additions in the manuscript. As a result of
this, new taxa, additions to the flora and nomenclature changes published since 1972
are not included. .

Perfection is an elusive quality, seldom achieved, and although the book is a
remarkable achievement it is not perfect. I would like to point out a few of its
imperfections and positive features.

In the introduction (p. 7) several questionable statements are made. First, the
authors state:

“The geographic area whose algal population is most like that of Cali-
fornia is Japan. The marine flora of Northern Honshu and Hokkaido would
seem very similar to a Californian .. . It is estimated that about 30-40
percent of the species occurring in California may also be found in Japan.
This is a larger percentage than a comparison of California with the North
Atlantic weuld realize, though temperature barriers in the Pacific are no
less formidable as obstacles than the land barriers separating California from
the Atlantic.”’

The figure of 30-40 percent species shared in common is a gross overestimate. Hom-
mersand (1972, Proc. Intern. Seaweed Symp. 7:66-71) very generously estimates
that perhaps 110 species are common to both Califcrnia and Japan. This is only
16% of the 669 species listed in MAC. A more realistic figure based on hardcore
evidence might reveal less than 10% are common between the two areas. Their
statement that temperature barriers are no less formidable obstacles than land
barriers cannot be taken seriously. Many temperate and cold water marine algae
can withstand wide ranges in temperature (5—20°C) for several days or weeks and
still remain viable. This seems a sufficient time for drift algae to be transported
short distances and allows for progressive establishment along a coast line. Moreover,
they apparently do not consider the possibility that migration of certain species may
have occurred at times when water temperature patterns differed from those of the
present, nor do they consider the possible involvement of ccntinental drift in
species distribution.
In the last paragraph on page 7 the authors state:

“Although California and Japan share many genera, there are usually
larger numbers of species in given genera on one side of the Pacific than on
the other, implying genetic if not ecological diversity. One genus, Laurencia,
has 12 eastern Pacific species and 23 western Pacific species.”

This is an unreasonable comparison because Laurencia is principally of tropical and
subtropical distribution. In these particular ocean temperature zones there is a
vastly greater coast line in the western Pacific than in the eastern Pacific. The greater
number of habitats in the western Pacific should permit a greater species diversity.
By contrast the coast line in temperate and colder water zones is closer to being

126 MADRONO [ Vol. 24

equal. Surprisingly, the authors do not consider upwelling as one of several possible
factors influencing species diversity, yet that is one of the factors considered by
zoogeographers in explaining the greater diversity of the Pacific North American
biota relative to the Atlantic North American bicta.

The integration of illustrations with species descriptions, the generally high quality
of reproduction of the figures, the illustration of every species and most subspecific
taxa included as well as the use of figures to show important diagnostic characters
are all very useful features which make this flora vastly superior to others I have
used. However, there are also a number of problems in regard to the figures. The
single most cumbersome and time-consuming aspect is that the figure numbers are
not directly next to the figure (although each figure in a composite plate is identi-
fied in the caption by a number and its position on the page, e.g., lower right, upper
center). Moreover, figure numbers are not cited with the species descriptions. For
rapid accurate reference it is essential that figure numbers be placed immediately
next to the figure and, what is especially important, that they be cited in the descrip-
ticn for each species. The captions generally fail to identify the types of reproductive
structures illustrated. For the inexperienced student, to whom this flora is especially
directed, these deficiencies could be very troublesome and frustrating.

In Fig. 23 (Lithothamnion californicum) the caption identifies the figure as a
section through a tetrasporangial conceptacle, yet the figure shows binucleate bispor-
angia. Tetrasporangia but not bisporangia have been reported for this species. For
Fig. 156 (Punctaria hesperia) the caption indicates that only the “unangia” are
illustrated but “plurangia” are also present in the figure. Fig. 310 and Fig 311 were
transposed (Abbctt, personal communication). Fig 310 is of Peysonnelia rubra var.
orientalis instead of P. profunda as indicated in the captain. Surprisingly, the
description of P. rubra var. orientalis does not refer to the zonately divided spor-
angia of this taxon, a unique character among the California species of Peyssonnelia.
This would have been useful as a keying character and should have been mentioned
in the generic descripticn as an exception to the cruciately divided sporangia.

I should note here that misspellings are very infrequent but one is Peyssonelliaceae
and Pevssonellia (sic). The accepted spelling is Peyssonneliaceae and Peyssonnelia.

Some descriptions tend to be incomplete or inaccurate with respect to information
on reproduction. For example, that of Farlow7a includes a description of tetraspor-
angia, the implication being that tetrasporophytes and gametophytes are ismorphic.
This report of tetrasporangia is based on Abbott’s description of tetrasporangia in
Leptocladia conferta which she transferred to the genus Farlowia in 1968 (J. Phycol.
4:180-98). However, tetrasporophytes are not recorded for any other species of
Farlowia. It may be possible to return F. conferta to Leptocladia on this basis, al-
though the female reproductive structures of F. conferta are more similar to those of
Farlowia than of Leptocladia. It appears that California species of the crustose genus
Cruoriopsis may be the tetrasporophyte for at least two species, F. compressa and
F. mollis (DeCew and West, unpublished observations). In contrast, the descrip-
tion for Gymnogongrus states correctly that a tetrasporangial phase is unknown in
certain species.

The red algal genus Besa was placed in the Phyllophoraceae, although it is tradi-
tionally placed in the Gigartinaceae. The authors do not explain the basis for this
transfer.

The northern limit of the geographic range for Pachydictyon coriaceum is given as
Cape Arago. This is, however, based on confusion in identification and synonymy
discussed by Dawson (1950, Wassman J. Biol. 8:267). The taxon involved is
Dictyota bringhamiae. Furthermore, the description of P. coriaceum in MAC indi-
cates that the thalli are 400-500 um thick, yet the scale shows a maximum thickness
of 300 um in the specimen illustrated.

1977 | REVIEW Ir,
The authors state on page 119 that:

“Scagel (1960) and Chihara (1960) indicated the ‘‘Collinsiella” may be a
stage in the life history of a species of Enteromorpha and/or Monostroma.”

It is correct that Scagel considered Collinsiella as a stage in the life history of
Enteromorpha, but Chihara’s work in no way suggests this. The three species
investigated by Chihara have multicellular gametophytic stages which are cushion-
shaped or open sacs whereas the sporophytic stage is a unicellular “zygocyst” resem-
biing Gomontia. Nowhere does Chihara state that Collinszella is a stage of Mono-
stroma, although it is known that Gomontia-like stages occur in the life histories
of some species of Monostroma.

In general the keys seem quite workable and are less troublesome than many of
those in Smith. However, it should be noted that the second part of the third
dichotomy of the green algae key (p. 748) is incorrect and confusing. The genera of
the Prasinophyceae, Chaetophoraceae, Ulvaceae, Ulotrichaceae and Monostromata-
ceae to which this choice leads are certainly not coenocytes.

The use of quotation marks with “Chlorochytrium” (dichotomy 30 of the master
key of green algae) implies that the authors do not accept the genus as being taxo-
nomically valid, probably because C. znclusum was shown to be the sporophytic stage
of a species of Spongomorpha by Chihara (1969, Phycologia 8:127-33) and other
workers. However, this does not invalidate the genus because the life histories of the
type species, C. lemnae (an endophyte of Lemna in freshwater), and other species,
including C. porphyrae have not been investigated yet.

The first dichotomy of the Acrochaetium key (p. 309) does not enable one to
identify A. rhizoideum because the erect filaments are considerably longer than the
endophytic system. In addition, the illustration of A. rhizoideum (Fig. 260) does not
show the diagnostic feature of the species, namely the numerous pyrenoids in each
cell. Other California species lack a pyrenoid or have a single pyrenoid per cell.

Although there are minor deficiencies such as those mentioned above, the Marine
Algae of California is exemplary in its overall quality and will serve phycclogists
well for many years. The authors and publisher should feel pleased with their
efforts—Joun A. West, Department of Botany, University of California, Berkeley
94720.

REVIEW

The Tactless Philosopher. Johann Reinhold Forster (1729-1798). By Micuaer E.
Hoare. x + 419, 13 illus. Hawthorne Press, Melbourne, Australia, 1976. $15.95.
Austr.

The author citation “Forst.”’ (or more accurately Forst. & Forst. f.) is unfa-
miliar to most California botanists. However, Dichondra is one of their 75 new
genera published in Characteres genera plantarum (1776), which included the first
descripticn of New Zealand plants, Queen Charlotte Sound being the probable
source of Dichondra repens. The want of a “full-length biography” was noted by
Michael Hoare in his sketches of the Forsters, father and son, in Dictionary of
Scientific Biography (1972). Now this thoughtful, accurate, and attractive account
of the foibles and fortunes of a fumbling Forster of two centuries ago is highly
recommended.

E. D. Merrill was Forster’s ‘most trenchant critic” of botanical matters. His
condemnation of Forster’s use of Solander’s generic names is, in Hoare’s opinion,
unjustified and could not have rested on access to the Banks and Solander specimens

128 MADRONO [Vol. 24

and manuscripts after Cook’s Second Voyage (1772-1775). “No such hint or state-
ment appears in even the most private correspondence” (p. 139), Hoare adding,
“to argue, as Merrill does, that either Forster depended upon Solander’s knowledge
and experience for their botany is absurd. Forster admired the Swede’s abilities
and intellect but scarcely felt beholden to him.”

Armchair navigators with Capt. Cook will find Hoare’s biography of Forster
high adventure. The story tells of New Caledonian “Cookpines,” actually Arau-
carias, mistaken at first by Forster for basalt columns; of Tahiti where the natives
named him Fatara; and much more. There are notes on the Americans, Mannasseh
Cutler and Samuel Vaughan, watching for locally published travel narratives that
Forster might translate; on Dr. Lettsom and Thomas Pernant in England; and on
Kurt Sprengel who was at Forster’s deathbed on December 9th, 1768, in Halle.
Today the graceful Forster’s Tern; a popular dining-hall palm, Howea forsterana;
the lowly perennial New Zealand genus Forstera; and, of all things, a short street
in the Fijian town of Suva are visible memories——JOsEPH EWAn, Department of
Biology, Tulane University, New Orleans, La. 70118.

REVIEW

A Gazetteer of the Chihuahuan Desert Region. By JAMES HENRICKSON and
RicHarp M. Straw. xxii + 272 pp., incl. 18 maps. Published by the authors. 1976.
Available from Dr. J. S. Henrickson, P.O. Box 8495, University Station, Austin,
Tex. 78712. $12.00 incl. postage; in Texas add applicable sales tax. [Note: A com-
plimentary copy of “Maps of the Chihuahuan Desert Region” (49 pp., offset, 43 by
28 cm.) compiled by José Garcia will be sent with each gazetteer. ]

This gazetteer is coordinated with and is intended to be a supplement to M. C.
Johnston’s work toward a flora of the Chihuahuan Desert Region (CDR). As
delimited for floristic treatment, CDR is the plateau between Sierra Madre Oriental
and Sierra Madre Occidental extending from ca 34° N in SE Arizona and S New
Mexico to ca 22° N in N Guanajuato and S San Luis Potosi.

The nearly 23,000 entries include place-names for natural (sierras, passes, canyons,
lakes, streams) and man-made (cities, villages, ranches, railroad stops, reservoirs)
localities. Each entry includes information as to type of locality (pass, village, lake,
etc.), state, latitude, longitude, elevation (in meters), source, and serial number.
Some peripheral, non-desert place-names are included. Sources of names include
numercus maps and published records. The computer programs used in producing
the gazetteer allow extraction of partial or alternate listings such as listings by
elevation or latitude or listings for subregions within CDR.

The gazetteer includes maps showing county or municipio boundaries for all
states in er bordering CDR. The topographic maps (from U.S.G.S. for United
States and A.M.S. for Mexico) compiled by Garcia are well reproduced and provide
a very useful and welcome supplement.

This compilation of information about CDR will be invaluable to naturalists
concerned with the area—Joun L. StrorHER, Botany—Herbarium, University of
California, Berkeley 94720.

Membership in the California Botanical Society is open to individuals ($12.00 per
year, regular; $8.00 per year, student). Members of the Society receive MaproNo
free. Institutional subscriptions to MaproNo are available ($14.00 per year).

Back issues of Madrofio are available at the following rates:

Vol. 1 (1916-1929) and Vol. 2 (1930-1934), each consisting of 17 numbers: $17.00
per volume, $1.00 per issue.

Vol. 3 (1935-1936) through Vol. 20 (1969-1970), each biennial, consisting of 8
numbers: $16.00 per volume, $2.00 per issue.

Vol. 21 (1971-1972) through Vol. 23 (1975-1976), each biennial, consisting of 8
numbers: $24.00 per volume, $3.00 per issue.

Vol. 24 (1977) et seq., one volume per year, each consisting of 4 numbers: $14.00
per volume, $3.50 per issue.

Applications for membership (including dues), orders for subscriptions, requests
for back issues, changes of address, and undelivered copies of MaproNo should be
sent to the California Botanical Society, Inc., Department of Botany, University of
California, Berkeley 94720.

INFORMATION FOR CONTRIBUTORS

Manuscripts submitted for publication in MaproNo should be sent to the Editor.
Membership in the California Botanical Society is requisite for publication in
Maprono.

Manuscripts and accompanying illustrative materials must be submitted in dupli-
cate and should follow the format used in recent issues of MApRONO. Original illus-
trations should not be submitted until paper is accepted for publication. All manu-
scripts MUST BE DOUBLE SPACED THROUGHOUT, including title, text, tables,
captions, lists, literature cited, etc. Footnotes (which should be avoided wherever
possible), captions, and tables must be typed on sheets separate from the text.
Presentation of nomenclatural matter (accepted names, synonyms, typification)
should follow the format used for Rhus integrifolia in MapRONO 22:288. 1974. All
measurements should be given in S. I. (metric) units. Where appropriate, scales
should be included on figures rather than in captions as estimates of ratio of repro-
duction such as X %4, X 1620, etc. Institutional abbreviations in specimen citations
should follow Holmgren and Keuken’s list (Index herbariorum, Part 1, The herbaria
of the world. Sixth edition. 1974. Regnum Veg. vol. 92). Abbreviations of names of
journals should be those in Botanico-Periodicum-Huntianum (Lawrence, G. H. M.
et al. 1968. Hunt Botanical Library, Pittsburgh). If the correct abbreviation cannot
be determined, the full title of the journal should be used. Titles of books should be
given in full, together with the place and date of publication, names of publisher,
and an indication of the edition, if other than the first.

Short articles such as range extensions and other brief notes are published in con-
densed form under the heading “Notes and News”. Authors of such articles should
follow the format used in recent issues of MaproNo.

Authors are allowed up to 10 pages per year without page charges; charge for
additional pages is $30.00 per page. Subject to approval by the Editors, articles may
be published ahead of schedule, as additional pages of an issue, provided the author
assumes complete costs of publication.

ADRONO

VOLUME 24, NUMBER 3 JULY 1977
Contents
TAXONOMY OF CHRYSACTINIA, HARNACKIA, AND LESCAILLEA

(CoMPposITAE: TAGETEAE), John L. Strother 129
INTRODUCTION OF Dr. REID Moran, John H. Thomas 140

NEW OR RENOVATED POLEMONIACEAE FROM. BAJA CALIFORNIA,
Mexico (Ipomopsis, LinaAntHus, NAVARRETIA), Reid Moran 141

THE FLoraL EcoLtocy oF ASCLEPIAS SOLANOANA Woops,
Steven P. Lynch 159

VARIATION IN THE HELIANTHUS EXILISs-BOLANDERI
CompLex: A REEXAMINATION, A. M. Olivieri and S. K. Jain 177

A NEw CoMBINATION IN CyMOPHORA (COMPOSITAE:

HELIANTHEAE: GALINSOGINEAE) ,Judith M. Canne 190
REVIEW
Harry D. Tuers, California Mushrooms, a Field Guide to the
Boletes (Kenneth Wells) 190
SPECIAL ANNOUNCEMENT Cover 4

A WEST AMERICAN JOURNAL OF BOTANY

UBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY

MaproNo is published quarterly by the California Botanical Society, Inc., and is
issued from the office of the Society, Herbarium, Life Sciences Building, University
of California, Berkeley. Established 1916. Second-class postage paid at Berkeley.
Return requested.

Editor — BarBARA D. WEBSTER
Department of Agronomy and Range Science
University of California, Davis 95616

Associate Editor — Gravy L. WEBSTER
Department of Botany, University of California, Davis 95616

Board of Editors
Class of:

1977—W1LL1uaM Louis CuLBerson, Duke University, Durham, North Carolina

Date M. SuitH, University of California, Santa Barbara
1978—SHERWIN CARLQUIST, Claremont Graduate School

Lestiz D. GortLis, University of California, Davis

DEnNNIs R. PARNELL, California State University, Hayward
1979—Puitre W. RunpDEL, University of California, Irvine

ISABELLE TAVARES, University of California, Berkeley
1980—JameEs R. GrirFin, University of California, Hastings Reservation

Frank A. Lane, Southern Oregon College, Ashland
1981—DanrEt J. CRAWFORD, University of Wyoming, Laramie

James HEnricKson, California State University, Los Angeles
1982—Dean W. Taytor, University of California, Davis

RICHARD VocL, California State University, Los Angeles

CALIFORNIA BOTANICAL SOCIETY, INC.

OFFICERS FOR 1977

President: Dr. WinsLow R. Briccs, Carnegie Institution of Washington,
Stanford, California 94305

First Vice President: Dr. Atva Day, Department of Botany, California Academy of
Sciences, Golden Gate Park, San Francisco, California 94118

Second Vice President: Dr. Jos Ku1jt, Department of Biological Sciences,
University of Lethbridge, Lethbridge, Alberta, Canada

Recording Secretary: Dr. CHARLES F. QUIBELL, Department of Biological Sciences,
Sonoma State College, Rohnert Park, California 94928

Corresponding Secretary: Dr. RupoLF Scomi, Department of Botany,
University of California, Berkeley, California 94720

Treasurer: Dr. G. Douctas BarsBe, California Department of Food and Agriculture,
1220 N Street, Sacramento, California 95814

The Council of the California Botanical Society consists of the officers listed above
plus the immediate Past-President, DENNIS R. PARNELL, Department of Biological
Sciences, California State University, Hayward, CA 94542; the Editors of Madrono;
and three elected Council members: L. R. HECKARD, Jepson Herbarium, Department
of Botany, University of California, Berkeley, CA 94720 (1975-1977); James R.
GriFFIN, Hastings Reservation, Star Route 80, Carmel Valley, CA 93924 (1976-
1978) ; Harry D. Tuiers, Department of Ecology and Systematic Biology, Califor-
nia State University, San Francisco, CA 94132 (1977-1979).

TAXONOMY OF CHRYSACTINIA, HARNACKIA, AND
LESCAILLEA (COMPOSITAE: TAGETEAE)

JOHN L. STROTHER
Botany—Herbarium, University of California, Berkeley 94720

Taxa reviewed here are remarkably diverse morphologically, still they
form a close, and probably long-distinct, alliance. They are distinguished
from other members of Tageteae by the following combination of charac-
teristics:

1) Involucres turbinate to hemispheric, usually ecalyculate;

2) Phyllaries free to base, narrowly ovate to linear, + carinate, persis-

tent in fruit;

3) Heads radiate (except Lescaillea) ;

4) Pappus of 20—40 bristles, free to the base; and

5) Style branches (disc florets) well-developed and stigmatic almost

to the truncate-rounded apex.

These plants have received little attention outside floras or floristic
lists. As part of monographic studies in Tageteae, I offer the following
key, descriptions, observations, and comments, including accounts of
nomenclature, typification, distribution, phenology, and reproductive bi-
ology. For poorly known taxa, all specimens seen are cited. For loans or

other courtesies, I thank members of the staffs of the following herbaria:
A, ASU, C, CAS, DS, E, F, GB, GOET, K, LD, LL, MICH, MO, NMC,
NY, P, PENN, PH, POM, RSA, S, SD, SMU, TEX, UC, UPS, US,

WIS.

_ Chrysactinia has five species, referred here to three sections. The sec-
tion with the greatest concentration of primitive character expressions
(sect. Phylloloba) includes C. pinnata and C. truncata, which are re-
_ stricted to Sierra Madre Oriental of Mexico (Fig. 1). Two relatively
advanced taxa form sect. Chrysactinia: C. acerosa, restricted to gypseous
soils on western slopes of Sierra Madre Oriental (Fig. 1), and C. mexi-
| cana, widespread in the Mexican highlands and beyond from Oaxaca to
_ New Mexico and Texas (Fig. 2). Chrvsactinia lehtoaec, which is known
from only one locality (northern Sinaloa in Sierra Madre Occidental,
| Fig. 1), is the sole member of sect. Tagetifolia.

| Harnackia and Lescaillea are monotypes endemic to Cuba on serpen-
_ tine soils (Prov. Oriente and Prov. Pinar del Rio, respectively). Lescaillea
_ seems certainly to have been derived from Harnackia (or an immediate
ancestor) by reduction of leaves and loss of ray florets. Harnackia is very
similar to Chrysactinia truncata with which it must have shared a recent

ancestor. These three taxa together with C. pinnata are very closely
' related.

Madroho, Vol. 24, No. 3, pp. 129-192. August 9, 1977.
129

130 MADRONO [Vol. 24

a C. acerosa 4 Sb oe
© C. lehtoae is meer er ee
@ C. pinnata i OU ear ee :
a C. truncata ee ZA
0) 400
ES

km

Fic. 1. Distribution of Chrysactina spp.

The Cuban plants probably should be transferred to Chrysactinia sect.
Phylloloba, but they are so little known that I am reluctant to make
transfers at this time. An alternative treatment of these four species as a
separate genus (Lescaillea) seemed appropriate until the recent discovery
of C. lehtoae, which bridges the considerable morphological gap between
sect. Chrysactinia and the ‘Phylloloba’/Cuban alliance. For the present,
a conservative, status quo treatment seems preferable.

The relatively restricted and mesic, montane or submontane distribu-
tions of C. lehtoae, C. pinnata, C. truncata, Harnackia, and Lescaillea
are interesting in view of the presumably primitive position of these taxa
within Tageteae. These may be relictual survivors of taxa delineated early |
in the history of Tageteae. The nearest allies of these plants are presently
referred to Porophyllum Guett., which has discoid heads and, for several ©
other characters, differs substantially from taxa treated here. Nicolletia
A. Gray and Leucactinia Rydb. may also belong with this group.

Key to Chrysactinia, Harnackia, and Lescaillea
a. Heads radiate; leaves linear to pinnately divided, not reduced to
SCAIESS Ae ut at he Pe. va. E ee
aa. Heads discoid; leaves reduced to short, appressed, opposite scales.
Lescaillea.

b. Erect shrubs or subshrubs; leaves pinnately divided into 3—20 lobes.
or undivided and linear to acerose (CArysactinia). . . . . . ©.
bb. Scandent, suffrutescent plants; leaves pinnately divided into (1-3) |
linear-cuneate lobes... . . . . . . . . . . ~~. Harnackia.

1977] STROTHER: CHRYSACTINIA 131
i

®
%e % %

e

ee
eae Sotcue secey esis
a Sc

Be %o%
\ j e ? @ 2 . fe

e lee e.° é
\ 2 eran, Pe ‘ eo % ”

Fic. 2. Distribution of Chrysactinia mexicana.

' c. Leaves simple, linear to acerose (C. sect. Chrysactinia). . . . a.
(ce. eaves pinnately divided... 2. . ss w «© «4 4 wae

_ d. Leaves mostly alternate, onge mostly 1-2 mm wide; phyllaries
| mostly 13) linear... ie. a Cae KICaILE:
dd. Leaves mostly opposite, acerose, ost 0. 2—0.4 mm wide; phyllaries

mostiy 8 -ovate. . . 2. & 2 s « a » ss. OC, adcerosa.

132 MADRONO [Vol. 24

e. Leaf lobes broadly cuneate to obliquely deltoid, 1—4 times longer than
wide, bearing a few orange to brownish, pellucid glands (C. sect.
aN oanr st ey ho a

ee. Leaf lobes fenees naeae 8- 42 mes feceen ‘dna fale bearing numer-
ous submarginal, preenishy pellucid glands (C. sect Tagetifolia).

C. lehtoae.

f. Leaf lobes 8-12(-—20), obliquely deltoid, acute; phyllaries mostly 8.
C. pinnata.

ff. Leaf lobes 3— 7(- 13), nena earn apiculate: phyllaries mostly
Bote ke a ee eee

CHRYSACTINIA A, Gray, Mem. Amer. Acad. Arts, ser, 2. 4:93. 1849.
[Plantae Fendlerianae]| Type: Chrysactinia mexicana A. Gray.

Evergreen, glabrous to puberulent shrubs or suffrutices, mostly 1-8 dm
high at anthesis. Leaves opposite or alternate, simple and acerose to +
linear or pinnately divided into lobes or leaflets, variously dotted with
few to numerous, marginal or submarginal pellucid glands containing
strongly scented oils. Heads solitary, terminal, peduncled to subsessile.
Peduncles slender, glabrous to puberulent, usually bracteolate. Involucres
turbinate to hemispheric, 3-8 mm high. Calyculum none. Phyllaries
mostly 8 or 13, free to the base, linear to ovate, usually carinate, persis-
tent and remaining erect or becoming reflexed at maturity, usually each -
bract bearing 1—5 pellucid oil glands. Receptacle slightly convex to hemi- |
spheric, alveolate, glabrous or erose-hispid around the sockets, rarely —
bearing a few slender, deciduous paleae (C. mexicana). Ray florets
mostly 8 or 13, pistillate, fertile; corollas yellow to orange, laminae —
mostly narrowly elliptic to linear-ovate. Disc florets 12—50(-—70), per- |
fect, fertile; corollas yellow, slender, cylindro-funnelform, glabrous to —
variously glandular-puberulent, tube much shorter than the throat, lobes
5, deltoid to lanceolate, erect to spreading or reflexed; anthers slender,
minutely sagittate, collars 3—5 times as long as wide, apical appendanges ©
ovate to lanceolate; style branches often distally papillate-hispidulous, ©
stigmatic almost to the truncate to rounded apex. Achenes slenderly to
stoutly cylindric to fusiform, blackish, striate, subglabrous to hispidulous |
with short, whitish, antrorse hairs. Pappus of 20-40 tawny, free, uniseri-
ate, subequal bristles, mostly longer than the achene.

CHRYSACTINIA A, Gray sect. PHYLLOLOBA S. F. Blake, Proc. Amer. Acad. |
Arts 51:525. 1916. Type: Chrysactinia pinnata S. Wats. |

CHRYSACTINIA PINNATA S. Wats., Proc. Amer. Acad. Arts 25:154. 1890.
Type: Nuevo Leon, “On limestone ledges of mountains near Mon- |
terey” (Saddle Mountain, near 25°40’N, 100°20’W, cf. Davis, 1936),
28 May 1889, Pringle 2524 (Holotype: US!; isotypes: BM! DS! E!
F! GH! K! MO! NY(2)! PH! UC!), |

1977] STROTHER: CHRYSACTINIA 133

Shrubs or suffrutices with erect, little branched, slender, terete stems
from a rhizomatous(?) base, 6-8 dm high; internodes 25-45 mm long.
Leaves mostly opposite, lance-elliptic in outline, 25-45 mm long, coarsely
pinnatifid into 8-12(—20) obliquely deltoid, acutely pointed, somewhat
coriaceous lobes, most lobes with a dark, pellucid gland in a sinus in the
basiscopic margin. Peduncles 30-65 mm long, bearing 3—5 subulate bract-
lets; heads held well above foliage. Involucres turbinate, 6-8 mm high.
Phyllaries mostly 8, narrowly ovate to lanceolate, basally carinate, mar-
ginally and distally scarious, each bract usually bearing a subapical
pellucid gland and 1-2 pellucid glands near the basal margins. Ray florets
mostly 8; corollas with golden yellow tube ca. 3 mm long, lamina nar-
rowly ovate, whitish above, golden orange below, 4-7 mm long, 1.4—1.7
mm wide, narrowed and minutely 3-toothed at apex, glabrous. Disc
florets 20-30; corollas orangish yellow, 4—6 mm long, tube 1.1—-1.8 mm
long, throat 2.2-3.5 mm long, marked with dark nerves below sinuses,
lobes 0.6—-0.7 mm long, lanceolate to lance-ovate, erect to spreading or
reflexed, minutely papillate, throat (distally) and lobes (abaxially) glan-
dular-puberulent; anthers ca. 2.6 mm long including collar (0.4 mm
long) and ovate, blunt apical appendage (0.3-0.4 mm long); style
branches ca. 1.8 mm long, papillate-hispidulous distally, stigmatic almost
to truncate-papillate apex. Achenes 3-4 mm long, narrowly cylindric to
fusiform, blackish, finely striate, sparsely and evenly hispidulous with
short, antrorse hairs on the striae. Pappus of ca. 40 fine, tawny, minutely
barbellulate, subequal bristles 4-5 mm long.

Distribution (Fig. 1): Lechugilla/Hechtia scrub to submontane ma-
torral and pinyon woodlands, mostly on relatively mesic, north-facing
limestone slopes, sometimes bordering streams, in Sierra Madre Oriental
of Nuevo Leon, Tamaulipas, and San Luis Potosi (ca. 28°—22°30'N) ;
600-1700 m; flowering May—Jun(—Nov).

CHRYSACTINIA TRUNCATA S. Wats., Proc. Amer. Acad. Arts 25:154. 1890.
Type: Nuevo Leon, “Summit ledges of the Sierra de la Silla” (Saddle
Mountain, near 25°40’N, 100°20’W, fide Davis, 1936), 5 Jun (labels)
or 16 Jul (see Davis, 1936) 1889, Pringle 2601 (Holotype: US!; iso-
types: BM! GH! K! MICH! MO! NY(2)! PH! UC!).

i)
}

Shrubs or suffrutices, compact, much branched, 3( +?) dm high; old
stems with thick, corky bark, young stems terete, striate, glabrous; inter-
_ nodes 10—-20(-—30) mm long. Leaves opposite or alternate, ovate to elliptic
in outline, 20-45 mm long, pinnately divided into 3-7(—13) cuneate,
_ entire or coarsely dentate, apically truncate-apiculate lobes, most lobes
, and some of the distal teeth bearing apical, subulate processes each sub-
_ tended by a dark pellucid gland. Peduncles 15—30(—65) mm long bearing
| S—5 subulate, glandless bractlets; heads usually held well above the
foliage. Involucres broadly turbinate, 4-6 mm high. Phyllaries mostly

134 MADRONO [Vol. 24

13, narrowly ovate to lanceolate, carinate, marginally and distally scari-
ous and erose-ciliolate, each bract usually bearing a conspicuous, oval
pellucid gland subapically and sometimes 1-2 pellucid glands near basal
margins. Ray florets mostly 13; corollas mostly golden yellow, tube
1.5—2.6 mm long, lamina linear, 7—8(—12).mm long, 2(-—3) mm wide,
apically shallowly 3-lobed, glabrous. Disc florets 35-50; corollas dull
yellow, 3.6—6.2 mm long, glabrous, tube 0.6—-1.7 mm long, throat 2.4-3.8
mm long, lobes 0.6—1.0 mm long, lanceolate, papillate; anthers 2.3-3.5
mm long including basal collar (0.3—0.4 mm long) and lanceolate to ovate
apical appendage (0.3—-0.6 mm long); style branches 1.0-1.3 mm long,
stigmatic almost to the conspicuously papillate-hispidulous apex. Achenes
2.8-3.9 mm long, weakly prismatic to fusiform, blackish, striate, sparsely
hirtellous with subapressed, antrorse hairs ca. 0.1 mm long on the striae.
Pappus of 25-30 coarse, tawny, barbellulate, subequal bristles 4-6 mm
long.

Distribution (Fig. 1): Chaparral or oak/pinyon woodlands, mostly on
relatively mesic north-facing slopes in limestone sierras along eastern
border of Chihuahuan Desert, Sierra Madre Oriental of Coahuila, Nuevo
Leon, Tamaulipas, and San Luis Potosi (ca. 28°—23°30’N) ;_ 1250-2550
m; flowering late May—mid Aug (—Oct).

One collection treated here as C. truncata (near 23°23’N,99°51’W,
Johnston et al., 11179, LL) is morphologically anomalous. Leaf form and
texture suggest that it may be a product of hybridization between typical
C. truncata and C. pinnata. The latter is known from a nearby locality
(near 23°21’N, 99°40’W, Johnston et al., 11162c, LL). Pollen stain-
ability for the aberrant collection is 47%. In four other collections of
C. truncata, pollen stinabilities range from 79-98%. A fifth collection
(near 23°35’30’N, 100°53’20’W, Johnston et al., 11081, LL), has
38% stainability and is morphologically typical C. truncata. Pollen stain-
abilities for five collections of C. pinnata range from 84-95%.

Chrysactinia A. Gray sect. Tagetifolia, sect. nov. Type: Chrysac-
tania lehtoae Keil.

A ceteris sectiones Chrysactiniae lobis foliorum lanceolato-linearibus

et glandulis pellucidis viridulis submarginalibus numerosis punctatis —

differt.

CHRYSACTINIA LEHTOAE Keil, Madrofo 23:374. 1976. Type: Sinaloa,

18 mi NE of Coix, near 26°50’N, 108°11’W, 1300 m, 25-26 Nov 1975, |

Nash, Landye, and Lehto L19551 (Holotype: ASU!).

Shrublets to 3 dm high; young stems terete, dark reddish brown; —

internodes 12(6—-14) mm long. Leaves opposite, lance-elliptic in outline,

25-40 mm long, pinnately divided into mostly 7—9 lance-linear lobes
12-20 mm long, 0.5—1.8 mm wide, bearing numerous, ovate, greenish |

1977] STROTHER: CHRYSACTINIA 135

pellucid glands along margins at 1-3 mm intervals. Peduncles ca. 2 cm
long, bracteolate; heads held scarcely above subtending foliage. Invol-
ucres broadly turbinate, ca. 5 mm high. Phyllaries 13, narrowly lanceo-
late to linear, somewhat carinate, very narrowly scarious-margined,
apical margins minutely erose-ciliolate, each bract bearing a prominent
pellucid gland subapically and 2—4 inconspicuous, lateral glands near the
base. Ray florets 12-13; corollas bright yellow, tube ca. 3.5 mm long,
lamina elongate-oblong, ca. 8 mm long, 2.8 mm wide, distally thickened
and minutely 3-toothed, tube (distally) and lamina (proximally) sparsely
glandular-puberulent; style brances ca. 1.8 mm long, apices papillate,
rounded. Disc florets ca. 40; corollas greenish yellow, ca. 5.5 mm long,
tube ca. 1.8 mm long, throat ca. 3.2 mm long, lobes ca. 0.6 mm long, del-
toid, erect, thickened, minutely papillate, throat and lobes glandular-
puberulent; anthers ca. 2.7 mm long including collar (0.4 mm long) and
ovate apical appendage (0.4 mm long); styles branches ca. 1.8 mm long,
stigmatic almost to papillate, rounded apex. Achenes (immature) ca. 3
mm long, slender, blackish, striate with white, antrorse hairs 0.05—0.09
mm long on the striae. Pappus of 25-30 coarse, tawny, barbellulate
bristles 4-5 mm long.

Distribution (Fig. 1): Known only from the type collection.
CHRYSACTINIA A. Gray sect. CHRYSACTINIA.

CHRYSACTINIA ACEROSA S. F. Blake, Proc. Amer. Acad. Arts 51:524.
1916. Type: San Luis Potosi, Sierra de Guascama, Minas de San
Rafael (near 23°13’N, 100°15’W, fide Sousa S., 1969), Jun 1911,
Purpus 5136 (Lectotype [here designated]: US!; isotypes: BM! E!
F! GH! MO! NY! UC!).

Compact shrublets 1(—2) dm high; young stems slender, terete, gla-
brous to minutely puberulent. Leaves mostly opposite, antrorse, acerose
(rarely with 1-2 lateral lobes), 4-12 mm long, 0.2-0.5 mm wide, some-
what succulent, pungent-tipped, glabrous to minutely hispidulous, dotted
with numerous pellucid glands. Peduncles short (2-15 mm), sparsely
glandular-puberulent, usually bearing 1-5 subulate bractlets. Involucres

broadly turbinate to campanulate, 3-4 mm high. Phyllaries 8, ovate,

Carinate, scarious-margined, erose-ciliolate distally, each bract usually
_ bearing a single, ovate pellucid gland subapically, bracts spreading but

_ not reflexed at maturity. Ray florets 7-9; corollas bright yellow, tube ca.
1.2 mm long, lamina linear, 7-8 mm long, 1.5—2.5 mm wide, often with
_ 1-3 orange pellucid glands near the minutely 3-lobed apex, glabrous:
style branches unequal. Disc florets 12-15; corollas yellow, ca. 5.5 mm
_ long, glabrous, tube 1.2-1.4 mm long, throat 3.4-3.7 mm long, lobes

|

0.6-0.8 mm long, lance-triangular, erect, 0-3 bearing orange pellucid

| glands; anthers ca. 3 mm long including collar (0.5 mm long) and lance-

136 MADRONO LVol. 24

ovate appendage (0.6 mm long); style branches ca. 1.3 mm _ long,
shaggy-papillate distally, stigmatic almost to the rounded-conical apex.
Achenes ca. 2 mm long, stoutly cylindric, blackish, striate, sparsely his-
pidulous with short (0.1 mm), antrorse hairs on the striae. Pappus of
20-30 coarse, tawny, scabrellous bristles 4-5 mm long.

Distribution (Fig. 1): Poorly known; gypseous outcrops in desert
scrub or pinyon woodlands in mountains of eastern Chihuahuan Desert
and southeast in San Luis Potosi (ca. 24°50’—22°20’N).

Specimens seen in addition to types: Nuevo Leon, W of San Roberto Junction,
near 24°36’ N, 100°38’ W, 2100 m, 19 Jun 1972 (anthesis), Chiang et al 8019, LL.
Nuevo Leon, W of Galeana, near 24°41’ N, 100°10’ W, 20 Jun 1972 (anthesis),
Chiang et al. 8038, LL. Nuevo Leon, 17 mi E of San Roberto Junction then 2 mi
S on dirt road, 24 Oct 1970 (anthesis), Turner and Crutchfield 6321, TEX. Zaca-
tecas, Concepcion del Oro, 2600-2700 m, 18-19 Jun 1934 (anthesis), Pennell 17431,
NYS:

CHRYSACTINIA MEXICANA A. Gray, Mem. Amer. Acad. Arts, ser. 2. 4:93.
1849. Type: In protologue, Gray cited, “Dry valley west of Saltillo,
April; and on high grounds near Buena Vista, May, Dr. Gregg. Also at
‘Ojo del Agua,’ near the city of Mexico? Dr. Halstead (in Herb.
Torr.).” I have seen specimens labeled as follows: ‘west of Saltillo”
at GH, MO, and NY; “near Buena Vista” at GH, MO, and NY; and
“Ojo del Agua” at GH, K, and NY. At GH, the Gregg collections are
apparently combined and are associated with a single label, which lists
both localities. A small portion of the Halstead material is also mounted
on that sheet. Although all these specimens are readily determined as
conspecific and likelihood of confusion seems small, I here designate
(in spite of questionable locality) Halstead s.n. (no date) at GH as
lectotype in order to provide a peg on which to hang the name.

Pectis taxifolia E. L. Greene, Leafl. Bot. Observ. Crit. 1:148. 1905.
Type: New Mexico, Sierra Co., Black Range, Kingston, 5 Oct 1904,
O. B. Metcalfe 1440 (Holotype: US!; isotypes: BM! CAS! E! F!
GH! MO! NMC! NY! UC!).

Strict, twiggy shrubs 2-3(—4) dm high; young stems slender, terete,
glabrous to puberulous. Leaves mostly alternate, crowded to well-spaced,
linear to narrowly oblanceolate or clavate, flattened or subterete and
somewhat succulent, 5—10(—23) mm long, mostly 1-2 mm wide, usually
apiculate, glabrous to sparsely hispidulous, margins minutely ciliolate,
conspicuously dotted with greenish pellucid glands near abaxial margins.
Peduncles 30—50(15—75) mm long, minutely hispidulous to glabrous, |
usually bearing 1-7 lance-subulate bractlets; heads usually held well |
above the foliage. Involucres turbinate to hemispheric, 3.5—5.0 mm high. |
Phyllaries 13(8-14), linear to lance-linear, often acuminate, carinate,
narrowly scarious-margined, ciliolate distally, each bract usually bearing
a single orange pellucid gland subapically. Receptacle rarely bearing a —

1977] STROTHER: CHRYSACTINIA 13H)

few linear-subulate, deciduous paleae (e.g., Kruckeberg 4746, UC). Ray
florets mostly 13(—8); corollas golden yellow (sometimes drying green-
ish), tube 1.6-2.5 mm long, lamina oblong to linear, 6.2—11.9 mm long,
1.8-3.6 mm wide, tube (distally) and lamina (proximally) sparsely glan-
dular-puberulent. Disc florets 25-40(15-70); corollas yellow, 4.6-6.9
mm long, tube 1.1—-2.6 mm long, throat 2.3-3.7 mm long, lobes 0.7—1.0
mm long, triangular-deltoid, acute or rounded, spreading, papillate,
rarely sparsely glandular-puberulent; tube (distally) and throat (prox-
imally) decidedly glandular-puberulent; anthers 1.8—3.3 mm long includ-
ing collar (0.4-0.7 mm long) and lanceolate to ovate appendage (0.4—0.7
mm long); style branches 1.3—2.3 mm long, stigmatic almost to the papil-
late-rounded apex. Achenes 3—4 mm long, slender, black, striate, hispidu-
lous with short (0.1-0.2 mm), antrorse hairs on the striae. Pappus of
30-40 coarse, tawny, barbellulate bristles 3.0-5.5 mm long.

Distribution (Fig. 2): Widespread and often common in many vegeta-
tion types, mostly on limestone, from southern New Mexico south and
east through sierras of central Mexican Highlands to northern Oaxaca
(ca. 33°30’-18°N), east on Edwards Plateau to Travis Co., Texas:
200-3100 m; flowering late Mar—early Nov.

Pollen stainabilities for 23 collections of C. mexicana from localities
throughout its range are low (O-8% ) and micrograins are present in all
preparations. I found irregular meiosis in all (five) collections studied.
Two of these were reported as 2n = ca. 45; III’s, II’s, and I’s were noted
at first metaphase and laggards were present at first anaphase in both
(Strother, 1976). Michael Powell recorded ‘“n = 20 + 2 ?” on a speci-
men at TEX (King 4487). The only other chromosomal observation for
Chrysactinia known to me is “na = 15” for C. pinnata (Powell and
Turner, 1963). I assume that x = 15 for Chrysactinia and that C. mexi-
cana is triploid.

In spite of irregular meiotic behavior and low pollen fertility, plants of
C. mexicana seem consistently to set abundant good fruit (fide appear-
ance in herbarium specimens—I have not tested germination). Collec-
tively, these observations suggest that reproduction in this taxon is
largely or wholly apomictic. Tests of this hypothesis are planned.

HaArnNaAckiA Urban, Feddes Repert. Spec. Nov. Regni Veg. 21:72-73.
1925. Type: Harnackia bisecta Urban.

A monotypic genus.

HARNACKIA BISECTA Urban, Feddes Repert. Spec. Nov. Regni Veg.
21:73. 1925. Typr: Cuba, Prov. Oriente, Sierra de Nipe, Ekman
15154 (Holotype: B(?), destroyed ?). Paratypes: Urban cited Ekman
2312, 9119, 9747, 15154, I located only 9747 (NY!, Cuba, Prov. Ori-
ente, Sierra de Nipe, in charrascales-tibisiales ad Brazos Dolores, ca.
800 m, ‘12.7.1919’’).

138 MADRONO [Vol. 24

Lescaillea nipensis Carabia, Mem. Soc. Cub. Hist. Nat., ‘Felipe Poey”’
17:16-17. 1943. Type: Cuba, Prov. Oriente, Pinal de Mayari, Sierra
del Nipe, Loma del Winch, 18 Apr 1940, Carabia 3628 (Holotype:
NY!).

Scandent suffrutices or vines; old stems with thick, corky bark; young
stems slender (ca. 1 mm diam.), striate-angular, glabrous; internodes
30-60 mm long. Leaves opposite, petioles 2-4 mm long, blades divided
into 3 narrow, cuneate lobes 3-6 mm long, each lobe + truncate and
usually bearing a lance-subulate process subtended by a conspicuous,
swollen pellucid gland. Heads solitary at ends of branches, peduncles
mostly 15-30 mm long, 0-1 bracteolate. Involucres turbinate, 3.5—4.5 mm
high. Calyculum of 0-2 lanceolate bractlets. Phyllaries 7-8, free to the
base, narrowly lanceolate to lance-ovate, weakly carinate, scarious-
margined, each bract usually bearing a conspicuous pellucid gland sub-
apically. Receptacle slightly convex, alveolate, naked. Ray florets 3-5,
pistillate, fertile; corollas yellow(?), tube slender, ca. 2 mm long, lamina
narrowly elliptic, ca. 4.5 mm long, 1.5 mm wide, entire; style branches
1.3 mm long, stigmatic almost to the rounded, papillate-hispidulous apex.
Disc florets ca. 10, perfect, fertile; corollas yellow(?), narrowly funnel-
form, ca. 5 mm long, glabrous, tube ca. 0.9 mm long, throat ca. 3.5 mm
long, lobes 5, ca. 0.6 mm long, lance-ovate, erect, minutely papillate,
glandless; anthers ca. 2.7 mm long including collar (0.3 mm long) and
lance-ovate appendage (0.3 mm long); style branches ca. 1.8 mm long,
stigmatic almost to the minutely papillate-hispidulous, conical-rounded
apex. Achenes 2-3 mm long, slender, cylindric to weakly angled, blackish,
striate, minutely hairy distally. Pappus of ca. 35 fine, uniseriate, sub-
equal, barbellulate bristles 3.5—4.5 mm long, weakly united at base.

Specimens seen in addition to types: All from Cuba, Prov. Oriente, Sierra de
Nipe, ca. 500-800 m, near 20°30’ N, 75°45’ W. Pinal Colorado, Cayo Rey, serpen-
tine barrens, 7 Jan 1956, Alain, 4927, NY. La Cueva, serpentine barrens, 27 Jul
1940, Alain 19276, NY. Same locality, 6 Apr 1941, Alain 19875, NY. Rio Naranjo,
about the falls, 3 Feb 1910, Shafer 3862, NY.

Carabia (op. cit.) cited the following as being at GH: “Oriente, Sierra de Nipe,
serpentine hills 15 km. from Woodfred, R. A. Howard no. 6092.” In spring 1975, Dr.
Howard reported (pers. comm.) “. .. we have been unable to find. . .” Howard
6092 at GH.

LESCAILLEA Griseb., Catalog. Pl. Cubens. 156-157. 1866.—Porophyllum
sect. Lescaillea (Griseb.) Gomez de la Maza, Anales Soc. Esp. Hist.
Nat. 19:277. 1890, “Porophyllon”. Type: Lescaillea equisetiformis
Griseb.

A monotypic genus.

St

1977] STROTHER: CHRYSACTINIA 139

LESCAILLEA EQUISETIFORMIS Griseb., Catalog. Pl. Cubens. 157. 1866—
Porophyvllum equisetiforme (Griseb.) Gomez de la Maza & Molinet,
Anales Soc. Esp. Hist. Nat. 19:277. 1890, ‘Porophyllon’. TyPE:
“Cuba occ.”, 1863, C. Wright 2868 (Holotype: GOET!; isotypes:
BM! G! GH! K! MO! NY! P! S! [Notes with specimen at GH indi-
cate that Wright 2868 is a pooling of at least four gatherings from
‘San Marcos” and “Bahia Honda” |.

Scandent suffrutices or vines; old stems with thick, corky bark; young
stems slender (1-2 mm diam.), striate-ribbed, glabrous; internodes
30-50 mm long. Leaves opposite, reduced to appressed, lance-ovate scales
ca. 1.5 mm long, each bearing a conspicuous pellucid gland. Heads soli-
tary at ends of branches, peduncles rather long, slender. Involucres turbi-
nate, 3.5—4.8 mm high. Calyculum of 0-3 lanceolate bractlets. Phyllaries
7-8, free to the base, linear to narrowly lance-elliptic, basally thickened,
navicular, persistent and reflexed at maturity, each bract bearing a pellu-
cid gland subapically. Receptacle flat to slightly convex, alveolate, gla-
brous. Ray florets none. Disc florets (8—)12—20; corollas greenish yellow,
tipped with purple, funnelform, ca. 4.5 mm long, glabrous, tube ca. 0.9
mm long, throat ca. 3 mm long, lobes 5, lance-ovate, erect, 0.6 mm long
minutely papillate, each bearing a pellucid gland subapically; anthers
ca. 2.2 mm long including collar (0.4 mm long) and ovate appendage
(0.3 mm long); style branches ca. 1.8 mm long, stigmatic almost to the
rounded, papillate-hirsutulous apex. Achenes ca. 4 mm long, slender,
weakly angled, dark brown, striate, sparsely hairy distally, otherwise
subglabrous. Pappus of ca. 35 reddish brown, free, uniseriate, subequal,
barbellulate bristles 3-4 mm long.

Specimens seen in addition to types: All from Cuba, Prov. Pinar del Rio, near
22°45’N, 83°30’W. Cajalbana La Palma, serpentine barren, 3 Dec 1949, Alain 1340,
NY. Loma de Cajabana, in cuabales, 10 Mar 1920, Eckman 10470, S. Same locality,
2 Jan 1921, Eckman 12718, NY, S. Pan de Cajalbana, grimpante, sur les buissons,
6 Apr 1915, Leon and Charles 4937, NY, P, S.

LITERATURE CITED

Davis, H. B. 1936. Life and work of Cyrus Guernsey Pringle Univ. Vermont, Bur-
lington.

Powe tt, A. M. and B. L. Turner. 1963. Chromosome numbers in the Compositae.
VII. Additional species from the southwestern United States and Mexico. Ma-
droho 17:128-140.

Sousa S., M. 1969. Las colecciones botanicas de C. A. Purpus en Mexico. Univ.
Calit, Publ. Bot, 51:1-36.

STROTHER, J. L. 1976. Chromosome studies in Compositae. Amer. J. Bot. 63:247—250.

140 MADRONO [Vol. 24

INTRODUCTION OF DR. REID MORAN
by Dr. Jonn H. THomas
as the featured speaker at the Annual Dinner Meeting of the
California Botanical Society, Inc., on Saturday evening, February 19, 1977
in Emeryville, California.t

It is indeed a pleasure to be allowed to introduce our speaker, Dr. Reid Venable
Moran. He is the third in a row of distinguished speakers at our annual dinner
meetings from the land to which all our water flows. Last year, as you will recall,
Dr. Harlan Lewis spoke to us, and the year before that, Dr. Sherwin Carlquist.

Reid is indeed in need . . . of introduction.

He was born, went to Stanford and received his A.B. degree in 1939. Cornell
University with an M.S. in 1942 was next. World War II saw Dr. Moran as a stout
defender of democracy in many parts of the world and in many heroic situations.
After that, he continued his formal education at the University of California, Berke-
ley, and received his Ph.D. in 1951.

His dissertation was “A Revision of Dudleya Crassulaceae’”’, and I would like to
quote from page 4 of that most celebrated work:

“T am deeply indebted to Mr. Karl Jakob, without whose kind cooperation this
thesis would have been submitted one hour and 35 minutes later.

While at Berkeley, and perhaps those were among the golden years there, he
distinguished himself in many ways, and here the historical record may need some
careful and detailed verification. But there are stories about an egg being mailed
through the Post Office to Phyllis Gardner and the sword through his (Reid’s)
head at Halloween.

Following Berkeley, Dr. Moran did a number of good things in the contem-
porary commendable “post doctoral” tradition. And in 1957 he became Curator of
Botany at the San Diego Museum of Natural History, a position which he has
filled ever since. His long-standing interest in the flora and vegetation of Mexico,
with emphasis on Baja California, began to blossom. For instance, his attention to
details resulted in significant papers on range extensions. In 1962 he published his
now classical paper on Cneoridium dumosum. The title of this paper is: “Cneoridium
dumosum (Nuttall) Hooker f. collected March 26, 1960, at an elevation of about
1450 meters on Cerro Quemazon, 15 miles south of Bahia de los Angeles, Baja Cali-
fornia, México, apparently for a southeastward range extension of some 140 miles.”
(Madrono 16:272. 1962.) The text of the paper is: “I got it there then(8068)”. The
acknowledgments go on for some 29 lines (8 pt. type), and I will quote only the
last sentence: “Last but not least, I cannot fail to mention my deep indebtedness
to my parents, without whose early cooperation this work would never have been
possible.” This cooperation would appear to have occurred in 1915!

Dr. Moran is also an expert in the families Cactaceae and Crassulaceae, and he
recently described a new genus and species in the latter family with Jorge Meyran:
“Tacilus bellus, un nuevo género y especie de Crassulaceae de Chihuahua, México.”
(Cactacea y Suculentas Mexicanas 19:75-84. 1974.) I quote further: “The genus is
named not for the Roman historian or emperor but for the peculiar form of the
corolla—from the Latin word tacitus, meaning silent. The corolla is scarcely more
silent than in most other plants; but compared with that of near relatives, it is
very close-mouthed.” (Page 82.) Is there perhaps something autobiographical in
everything one writes? In the generic description itself, one finds a compelling ex-
ample of Dr. Moran’s compassion: “Tacitus Moran: genus novum mexicanum,

1 The Editors of Madrofio, with concurrence of Dr. Thomas and the Council of
the California Botanical Society, take special pleasure in sharing with the member-
ship this part of the annual meeting of the Society.—Eds.

1977] MORAN: POLEMONIACEAE 141

Graptopetalo Rose proximum, a quo calycis segmentis reflexis, corollae ore clauso
sementisque concoloribus basi angustatis enatioaibus ornatis, fiillamentis aetate non
reflexis, stylis elongatis differt. Pax vobiscum. Herba perennis succulenta glabra... .”
(Page 76.) And so on for 13 more lines of impeccable Latin prose.

Well, I would cite many more examples of Dr. Moran’s contributions and philan-
thropies, but perhaps it is time to let him tell us about the “PLANT LIFE OF
BAJA CALIFORNIA”.

NEW OR RENOVATED POLEMONIACEAE
FROM BAJA CALIFORNIA, MEXICO
(IPOMOPSIS, LINANTHUS, NAVARRETITA)

Ret Moran
Natural History Museum, San Diego, CA 92112

In collecting in Baja California I have found three Polemoniaceae that
seem to be unnamed, three whose generic position needs changing, and a
few otherwise worth noting. The first set of my specimens is in the her-
barium of the San Diego Society of Natural History (SD); duplicates
will be distributed. In this account, my field numbers are prefixed with
“M”’. I am grateful to Drs. Alva Day and Charles H. Uhl for chromo-
some counts and to Dr. Day also for the drawings and for reviewing this
paper. Also, I thank the curators at POM, RSA, UC, and US for the loan
of specimens, and at GH for photographs.

IPpoMopsis

In his reclassification of the Polemoniaceae, Grant (1959) maintained
Ipomopsis Michx. as a genus distinct from Gilia R. & P., with the ex-
panded limits he had proposed before (Grant, 1956). Between these two
rather large and variable groups he found general differences in duration,
leaf distribution, leaf texture and dissection, flowering season, corolla
form and venation, seed size and shape, etc.; and despite some specific
exceptions, the two genera appear distinct. He also found a supporting
cytological difference: [pomopsis has a basic chromosome number of
x = 7, as in Eriastrum and Langloisia, whereas Gilia has x = 9 as in
Navarretia, Leptodactylon, and Linanthus. Thus Ipomopsis seems well

_ maintained in this expanded sense.

Asa Gray named three species of Loeselia from the Sierra Juarez of
northern Baja California: L. effusa (1876), L. tenuifolia (1876), and L.
guttata (1885). He placed them in their own section, GILtIopsis, ‘‘con-

_ hecting with Gilia”. Gray (1886) transferred all three to Gilia section
_ITpomopsis (Michx.) Benth.; but Brand (1907) and Standley (1924)
| kept them in Loeselia, and likewise Jepson (1943) and Mason (1951)
_ kept in Loeselia the one species extending into Alta [upper] California.
Grant (1959) defined Loeselia to exclude these species, placing L. effusa

i
)
|

in Gilia section Grt1astruM Brand, L. tenuifolia in Ipomopsis section

142 MADRONO [Vol. 24

PHLOGANTHEA (A. Gray) V. Grant, and L. guttata in synonymy under
Ipomopsis tenuifolia. Study of these plants suggests that Gray was right
in recognizing three species and in placing them in one genus but that
the genus now should be /pomopsis rather than Loeselia or Gilia.

These three species appear to form a close natural group within /po-
mo psis, distinct from other species of section PHLOGANTHEA but not dif-
ferent enough to be treated as a separate section. They might be called a
subsection of PHLOGANTHEA, but without first-hand information about
other members of the genus, I will not attempt a reorganization at this
level. For convenience, therefore, I refer to them informally as the Giliop-
sis group, without proposing any change in nomenclatural status. The
group may be described as follows.

Ipomopsis section PHLOGANTHEA, Giliopsis group

Loeselia section Giliopsis A. Gray, Proc. Amer. Acad. Arts 11:86. 1876.

Loeselia subgenus Giliopsis [attributed to Gray by] Peter, Nat. Pflan-
zenfam. 4(3a):54. 1891.

Annual with basal rosette and leafy stems or perennial with several
leafy stems from woody base, glandular puberulent at least above, more
or less pubescent with white multicellular hairs or glabrate in age, the
herbage with odor recalling tomato plants. Leaves horny—mucronate,
linear and entire or the lower pinnatifid with a few linear to ovate, mucro-
nate lobes. Flowers cymose, erect to horizontal, borne spring to autumn.
Calyx regular, with scarious intervals equalling or wider than herbaceous
ribs, rupturing in fruit. Corolla either red and concolorous or white to
pink and irregularly spotted, nearly regular to strongly irregular; when
corolla irregular, the anterior (lower) one or two sinuses deeper than the
others and the posterior or anterior segment respectively in plane of
symmetry; throat (measured to base of deepest sinus) much shorter to
slightly longer than tube, narrowly funnelform; segments variously
spreading, oblong to linear, cuneate, truncate and irregularly tridentate,
muricate—papillose ventrally especially near base of posterior segments.
Stamens subequally inserted at base of throat, well exserted from throat,
nearly equalling to well exceeding segments, declined with tips upcurved
when corolla irregular; anthers versatile, lobed from base nearly to inser-
tion; pollen blue. Seeds pale brown, mucilaginous when wet, oval, rounded
on back and pitted, flattish on face; or when crowded, seeds irregularly
angulate. Chromosome number: x = 7.

The type species of Loeselia section Gitiopsis is L. tenuifolia, desig-
nated by Grant (1956).

In all three species of the Giliopsis group, the flowers range from nearly
regular to strongly irregular; and when irregular, they follow either of
two variable patterns. To some extent the variation is between individual
plants, the flowers of one being generally more regular than those of an- |
other; but also, the flowers of one individual may vary markedly. The |

|
|

\

1977] MORAN: POLEMONIACEAE 143

attitude of the flowers varies from erect to horizontal, and the erect
flowers tend to be more regular. In nearly regular flowers, the lower
(anterior) sinuses of the corolla are scarcely deeper than the upper (pos-
terior); the corolla segments are about equal and diverge about equally;
and the stamens are not declined but equally spaced about the mouth of
the corolla. In irregular flowers, either two lower sinuses of the corolla are
conspicuously (to 6 mm) deeper than an uppermost one, or one lower
sinus is conspicuously deeper than the upper two; the segments may be
unequal, and they spread variously in bilateral symmetry; and the sta-
mens are declined. Also, the two lateral sinuses vary in position and
depth. When two lowest sinuses are deepest, the lowermost segment lies
in the midplane of the flower, and the other four segments diverge from
it to varying degrees, depending on the position and depth of the lateral
sinuses: in extreme flowers the other four segments are within an upper
arc of about 180°, thus forming an upper lip. When one lower sinus is
deepest and marks the midplane, the two lowest segments diverge from
it to varying degrees, again depending on the position and depth of the
lateral sinuses: in extreme flowers all five segments are within an upper
arc of 180°, thus forming an upper lip with no lower lip. Although the
symmetry varies in all three species, the modes are different, the flowers
of J. tenuifolia being more often nearly regular, those of 7. effusa and J.
guttata more often one— or two—lipped and seldom nearly regular.

The herbage of all three species has a characteristic odor, which on an
early label I compared with that Lantana camara L. but which I settled
on comparing with that of Lycopersicon esculentum Mill.

The three species of the Giliopsis group differ from other members of
I[pomopsis in their cuneate-truncate and irregularly tridentate corolla
segments. No other thoroughgoing distinction from PHLOGANTHEA is
evident. Whether the odor of the herbage is distinctive is not known since
no comparable information is available for the other species. Like Gili-
opsis, the other five species of PHLOGANTHEA also have more or less
irregular flowers; and floral symmetry is similarly variable at least in /.
havardi (A. Gray) V. Grant, with the midplane of the flower passing
through either a lower segment or a lower sinus. That species, of western
Texas, is perhaps especially close to the Giliopsis group, as suggested by
the treatment of Brand (1907).

Ipomopsis effusa (A. Gray) Moran, comb. nov.

_ Loeselia effusa A. Gray, Proc. Amer. Acad. Arts 11:86. 1876.
_ Gilia dunnii Kellogg, Pacific Rural Press 17:354. 1879. (Based on speci-

men sent by G. W. Dunn “from the southern part of [California ]”’.)
| Gilia effusa Macbride, Contr. Gray Herb. 56:57. 1918.

Annual with basal leaf rosette and 1—several slender leafy stems, 0.5—3
dm high, moderately pubescent. Cotyledons basally connate, Shion to

144 MADRONO [Vol. 24

linear—oblanceolate, acute, entire, 3-8 mm long. Rosette leaves 0.5-3 cm
long, simple or mostly pinnatifid, with 2-10 oblong to ovate lobes mostly
in upper half and sometimes crowded near apex, the rachis 0.5-1 mm
wide, the lobes ca as wide and 0.5—2 mm long; cauline leaves similar,
fewer—lobed and smaller upward, the upper entire, linear. Flowers May to
October. Calyx 3—4 mm long, the lobes 0.5—1 mm long, the scarious inter-
vals often marked with purplish red. Corolla commonly one-—lipped,
sometimes two—lipped, rarely nearly regular, pink and white, 9-14 mm
long, the tube and throat white to pinkish with small darker pink spots,
the tube 1.5—-2 mm long, ca 1 mm wide (unflattened), the throat 1-1.5
(—3) mm long, ca 1.5 mm wide, the lower sinus(es) 0.5—3.5 mm deeper
than upper, the segments deep pink, darker below, at base white with
pink spots, 4-7 mm long, 2—4.5 mm wide above, 0.7—-1.5 mm wide at
base. Filaments white, 6-9 mm long, exserted 5-8 mm from throat and
about equalling corolla segments. Style 5-10 mm long. Capsules 3-5 mm
long, 1.5—2.5 mm thick. Seeds ca 1 mm long. Chromosome number n = 7.

Type: Tantillas Mountains, Lower California, [ca 10 September |
1875, Edward Palmer 767 (GH, photo SD). This would be the Sierra
Juarez, probably: somewhere near El] Progreso. On the same sheet is a
later collection by C. R. Orcutt from nearby El Topo.

Distribution: On gravelly flats often with Pinus quadrifolia Parl. and
in mountain meadows with Pinus jeffrevt Grev. & Balf., or occasionally
in upper chaparral, at 1000-2600 m and straggling to lower elevations
along streams, northern Baja California Norte: Sierra Juarez, 1000-1600
m; Sierra San Pedro Martir, 875-2600 m.

Dr. Uhl reports a gametic chromosome number of ” = 7 probable for
a collection of /. effusa (M17887) from Yerba Buena, in the Sierra San
Pedro Martir; and Dr. Day, from many clear cells, reports 7 = 7 for a
later collection (718467) from the same place.

Grant (1959) placed this species in Glia section GILIASTRUM Brand,
but his criteria (1956: 351-352:1959:79) point to /pomopsis. In J. effusa
the stem is leafy, the leaves are simple or once—pinnate, and the segments
are horny—mucronate. Flowering is in summer, from May to September.
The flower is strongly irregular, the corolla is irregularly spotted, and its
veins do not anastomose. The seeds are oblong, not small and spheroidal,
and the chromosome number is 7 = 7. In all these respects, /. effusa is
like 7pomopsis, not like Gilia. Although it is annual, like most species of
Gilia and like no others of PHLOGANTHEA, both other section of /pomop-
sis do include annuals. Thus there seems no reason to put it in Gilza. And ©
finally, 7. effusa agrees with the other two species of Giliopsis in types of |
pubescence, odor of herbage, and patterns of floral irregularity, and, most
notably, in the cuneate—truncate and irregularly tridentate corolla seg- |
ments.

From the other two members of the Giliopsis group, /. effusa differs in |
its annual habit and basal leaf rosettes, its more consistently pinnatifid |

1977] MORAN: POLEMONIACEAE 145

leaves, with more lobes, and its smaller flowers, with much shorter
corolla tube. The corolla is often one—lipped and seldom nearly regular;
hence it is generally more strongly irregular than that of J. tenuifolia and
perhaps also than that of /. guttata.

Ipomopsis guttata (A. Gray) Moran, comb. nov.
Loeselia guttata A. Gray, Proc. Amer. Acad. Arts 20:302. 1885.
Gilia guttata A. Gray, Syn. Fl. N .Amer. Ed. 2. 2(1):411. 1886.

Perennial with several slender leafy stems from woody base, 1-4 dm
high, sparsely pubescent. Leaves mostly entire but the lower occasionally
1-3 lobed, 0.5—2.5 cm long, ca 0.5 mm wide or a little more, the lobes ca
as wide, to 3 mm long. Flowers May to October. Calyx 3—5 mm long, the
lobes 1-2 mm long, the scarious intervals purplish. Corolla commonly
two-lipped or one—lipped but sometimes nearly regular, deep pink to
white, lighter with age, irregularly spotted deep pink to purplish red,
withering bluish to purplish, 12-25 mm long, the tube 5—11 mm long,
ca 1 mm wide (unflattened), slightly wider at base, the throat 1-2 mm
long, 1.5-2 mm wide, the lower sinus(es) 0.5—6 mm deeper than the
upper, the segments 5—9 mm long, 2—3.5 mm wide above, 1—1.5 mm wide
at base. Filaments white, 6-10 mm long, exserted 5—9 mm from throat
and so about equalling corolla segments. Style 8-18 mm long. Capsules
3—5 mm long, 1.5—2.5 mm thick. Seeds ca 1.5 mm long. Chromosome
number: 2 = 7.

Type: from ‘near Hanson’s Ranch”, 18 September 1884, C. R. Orcutt
1225 (GH, photo SD). Since Orcutt (1893) reported returning from
Hanson’s by San Rafael, and since another herbarium label puts him in
San Rafael on 19 September, he could well have collected the type about
12-15 km southwest of Laguna Hanson, where the plant is known to
occur.

Distribution: Openings in chaparral, with Adenostoma fasciculatum
H. & A., and often also A. sparsifolium Torr., on the west slope of the
Sierra Juarez and the Sierra San Pedro Martir, Baja California Norte, at
800-1650 m. Apparently local, known only from two areas: 3-6 km W
to SW of El Rayo, Sierra Juarez, 1450-1650 m, M13536, 13552, 16636,

—-18483, 22738; foothills of Sierra San Pedro Martir from 1 mi S of
Rancho Santa Cruz, 1050 m (Wiggins 10031), southward at intervals
| for 15—20 mi ChnGaine 1944); divide between Arroyos Santa Cruz and
San Antonio, 1000 m, 16417, 23474; 5.5 km W of Santa Cruz, 800 m,
—M16280.

__ Dr. Day reports a gametic chromosome number of 2 = 7 for a collec-
| “tion (M18483) of I. guttata from west of El Rayo, in the Sierra Juarez.
_In some cells she found a ring or chain of four chromosomes, and in some
| the pairing appeared to be poor.

Grant (1959:137, 145) referred this species to snyonymy under J.

146 MADRONO | Vol. 24

tenutfolza, but it is clearly distinct. It is similar in size and habit, but with
leaves a little narrower and more commonly entire. The flowers are strik-
ingly different in color—white or pink and irregularly spotted rather than
bright red; they are somewhat smaller, with narrower corolla tube and
segments, much shorter throat, and much shorter stamens; and generally
they are more markedly irregular. So far as known, the areas of the two
species are distinct, but they are close enough that some overlap may
occur.

Wiggins (1944) reported this species from the foothills of the Sierra
San Pedro Martir and gave notes about the flowers.

Ipomopsis tenuifolia (A. Gray) V. Grant, Aliso 3:357. 1956.

Loeselia tenuifolia A. Gray, Proc. Amer. Acad. Arts 11:86. 1876.

Gilia tenuifolia A. Gray, Syn. Fl. N. Amer. Ed. 2. 2(1):411. 1886.

Gilia truncata A. Davidson, Bull. S. Calif. Acad. Sci. 22:72, pl. 19. 1923.
(Based on Payne & Kessler 3572, from near Jacumba, San Diego Co.,
Calif.)

Perennial with several slender leafy stems from woody base, 1-4 dm
high, sparsely to moderately pubescent. Cotyledons basally connate, line-
ar—oblanceolate, acute, horny—apiculate, ca 10-12 mm long and 1-2 mm
wide. Leaves entire or lower often 1—4 lobed, 5-35 mm long, mostly 0.5—1
mm wide, the lobes ca as wide, to 7 mm long. Flowers March to Decem-
ber. Calyx 5—9 mm long, the lobes (1—)2—3 mm long, the scarious inter-
vals reddish. Corolla nearly regular to two—lipped or sometimes more or
less one—lipped, bright red except for irregular white guide—marks at very
base of segment and continuous with white or light pink color within
tube, (12—) 16-28 mm long, the tube 5-10 mm long, 1.5—2 mm wide (un-

flattened), slightly wider at base, the throat 6-11 mm long, 2.5-3 mm —

wide, the lower sinus(es) 0.5—4 mm deeper than upper, the segments 4—9

mm long, 2-5 mm wide above, 1—2.3 mm wide at base. Filaments red |
above, white below, 14-22 mm long, exserted 8-14 mm from throat and |

so exceeding corolla segments. Style 20-30 mm long. Capsules 5—7 mm
long, 2—3.5 mm wide. Seeds ca 2 mm long. Chromosome number: ” = 7.

Type: from “northern borders of Lower California, Tantillas Moun- |

tains, especially at the entrance of the Great Canyon, W. Dunn, E. Palm-

er”, [ca 10 September] 1875 (GH, photo SD). Presumably this is near |
E] Progreso, in the Sierra Juarez. Though not mentioned on the label, G.

W. Dunn was on this trip (McVaugh, 1956).

Distribution: On open gravelly slopes and in arroyos, associated with |

pinyon—juniper woodland, chaparral, or desert scrub, from southern Cali-

fornia to north-central Baja California (Baja California Norte) at
100-2300 m elevation: Mammoth Wash, Chocolate Mts., Imperial Co., |
[ca 100 m| E. Gray (SD); SW Imperial Co. and SE San Diego Co., |
Calif., ca 450-1200 m; Sierra Juarez at 1100-1600 m on the west slope, _

1977 | MORAN: POLEMONIACEAE 147

with stragglers down to 700 m in eastside canyons; west slope of Sierra
San Pedro Martir, 800-2300 m; Cerro Matomi, 1375 m, M@20806; Cerro
San Miguel, 1125 m, M19519; Cerro San Luis, 1300 m, Moran & Hen-
rickson 10300; Ubi [ = Yubay, near 29°11’N, ca 650 m]|, Brandegee in
1889 (SD, UC).

Grant (1959) cited a chromosome number of 2” = 14 for J. tenutfolia,
based on a collection from Jacumba, California.

Grant and Grant (1965) reported that plants from Jacumba grown at
Claremont, California, were commonly visited by hummingbirds: “Their
bills slip easily into the tube as they hover and probe for nectar, and their
heads become dusted with pollen at the same time.” They concluded that
hummingbirds are the animals best fitted to feed on and pollinate these
flowers. This is the only species of section PHLOGANTHEA with flowers red
like the hummingbird—pollinated flowers of section Tpomopsis. Grant and
Grant (1965) found the Jacumba plants self-incompatible.

The corolla of J. tenuzfolia is often nearly regular, as shown by Mason
(1951: Fig. 4007), by Grant (1959: Fig. 45, from Brand), and by Grant
and Grant (1965: pl. 2F); but it may also be markedly irregular. In the
other two species of the Giliopsis group, presumably pollinated by insects,
most flowers are markedly irregular. The red hummingbird—pollinated
flowers of /pomopsis section IPpoMopsts are quite regular.

Like some other hummingbird flowers, /. tenuzfolia is sometimes called
“chuparosa” in Mexico. Also as common names, Martinez (1937) cited
“ubi” and ‘“‘agua bonita’: clearly both are from Brandegee (1889), who
gave them not as plant names but as place names.

With /. tenuifolia in section PHLOGANTHEA, V. Grant (1956, 1959)
placed the Baja California shrub commonly called /. gloriosa (Brande-
gee) A. Grant. However, Alva Day and I are studying this plant and con-
sider that it probably does not belong to /pomopsis.

Ipomopsis sonorae (Rose) A. Grant ex V. Grant, of section Micro-
GILIA (Benth.) V. Grant, also occurs in Baja California: vernally moist
depressions, sandy brush—covered flats north of the bay, San Quintin,
30 m, Raven, Mathias, & Turner 12378 (UC), det. by V. Grant; La
Bocana, east base of Sierra San Borja, 250 m, 12494 (SD), det. by
Alva Day.

LINANTHUS AND NAVARRETIA SPECIES

_ Linanthus jamauensis Moran, spec. nov. Fig. 1.

_ Planta annua hispidula supra glandulo—puberulenta 2—13 cm alta ple-
| rumque ramosa, internodiis infernis brevibus, supernis elongatis. Folia
1-9 mm longa 3-5 partita, segmentis linearibus spinuloso—apiculatis.
_Inflorescentia aperta, floribus terminalibus subsessilibusque vel axilliari-
bus longeque pseudopedicellatis. Calyx tubulo-campanulatus 3-4 mm

148 MADRONO [Vol. 24

i|
=
7 3

Fics. 1-5. Drawings by Dr. Alva Day. Fig. 1. Linanthus jamauensis, corolla,
pistil; Moran 20930 (type). Fig. 2. Linanthus uncialis, part of corolla, pistil; Raven
Mathias, & Turner 12528. Figs. 3,4. Linanthus viscainensis, corolla, seeds; Moran &
Reveal 19868 (type). Fig. 5 Navarretia fossalis, corolla, pistil; Moran 16014 (type).

longus 13 lobatus, tubo infra sinua scarioso. Corolla tubulo—infundibuli-
forma 10-17 mm longa, tubo gracillimo 6-10 mm longo intus supra basim |
puberulo—annulato, fauce 1-2 mm longa, lobis rubellis obovatis 3-6 mm »
longis. Filamenta sub sinubus inserta valde inaequilonga, brevissimis ca _
0.3 mm longissimis 1.2—2.3 mm longis. Typus: Moran 20930 (SD 83887).
Species staminibus valde inaequalibus notabilis, corollae tubo gracillimo
calyce 2—3—plo longiore in sectione Dactylophyllo praeterea distincta, L. —
rattanu et L. ambiguo fors proxima qui autem foliis 3—7—partitis, calyci- |
bus grandioribus, corollis dissimiliter coloratis, et staminibus aequalibus
valde exsertis differunt.

1977] MORAN: POLEMONIACEAE 149

Much branched or rarely simple annual herb, 2-13 cm high and to
17 cm wide, hispidulous with stiff, tapering, whitish non—glandular tri-
chomes to 0.15 mm long on stems and to 0.3 mm long or more on leaves,
and glandular—puberulent with colorless 2—3—celled trichomes ca 0.1 mm
long, each tipped with a yellowish globule to 0.05 mm thick, which
shrinks and dries reddish. Hypocotyl papillose. Cotyledons subsessile and
connate, elliptic, rounded at apex, 1.5—-3 mm long, ciliate at base, other-
wise glabrous. Stem reddish, to ca 1 mm thick at base, the main axis ca
2-6 cm high, with 3-10 nodes, in favorable years overtopped by 1-5
orders of axillary branches, hispidulous and scarcely glandular below,
often closely leafy below, the basal internodes often only 1—3 mm long;
branches mostly rebranching at each node; upper internodes more slender
and elongate, commonly 0.5—3(—4) cm long, subglabrous except often
glandular just below nodes, the ultimate to 0.2 mm thick. Leaves opposite
and connate, palmate; lower 1-9 mm long, 1-12 mm wide, moderately
hispidulous ventrally and on margins of segments, less so towards apex,
scarcely glandular, mostly 5—parted (or lowermost pairs 3—parted or
rarely simple), the segments linear, spinulose—apiculate, to 0.75 mm
wide, ca half as thick or more, the middle 1-7 mm long, the lateral
slightly shorter; upper leaves smaller, less hispidulous and more glan-
dular towards base, mostly 3—parted. Flowers pink, diurnal, terminal
and sessile or subsessile in a leaf pair or axillary on slender pseudopedicels
to 15 mm long. Calyx tubular-campanulate, (2—)3—4 mm long, ca 1 mm
wide, lobed ca one-third, moderately glandular throughout or at least in
basal half, the tube with herbaceous ribs and scarious intervals about
equally wide, the lobes equal, triangular—lanceolate, spinulose—apiculate,
herbaceous, with scarious margins only at base. Corolla tubular—funnel-
form, (7—)10-17 mm long; tube pinkish, slender, 0.3-0.5 mm _ wide
(flattened), scarcely widened upward, 6-10 mm long, glabrous or com-
monly glandular—puberulent without, with puberulent ring ca 2 mm from
base within; throat funnelform, 1-2 mm long, 1.5—-2 mm wide above,
glabrous; lobes pink, of various shades in different plants, drying blue—
violet, narrowly obovate, rounded at apex, 3-6 mm long, 2—3 mm wide.
Filaments subequally inserted in upper throat ca 0.3-0.6 mm _ below
sinuses, markedly unequal, the shortest ca 0.3 mm long, the longest
1.2-2.3 mm long; anthers 0.8-1.0 mm long; pollen yellow. Ovary ca
1 mm long; style reaching corolla mouth, the stigma lobes exserted, 1—2.5
mm long. Capsules oblong, light tan, ca 2-3.5 mm long, 1—1.5 mm thick,
with ca 2-3 seeds per cell. Seeds tan, irregular, verrucose, ca 1 mm long,
mucilaginous when wet.

Type: Abundant on gentle north slope at 1250 m, with Juniperus
californica Carr., Pinus quadrifolia Parl., and Yucca schidigera Roezl,

_ ca 1 km W of Portezuelo de Jamau, Sierra Juarez, Baja California Norte,

Mexico (near 31°37’N, 115°39’W), 19 May 1973, Moran 20930 (Holo-
type SD 83887).

150 MADRONO [Vol. 24

Distribution: Known only from the southern Sierra Juarez. Other col-
lections: gentle open north slope 3 mi NE of El Rincon, 1250 m, M21257;
Portezuelo de Jamau, 1300 m, M13886; type locality, M@21221; openings
in chaparral, mesa 1 mi S of Portezuelo de Jamau, 1450 m, M20937,
21224; flat divide 4 km NW of Cerro el Saiz, 1300 m, M23282.

Whereas plants collected in the relatively rainy year of 1973 (20930,
20937) were bushy—branching, those collected at the same places (21221,
21224) in the drier year of 1974 were mostly simple, with only 3—5 nodes
and a single flower each. In May 1976, another dry time, I found no
plants at either place.

This plant, with its short few—lobed leaves and open inflorescences,
falls in the section DACTYLOPHYLLUM (Benth.) V. Grant (Grant 1959:
108, 109). It may have one to five generations of axillary branches, each
of one elongate internode, or of very few, and ending with a leaf pair and
a terminal sessile or subsessile flower. In the ultimate branchlets the
upper leaf pair may be reduced or quite absent: with no leaves, the
branchlet is like a pedicel and would commonly be called one. However,
Grant used the term ‘“‘peduncle” for the stalk of the flower in this section.

Within DactyLoPpHyYLiuM, the new species falls in the group of species
(L. ambiguus [Rattan] Greene, L. aureus [Nutt.| Greene, L. bakers
Mason, L. bolanderi {[A. Gray| Greene, L. lemmoniu |A. Gray] Greene,
and L. rattan |A. Gray| Greene) with glabrous filaments but with a
puberulent ring within the corolla below the insertion of the stamens.
From all these species, and apparently from all other members of the
genus, it differs in its markedly unequal filaments, the longest about 4-8
times the shortest. Within the section it is remarkable also for its long
slender corolla tube, 2-3 times the calyx and about 13 mm wide: in most
other species the corolla is shorter or at least has a much shorter tube.
Within the section, only L. rattanii, of the North Coast Ranges of Cali-
fornia, and L. ambiguus, of the South Coast Ranges, have corollas that
may be as long, though commonly they also are shorter. Both species
differ further in their 3—7—parted leaves, their flowers more commonly
borne on long pseudopedicels, their generally longer calyx, their differ-
ently colored corollas, and their exserted and more nearly equal stamens.

Linanthus orcuttii (Parry & Gray) Jeps., Man. FI. Pl. Calif. 804. 1925.

Gilia orcuttu Parry & Gray, Proc. Davenport Acad. Nat. Sci. 4:40. 1884.

Linanthus pacificus Milliken, Univ. Calif. Publ. Bot. 2:53. 1904.

Linanthus orcuttii ssp. pacificus (Milliken) Masin, in Abrams, Ill. FI.
Pac. States 3:426. 1951.

The type of G. orcutta is from a “high mountain ridge in Lower Cali-
fornia, collected by C. R. Orcutt, June 1883”. From two accounts (Or-
cutt, 1883, 1893), it came from the north slope of the Guadalupe Moun-
tains, “credited with an altitude of 4000 feet [ca 1200 m]|” and with

1977 | MORAN: POLEMONIACEAE ile

Pinus coulteri near the summit, up valley from Rancho Guadalupe, 75
mi [120 km] from San Diego by road. This is the 1350—m peak now
mostly called Cerro Blanco or Sierra Blanca, 8 km SE of Guadalupe
(near 32°03’N, 116°30’W), with the only stand of P. coulteri within
50 km. Linanthus orcutti is rather scarce in the open Coulter pine wood
on the north slope, at ca 1000-1200 m (M16160, 23240). It also occurs,
rarely, in the Sierra San Pedro Martir (damp soil along streamlet, La
Concepcion, 1500 m, M15019; canyon of Rio Santo Domingo, 950 m, /.
L. Wiggins 10023A [DS]; Jeffrey pine forest, Santa Eulalia, 1850 m,
M11137) and near the mouth of Rio Santo Domingo (near Hamilton
Ranch, J. H. Thomas 104 |DS|). (Dr. Day tells me of the two DS
specimens, which I have not seen. )

Milliken (1904) described L. pacificus from Palomar Mountain, San
Diego Co., without reference to L. orcutit. It occurs also on Monument
Peak, Laguna Mountains, 1800 m (Beauchamp & Williams 2754, SD).
Jepson (1925, 1943) and Munz (1935) listed it in synonymy under L.
orcuitu. Mason (1951) made it a subspecies of ZL. orcuttu, but with no
comparison; and he was followed by Munz (1959, 1974). Grant (1959)
referred L. pacificus to L. orcutti with no comment as to subspecific
status; but while he placed L. orcutti in section DIANTHOIDES ( Endl.)
V. Grant, he named L. pfacificus type of section PActFicus (Jeps.) V.
Grant.

The specimens at hand show no basis for separating L. pacificus even
subspecifically.

Linanthus uncialis (Brandegee) Moran, comb. nov. Fig. 2.
Gilia uncialis Brandegee, Zoe 5:107. 1897.

Delicate annual 2—7 cm high, commonly simple, + lightly puberulent
with slender several—celled trichomes 0.1—-0.3 mm long, those of upper
parts gland—tipped. Lower leaves opposite for 3-6 pairs, connate at base,
simple, narrowly linear, horny—tipped, obtuse and apiculate, 3-18 mm
long, ca 0.3-0.6 mm wide, channelled ventrally especially towards base,
puberulent ventrally in lower half; upper few alternate, similar. Flowers
February to April, terminal and solitary or also with 1-2 from upper
axils on pedicels 5-10 mm long. Calyx 3-7 mm long, the tube 1-3 mm
long, with scarious intervals slightly wider than herbaceous ribs, the
sinuses V-shaped, the lobes linear, obtuse and apiculate, membranous—
margined in lower ca 1 mm, channelled and puberulent ventrally, 2-5 mm
long, in larger calyces often markedly unequal. Corolla white, 3-6 mm
long, funnelform, glabrous within, the tube and throat each ca 1 mm long,
the lobes obovate or narrowly so, + erose to irregularly rounded—toothed,
ca 2-3 mm long. Filaments inserted at base of throat, equal, glabrous, ca
0.8 mm long; anthers oval, ca 0.3 mm long after dehiscence; pollen yel-
low. Style ca 0.5-1.0 mm long; stigmas ca 0.5—0.8 mm long. Capsules
3—4 mm long.

152 MADRONO [Vol. 24

Type: Abundant on sides of gulches and in shade of bushes near sum-
mit of highest mountain [ca 1200 m], Cedros Island, Baja California
Norte, México [near 28°08’N, 115°13’W], 7 Apr 1897, T. S. Brandegee
s.n. (Holotype UC 125003).

Distribution: Near west coast of central and north-central Baja Cali-
fornia at 500-1200 m. Other collections: Baja California Norte: Aguajito
gerade 2 km E of Aguajito, 540 m, Raven, Mathias, & Turner 12528
(LA?, RSA, SD); 0.5 mi SW of southernmost pine grove, 1600 ft, Ced-
ros Island( Haines & Hale sn. (LA?, UC). Baja California Sur: upper
N slope of Cerro Azul [ca 100 km SE of type locality], 700 m, Moran
é& Reveal 19991.

Brandegee wrote that this plant was closely allied to Gilia dianthoides
Endl. [Linanthus dianthiflorus (Benth.) Greene], differing most obvi-
ously in its small corolla but not resembling the depauperate form of that
species seen about San Diego in dry seasons. Grant (1959) kept it in
Gilia, in section GIL1aAstRUM Brand, whose type is G. rigidula Benth.
However, it does seem closer to L. dianthiflorus than to any species of
Gilia, especially in view of the opposite leaves; and I place it in Linan-
thus section DIANTHOIWES (Endl.) V. Grant.

Linanthus uncialis resembles L. dianthiflorus in its leaves and calyx
and its often pedicellate flowers. It differs in its glabrous filaments and its
smaller and apparently unmarked corolla, glabrous within, with less
prominent veins and no regular denticulations. Other species placed by
Grant in this section have mostly palmately parted leaves (except L.
maculatus |Parish| Milliken), subsessile flowers, and calyx segments
hyaline—margined over most of their length; and in some the calyx is
more deeply divided. In most species the corolla is larger than in L. unci-
alis. The only other species with simple leaves and small flowers is L.
maculatus, a rare endemic of the northwestern Colorado Desert, Califor-
nia. That is also a small! plant but otherwise quite different: with compact
branching habit; coarser pubescence; shorter, broader, thicker leaves;
more flowers, on short pedicels; calyx divided to the base; corolla lobes
spreading, subtruncate, maculate.

Linanthus viscainensis Moran, spec. nov. Figs. 3, 4.

Planta annua 3-15 cm alta irregulariter multiramosa plus minusve
villosa et glanduloso—puberulenta. Folia 5-18 mm longa, inferioribus op-
positis linearibus, superioribus saepe alternis plerumque tripartitis, seg-
mentis linearibus. Flores vespertini in cymulis vulgo trifloris conferti. Ca-
lyx tubularis 4-7 mm longus plus minusve ad medium lobatus, sinubus
acutis, tubo infra sinus et segmentorum marginibus inferioribus scariosis.
Corolla calycem subaequans tubulo-infundibularis, segmentis albis 1-2
mm longis. Stamina inclusa brevia faucis basi inserta. Semina rubiginosa
subreniformia irregulariter alboangulata foveolataque hilo constricta.

1977] MORAN: POLEMONIACEAE 153

Typus: Moran & Reveal 19868 (SD 92324). Species L. arenicolae
affinis, sed ille parvior densiorque fere e basi dichotome ramosus, villosus
sed non glandulo—puberulentus, foliorum segmentis altior insertis, semini-
bus sub aqua non mucilaginosis.

Erect bushy annual 3—15 cm tall and to 15 cm wide, glandular—puberu-
lent with 2—3-celled trichomes mostly 0.05—0.1 mm long and each tipped
with a yellowish globule, also more or less pubescent with multicellular
white trichomes mostly 0.3-0.6 mm long. Stems slender, reddish, glandu-
lar, pubescent especially above, the main axis with ca 5-8 nodes, the
lower internodes 1-3 mm long, the upper to 2 cm; lower branches also
commonly with several nodes. Leaves opposite and basally subconnate
below, often alternate above or at least with some pair—members well sep-
arated, 5-18 mm long, strongly nerved, pubescent ventrally at least near
base, the lower simple, the upper simple or mostly 3—parted with lateral
segments commonly ca half the mid—segment, the leaf or segments nar-
rowly linear, 0.25—-1 mm wide, horny—apiculate. Inflorescence cymose,
the main axis and each branch ending in a flower, the main branches
overtopped by 2-4 generations of axillary branchlets, each commonly of
one internode, the lower elongate, to 4 cm, the ultimate very short, thus
forming cymules of mostly 3 crowded subsessile or short—pedicellate
flowers. Calyx tubular-campanulate, 4-7 mm long, 1-1.5 mm wide,
sparsely glandular, pubescent at base and on pedicels, also in mouth and
ventrally on lower half of lobes, the tube 2-4 mm long, with hyaline inter-
vals equalling or slightly exceeding herbaceous ribs in width, the lobes
erect or slightly outcurved, often unequal, subulate, hyaline—margined
below, apiculate, 1-4 mm long, the sinuses sharply V-shaped. Corolla
vespertine, tubular—funnelform, 4—6.5 mm long, the tube white, ca 2-3
mm long, ca 0.5 mm wide, the throat poorly delimited and scarcely wider,
yellow, 1—1.5 mm long, the segments white (or light yellow?), narrowly
obovate, subacute, 1-2 mm long, 0.5-1 mm wide. Stamens included, the
filaments glabrous, 0.3-0.5 mm long, inserted somewhat unequally at
base of throat, the anthers ca 0.25 mm long. Style 0.5—1 mm long; stigma
lobes 0.5—1 mm long. Capsules oblong, light tan, 2-4 mm long, 1.5—2 mm
thick, with ca 15-20 seeds per cell. Seeds red—brown, 0.4—0.6 mm long,
subreniform, irregularly angled and pitted, notched and white at the
hilum, the testa closely adherent but projecting as narrow whitish wings
on some angles, mucilaginous when wet.

Type: Few and scattered in sandy bed of Arroyo Malarrimo 18 km
S of mouth, at 75 m, Baja California Sur, México (near 27°39’N,
114°290’W), 6 Feb 1973, Moran and James L. Reveal 19868 (Holotype
SD 92324),

Distribution: Known only from sandy arroyo beds east and southeast
of Bahia Tortuga, on the Viscaino Peninsula, northwest part of Baja
California Sur. Other collections: scarce in bed of Arroyo Largo 6.5 km

154 MADRONO [ Vol. 24

E of mouth, 110 m, Moran & Reveal 19931; scarce in arroyo bed 13 km
by road NW of Asuncion, 70 m, Moran & Reveal 19786.

At all three localities L. viscainensis was scarce, widely scattered along
the sandy arroyo beds. Plants collected in mid—morning and early after-
noon had no open flowers; only the late afternoon collection had any
flowers open. These flowers had a white tube and segments and yellow
throat; but closed ones at another locality looked as if the segments
might have been light yellow.

The new species appears most closely related to L. arenicola (M. E.
Jones) Jepson & Bailey, reported as a rare gypsophile occurring from the
eastern Mojave Desert of California to southern Utah. That plant also
has simple or 3—parted leaves, and it is similar in the size and form of the
calyx and of the corolla, which again is vespertine. However, it is a
smaller and more compact plant, branching dichotomously from near
the base, the main axis having only one or two nodes, and flowering from
even the lowest dichotomies. In L. viscainensis the main axis and its
lower branches each have several nodes; and the upper branching is not
regularly dichotomous because the leaves are not regularly opposite.
Linanthus arenicola is similarly pubescent or more so, but it is not also
gelandular—puberulent; the lateral lobes of the leaves are inserted higher
above the base; and the seeds are slightly larger and not mucilaginous
when wet.

Mason (1938) considered L. arenicola (as L. mohavensis Mason)
most closely related to LZ. jonesti (A. Gray) Greene, which it resembles
in its vespertine flowers and its subreniform seeds, notched at the hilum.
Likewise, Jepson (1943) placed L. arenicola in his subgenus EULINAN-
THUS, with L. dichotomus Benth. (generitype), L. bigelovi (A. Gray)
Greene, and L. jonesii—all with vespertine flowers. Grant (1959) kept
only L. dichotomus, L. bigelovii, L. jonesu, and L. concinnus Milliken in
section LINANTHUS, characterized in part by flowers vespertine, calyx
sinuses broad and more or less truncate, and filaments inserted on the
corolla tube, i.e., below the throat. Linanthus arenicola, with sinuses
V-shaped and filaments inserted low in the throat, he placed in section
DIANTHOIWES (Endl.) V. Grant.

The vespertine flowers of L. arenicola and L. viscainensis do not alone
necessarily show close relationship with the section LINANTHUS. How-
ever, the notched seeds are remarkably similar to those of L. jonesii and
are different from those of section DIANTHOIDES. Furthermore, the calyx
sinuses, though V-shaped, are not so deep as in most members of DIAN-
THOIDES. And other members of that section have broader corollas, mostly
campanulate to funnelform (with a comparatively long tube and throat
only in L. killipit Mason and L. orcuttii [Parry & Gray| Jepson) and
mostly spotted at the throat. Thus there is reason to think that L. arent-
cola and L. viscainensis may belong in the section LINANTHUS.

Jt

1977] MORAN: POLEMONIACEAE 15:

Navarretia fossalis Moran, spec. nov. Fig 5.

Herba annua erecta prostratave 1-15 cm alta et lata plus minusve
villosa et glanduloso—puberulenta, capitulo florali primario sessili caules-
centive, vulgo ramos patulos in capitula similaria desinentes infra emit-
tenti. Folia 1-5 cm longa remote pinnatifida infimisve integris. Bracteae
5-15 mm longae pinnatifidae, lobis acerosis. Calyx 4-8 mm longus, lobis
inaequalibus integris, tubi ore arte ciliato. Corolla alba 4.5-6.5 mm
longa, lobis uninervatis. Filamenta proxime infra sinua corollae inserta,
0.3-1.0 mm longa. Stigmata duo 0.2—0.3 mm longa. Capsula bilocularis
diaphana indehiscens. Semina 5—25 rubiginosa 0.7—1.1 mm longa foveo-
lata sub aqua mucilaginosa. Typus: Moran 16014 (SD 70313). Ab aliis
speciebus quarum stamina in sinibus corollae pariter inserta sunt calyce
longiore (in illis plerumque 4-5 mm) fiilamentisque brevioribus (in illis
plerumque 1.5—3 mm) differt. A V. bakeri, N.. pauciflora, et N. pleiantha
capsulis bilocularibus pluriseminalibus, a N. prostrata foliis basalibus
parvioribus, calycis lobis simplicibus, corollaque parviore alba praeterea
differt.

Annual 1-15 cm high and wide, the primary head sessile or caulescent,
solitary or with 1—several spreading branches below, each bare below and
ending in a similar head or also with several lateral heads; some parts
pubescent with somewhat crinkly white pluricellular trichomes ca 0.5—1.0
mm long and glandular—puberulent with 2—3—celled trichomes ca 0.05
mm long, each tipped with a yellowish globule to ca 0.05 mm thick. Coty-
ledons connate at base, filiform, ca 5-13 mm long, bluntly horny—tipped.
Stems whitish, retrorse puberulent or glabrous below. Leaves and folia-
ceous bracts soft-herbaceous when fresh, the bracts and upper leaves
drying stiff and spinose. Lower leaves opposite, glabrous, 1-5 cm long,
less than 0.5 mm wide, filiform and entire or remotely pinnatifid with
linear or bifurcate lobes, the apices spinulose—apiculate. Middle leaves
alternate, similar, remotely pinnatifid. Foliaceous bracts 5-15 or more
mm long, + pubescent in lower half, the rachis oblong, 0.5-2 mm wide,
several nerved, scarious margined and ciliate, with 1-7 pairs of simple or
basally branching acerose lobes mostly below middle, the terminal lobe
linear, acerose. Heads 1-2 cm wide, each a compact compound cyme of
15—50 subsessile flowers, flowering in May and June. Calyx 4-8 mm long,
the tube 2-3 mm long, scarious between ribs, = pubescent and glandu-
lar, closely ciliate at mouth, the sinuses truncate, the lobes glabrous,
linear, acerose, unequal by — 2 mm, commonly 2 exceeding and 3 shorter
than corolla, drying purplish. Corolla white, 4.5—6.5 mm long, in age cir-
cumscissile at base and pushed up by developing ovary, the tube 3-4 mm
long, 0.2-0.4 mm wide, widened at base, the throat ca 0.5—1.0 mm long,
the limb 3-4 mm wide, the lobes oblong, rounded at apex, 1-2 mm long,
0.4-0.7 mm wide, the single nerve sometimes with 1-2 ascending weak
branches. Filaments inserted just below sinuses, 0.3—1.0 mm long; an-

156 MADRONO [Vol. 24

thers oblong, sagittate, 0.5-0.9 mm long before dehiscence and 0.3-0.6
after. Ovary ovoid, green, ca 1 mm long, 2—celled; style slender, 2.5—4
mm long; stigma lobes 2, 0.2—-0.3 mm long. Capsules 2-3 mm long, very
narrowly attached, 2—celled, with diaphanous walls, indehiscent. Seeds
5-25, oval to irregular, 0.7-1.1 mm long, red—brown, finely pitted, muci-
laginous when wet. Chromosomes: ” = 9.

Type: Common in dry adobe soil, bank of ranch pond, Rancho Mesa
el Tigre, 14.5 km SE of La Mision, Baja California Norte, Mexico, 375
m (near 32°00’N, 116°44’30”W), 31 May 1969, Moran 16014 (Holo-
type SD 70313).

Distribution: Locally common in a few vernal pools and roadside de-
pressions below 450 m, western Riverside and southwest San Diego Coun-
ties, California, and northwest Baja California. Other collection: River-
side Co.: 1 mi E of Perris, R. Hoover 11152 (UC). San Diego Co.: 1 mi
N of San Marcos, F. Gander 3809 (SD); National Ranch, D.. Cleveland
in 1882 (SD, UC), Orcutt sn. (UC); Loma Alta, Otay Mesa, F. Gander
217 (SD); Siempre Viva Rd., Otay Mesa, R. M. Beauchamp 405 (SD);
La Media Rd., Otay Mesa, 150 m, Moran & Witham 16041, M23576.
Baja California Norte: Tijuana Airport, 150 m, M16054, 16105; mesa
5 km WNW of Ejido Matamoros, 150 m, M@17535; 2.2 km SW of Re-
dondo Sta., 220 m, M17835; mesa near canyon rim SE of La Mision,
250 m, M15808a, 16004; 18 km SE of La Mision, 325 m, 14993,
14999; 2 km NW of Ejido Ruben Jaramillo, 30 m, M23503; 2 km N of
Rancho Ibarra, 50 m, M@22088,; 3.5 km E of old San Quintin, 10 m,
IM 23515.

Dr. Day reports a gametic chromosome number of 2 = 9 for a collec-
tion (M23503) of N. fossalis from near Ruben Jaramillo. This count
agrees with earlier ones for Navarretia (e.g. Grant, 1959).

In view of the truncate and closely ciliate calyx sinuses, the one-veined
corolla lobes, the indehiscent capsules, the minutely two-—lobed stigma,
and the vernal poo! habitat, NV. fossalis clearly belongs in section NAVAR-
RETIA (= section FRAGILES Crampton) (cf. Crampton, 1954; Grant,
1959). Here it falls in the group of species (NV. bakeri Mason, N. pauct-
flora Mason, N. plezantha Mason, N. prostrata | A. Gray| Greene) whose
filaments are inserted in or just below the sinuses of the corolla. Navar-
retia bakeri, N. pauciflora, and N. pleiantha all occur in northern Cali-
fornia, in Lake County and NV. dbakeri also more widely. Like these three,
N. fossalis has entire calyx segments; from them it differs in its two-—
celled capsules with more numerous seeds (5—25 vs. 1-5), its longer calyx
(mostly 5-7 vs. 4-5 mm), and its shorter filaments (0.3—-1.0 vs. 1-3 mm).

Navarretia fossalis kas been confused with NV. prostrata, which extends
from Merced and Monterey to Los Angeles and Riverside Counties, Cali-
fornia. Thus Jepson (1943) referred a San Diego County collection to NV.
prostrata, though he called it ‘‘a partial departure from the usual form”
in its elongate main axis; and Mason (1957) also gave the range of NV.

1977] MORAN: POLEMONIACEAE 157

prostrata as south to San Diego County. Navarretia prostrata agrees
with .V. fossalis and differs from the three northern species in having
two—celled capsules with more numerous seeds. Typically, NV. prostrata
has a distinctive aspect, with its main floral head sessile in a broad basal
rosette, the leaves 3—8 or even 13 cm long, with a rachis often 1 mm and
sometimes 3 mm wide, the long-stemmed lateral heads, when present, also
subtended by conspicuous leaves; and few collections include caulescent
plants. In .V. fossalis the main floral head is more commonly caulescent,
though in some populations often shortly so and in a few commonly ses-
sile; and the leaves are mostly less than 3 cm long and 0.5 mm wide. In
N. prostrata the calyx tends to be a little smaller (mostly 4-6 vs. 5-7
mm); and its larger segments often are tridentate, whereas in NV. fossalis
all are entire. Also, in NV. prostrata the corolla is longer (6 or mostly 7—9
vs. 4.5-6.5 mm) and commonly blue or lavender rather than always
white; the filaments are longer (1 or mostly 1.5—3 vs. 0.3-1.0 mm); and
the anthers are longer (to 1.2 mm before dehiscence and 0.5—0.8 mm after
vs. to 0.9 mm before and 0.3—0.6 mm after). Because of the longer corolla
and shorter calyx, the corolla of N. prostrata typically is well exserted,
whereas that of NV. fossalis is included.

Jepson (1943) commented on the variation in vegetative characters in
Navarretia and also on the variation in the calyx in some species and the
similarity of calyx lobes to bractlets and leaves. In NV. fossalis, so far as
seen, the calyx segments are always entire. In V. prostrata the one to four
largest segments often each have one or mostly two lateral teeth near the
apex. Sometimes most flowers of one plant have tridentate calyx seg-
ments, sometimes relatively few, and sometimes none. Apparently, how-
ever, most if not all populations include plants with tridentate segments:
for every herbarium sheet examined had such plants.

So far as noted on labels, the corolla color of NV. prostrata is blue or
violet, or blue or violet to white; but there is not enough information to
surmise that all populations include blue or violet flowers.

So far as known from collections, NV. fossalis occurred in relatively few
of the many vernal pools formerly in San Diego County, and in just three
areas. From two of these areas the vernal pools and the plant probably
are gone, and on Otay Mesa it is now known only in a few of the few
remaining pools; nor has it been found in artificial depressions. Possibly
it survives also near Perris, Riverside County, but in any case it must be
counted in Alta California a rare and endangered species. In Baja Cali-
fornia, however, its future seems more hopeful. Although I found it there
in only three areas of natural vernal-pools, and some of these pools are
since destroyed, it is well established in several rather widely scattered
artificial depressions. Thus as the natural vernal pools disappear, it is
becoming more a plant of roadside ditches; and I name is accordingly.

My three recent southern collections (22088, 23503, 23515), from 120

_ km or more below any of the northern, are all from ditches made during

158 MADRONO LVol. 24

construction of the highway completed about 1972—though the roadbed
was started and these ditches perhaps dug several years before. The dis-
tribution of vernal—pool plants in these relatively new roadside ditches is
still irregular and presumably unstabilized. An intriguing question is the
source of these new colonists—whether from far to the north or from
some unknown local population. With JV. fossalis at one southern locality
I was surprised to find Orcuttia californica Vasey, a very rare grass (cf.
Moran, 1969) apparently known before in Baja Caifornia only from the
type collection from San Quintin in 1886 and from my recent collections
at Tijuana Airport. Having found neither the grass nor any natural ver-
nal pools about San Quintin, I had wondered whether it persisted in that
area. The recent collection suggests that it does and thus further suggests
some refuge in the area where the Navarretia may also have persisted.
Thus it seems reasonable to suppose that NV. fossalis may be native as
far south as San Quintin, even if it has perhaps not been found there by
earlier collectors.

LITERATURE CITED

Branp, A. 1907. Polemoniaceae. Zn A. Engler, Das Pflanzenreich. Ser. 4, vol. 250
(Heft 27).
BRANDEGEE, T. S. 1889. A collection of plants from Baja California, 1889. Proc.
Calit.<Acad. Sci; Sere2, 2: 117-232.
Crampton, B. 1954. Morphological and ecological considerations in the classification
of Navarretia (Polemoniaceae). Madronio 12:225-238.
Grant, K. A. and V. Grant. 1965. Flower pollination in the Phlox family. Colum-
bia Univ. Press.
GRANT, V. 1956. A synopsis of Ipomopsis. Aliso 3:351-362.
. 1959. Natural history of the Phlox family: 1, Systematic botany. The
Hague, Martinus Nijhoff.
Gray, A. 1876. Miscellaneous botanical contributions. Proc. Amer. Acad. Arts 11:
71-104.
. 1885. Contributions to the botany of North America. Proc. Amer. Acad.
Arts 20:257-310.
—. 1886. Polemoniaceae. Synoptical flora of North America. Ed. 2. 2(1):
406-412. Smithsonian Inst.
Jepson, W. L. 1925. Polemoniaceae. Jn A manual of the flowering plants of Cali-
fornia. 781-809. Univ. Calif.
. 1943. Polemoniaceae. In A flora of California. 3:131-222. Univ. Calif.
Martinez, M. 1937. Catalogo de nombres vulgares y cientificos de plantas Mexi-
canas. México, Ediciones Botas.
Mason, H. L. 1938. Two new species of Linanthus from western North America.
Madronfio 4:157-162.
————., 1951. Polemoniaceae. In L. R. Abrams. Illustrated flora of the Pacific
states 3:396-474.
. 1957. Polemoniaceae. Jn A flora of the marshes of California. 658-666.
Univ. Calif. Press.
McVaucu, R. 1956. Edward Palmer: plant explorer of the American West. Univ.
Oklahoma Press.
Mitriken, J. 1904. A review of Californian Polemoniaceae. Univ. Calif. Publ.
Bot. 2:1-71.

1977] LYNCH: ASCLEPIAS 159

Moran, R. 1969. Orcutt grass is where it is (but why?). Environment Southw.
415:3-5.

Mouwnz, P. A. 1935. Polemoniaceae. /n A manual of Southern California botany.
388-403. Claremont Colleges.

—-—. 1959. Polemoniaceae. In A California flora. 468-515. Univ. Calif. Press.

_ 1974. Polemoniaceae. Zn A flora of Southern California. 637-669. Univ.
Calif. Press.

Orcutt, C. R. 1883. Among the palms and pines. Bull. Torrey Bot. Club 10:81-82.

————. 1893. Heman Chandler Orcutt. W. Amer. Sci. 8:31-35.

STANDLEY, P. C. 1924. Polemoniaceae. Jn Trees and shrubs of Mexico. Contr. U.S.
Natl. Herb. 23:1208-1213.

Wiccins, I. L. 1944. Notes on the plants of northern Baja California. Contr. Dudley
Herb. 3:289-305.

THE FLORAL ECOLOGY OF
ASCLEPIAS SOLANOANA WOODS.

STEVEN P. LYNCH
Department of Botany, University of California, Davis 95616

Asclepiads possess a mechanism for pollen transfer which requires
insects to remove discrete packets of pollen (termed pollinia) and later
to insert these into receptive stigmatic chambers on other flowers. Several
workers (Robertson, 1929; Grant, 1949; Woodson, 1954; Macior, 1965;
Stebbins, 1970) have speculated that mechanical and ethological exclu-
sion devices may be significant in the present reproductive isolation of
certain Asclepias species. Stebbins (1974) further suggests that these
exclusion devices, coupled with oligotrophy, may have been instrumental
in the diversification of taxa throughout the Asclepiadaceae.

This paper is part of a study in which the floral ecology of a number
of Californian and Mexican Asclepias species is being examined to deter-
mine pollination mechanisms and to assess the roles of insect behavior,
floral morphology, and habitat specificity in reproductive success. The
emphasis of this paper is to elucidate the life history and floral ecology
of Asclepias solanoana as it occurs in the North Coast Ranges of Cali-
fornia. Particular attention is given to the efficiency of the renroductive
process with regard to flower, fruit, and seed production. Behavioral
patterns and morphological adaptations both of the plant host and its
insect vectors are analyzed in detail and correlated to the efficiency of
the actual pollination process.

Geography and Site Description: Asclepias solanoana is endemic
to northern California, occurring within the North Coast Ranges from
Trinity County near Peanut in the north to Napa County near the Lake
County boundary in the south (Fig. 1). It is known only from a few small

160 MADRONO [Vol. 24

122°
l
° 124°
Aigo) UGS apo Nees eerie aot,

aS
Cc
/
\
C
c
‘
S
\-3 3

ACs
40°

38° a
; | DN
© at is ‘y a
g aoa es i SS vie? pa ee
- _— | \ 7
! xe
We Poa \
\ 4 a.
Wee a ane ant “7 ‘
0 10 20 50 100 Miles
ooo
0 25 50 100 150 Kilometers

Fic. 1. Geographical distribution of Asclepias solanoana in northern Cali-
fornia. Numbered dots correspond to study sites at Stanton Creek, Lake County
(site 1), and at Tedoc Mountain, Tehama County (site 2).

local populations on ultrabasic soils formed from extrusive parent rock
materials (serpentine). Each observed population is quite small, usually
consisting of fewer than fifty plants confined to a single slope or clear-

1977] LYNCH: ASCLEPIAS 161

ing on gently to steeply sloping hillsides with southerly exposures. Only
two of the populations of Asclepias solanoana provided a sufficient num-
ber of plants for study. They are described as follows:

Site 1 (Fig. 2) is in Lake County along Stanton Creek just west of the
Colusa County line near the northern edge of the new Indian Valley Res-
ervoir at approximately 550 m. The population consists of about 45
plants on a south facing serpentine rock outcrop with a slope of 60-90%.
The stony clay loam soil, developed from a mesozoic ultrabasic formation,
belongs to the Henneke series of serpentine soils.

Vegetation surrounding the site is a foothill woodland—chaparral tran-
sition. Within the site, an area approximately 15 x 30 m, the only other
perennials are two small leather oaks (Quercus durata); herbaceous spe-
cies rarely occur within the confines of the site. Lack of competing vege-
tation in the proximity of the A. solanoana population is probably related
to the chemical nature of the soil (Walker, 1964; Whittaker, 1954;
Kruckeberg, 1954, 1969) as well as to slope and exposure.

Site 2 is on the southwestern slope of Tedoc Mountain in Tehama
County at 1400 m. The study area, in a clearing in open jeffrey pine
forest, encompasses the most extensive known A. solanoana population.
The population consists of over 150 individuals covering several hundred
meters square on a steep south facing slope (60-90% ). The stony loam
soil belongs to the Dubakella series of serpentine.

Life History and Morphology: Asclepias solanoana is a small,
prostrate herbaceous perennial having one to several thin stems up to
2.5 dm long (Fig. 4). Each stem has one to several opposite pairs of firm,
oval to ovate leaves 3—4 cm long and 2—3 cm broad. Flowers are arranged
in terminal or subterminal umbelliform cymes with an average of 40
flowers per inflorescence.

All flowers in an inflorescence mature and open at approximately the
same time, the entire inflorescence producing, in effect, an erect hemi-
spherical blossom 3—4 cm in diameter. Individual flowers (Fig. 8) are
small; the purplish—rose, 5—lobed and strongly reflexed corolla is approxi-
mately 6 mm in length and nearly obscures the small, 5 parted calyx. The
five highly modified stamens are antisepalous, and are adnate to the en-
larged stigmatic head to form the gynostegium, a structure peculiar to all
members of the Asclepiadaceae. The filaments are fused into a tube, the
staminal column, and are adnate to the base of the corolla. The hoods,
collectively termed the corona, are actually appendages of the filaments.
They are cream to pink, saccate, and sessile, originating from the sta-
minal column. Each hood is bifid dorsally, providing a protected entrance
to the nectar stored in its hollow interior. The five bilocular brownish
anthers are expanded laterally, the resultant “anther head” covering the
enlarged stigma. Anther margins are bordered by leathery wings arranged
parallel with the wings of adjacent anthers to form five slits around the

MADRONO

1977] LYNCH: ASCLEPIAS 163

periphery of the anther head. Each slit is situated over a depression in
the stigmatic head; the resultant ‘“‘stigmatic chambers” contain the five
receptive stigmatic surfaces or grooves of the gynoecium. At the apex of
each slit is a dark, notched body, the corpusculum, which is attached
laterally to two thin, somewhat hyaline appendages, the translator arms.
Each of these is attached through the apical pore of the adjacent anther
locule to a discrete pollinium. The corpusculum, translator arms, and the
two attached pollinia form a pollinarium.'

When an insect gathering nectar inadvertently introduces an appen-
dage, hair, or spine into a slit, this body part may become attached to
the notched corpusculum. In pulling free, the insect carries away the
pollinarium which remains attached to its body (Figs. 5-7). The pollina-
tion process is completed when the insect, again inadvertently, inserts
one, or rarely both, pollinia of the attached pollinarium into the basal
region of a slit on a flower. As the insect pulls free, the pollinium becomes
wedged into the stigmatic chamber and breaks away from the translator
arm. After a pollinarium has been extracted, insertion of the pollinia into
a slit is facilitated by twisting of the translator arms, which places the
flattened planes of the pollinia in parallel (Fig. 8J illustrates the position
of the pollinia on the pollinarium after twisting, which usually occurs one
to several minutes after extraction).

The gynoecium consists of two free, superior ovaries and two separate
styles united only by the common stigmatic head. Each carpel contains
numerous multi—seriate, anatropous ovules on a ventral placenta. While
each carpel, once fertilized, is independently capable of maturing, usu-
ally only a singie follicle is produced per flower. Follicles, when mature,
lie prostrate on deflexed pedicels (Fig. 3). They are narrowly fusiform,
reaching 10 cm in length, and producing an average of 5O flattened,
comose seeds (Fig. 8). Mature seeds are dark brown, broadly oval, and
approximately 6-7 mm long; the white, loosely attached coma is about
18-20 mm long.

1 The term pollinarium is also commonly used to describe the pcllinium apparatus
in the Orchidaceae (Faegri and Van de Piji, 1971; Proctor and Yeo, 1972). An
orchid pcliinarium, however, consisting of viscidium, caudicle, and pollinium is
formed frcm the ccntents of a single anther; the two pollinia cf the asclepiad
pollinarium arc prcduced by adjacen* anthers.

Fic. 2-7. Asclepias solanoana. 2. Site of the study population at Stanton
Creek, Lake Ccunty. Mcst plants occur in the clearing shown in the center of the
photograph. 3. Mature follicles. 4. Habit of individual plant. 5. Apis mellifera
worker searching for nectar on an inflorescence. 6. Xylocopa californica male on an
inflorescence; note the position of the abdomen as the insect crawls over the
flowers. 7. Bombus vosnesenskii foreleg with attached pollinaria.

164 MADRONO LVol. 24

Fic. 8. Asclepias solanoana flower, pollinarium, and seed. A = corolla; B =
gynostegium; C =staminal column; D=hood; E=anther; F = anther wings;
G = corpusculum; H =translator arms; I= pollinitum; J = pollinarium showing
the orientation of pollinia after twisting of the translator arms has occurred; K =
mature seeds; L = seed coma.

A. solanoana perennates through establishment of a subterranean cau-
dex or “rootstock” derived from the hypocotyl (Woodson, 1954). Growth
of established plants and seedlings commences in the spring, inflores-
cences being initiated early in the new growth phase. Seedlings do not
normally flower in the first year; two or sometimes three years are neces-
sary. Flowering commences mid—May to mid—June, most inflorescences
within a given population opening within a two to three week period.
Individual blossoms may retain their attractiveness to nectar—seeking
insects for over a week, after which nectar production ceases and the
flowers shrivel and become discolored. Several fruits may mature on a
single inflorescence. At maturity the follicles dehisce along the ventral

1977] LYNCH: ASCLEPIAS 165

suture. Seeds, loosely attached to their comas, usually fall near the par-
ent plant. The few seeds which remain attached to their comas are wind
dispersed. After follicle dehiscence the leaves and stems wither and dry,
so that by mid-August to early September only the perennating rootstock
remains.

The phenology of A. solanoana is influenced by numerous environ-
mental factors and varies among populations and from year to year. In
general, however, the population at Stanton Creek enters each phase of
development much earlier than the population at Tedoc Mountain,
which is much higher in elevation and usually experiences lower tem-
peratures in the early spring.

METHODS

In each of the two study sites, experimental plants were selected and
marked in the early spring of 1973. The distribution of milkweeds within
each study area was mapped. Individually marked plants, 35 at each
site, were then observed during selected stages of their development
throughout the year. The Stanton Creek population was additionally
observed through the entire 1974 and 1975 seasons and into the spring of
1976. The Tedoc Mtn. population was inacessible from the winter of
1974 until mid—summer 1975, when the site was again visited.

Field observations included examination of each population to deter-
mine phenology, coordination of growth phases, establishment of new
individuals, and the death of established plants. Individual plants were
examined to determine stem number, inflorescence number, fruit initia-
tion, and fruit maturation throughout each growing season. Further
observations were made on pollinator activities, including frequency of
visitation by each insect species, duration of visits, behavior of the in-
sects in locating blossoms and obtaining nectar, and any activities of
these insects in visiting other plant species in the vicinity.

Old inflorescences (those no longer attracting nectar—seeking insects)
were examined to determine the number of pollinaria pulled per flower
and the number and location of pollinia inserted into stigmatic chambers.
Occurrence of double or triple insertions into a single stigmatic chamber
and presence or absence of an intact (unremoved) pollinarium above an
inserted pollinium were recorded. The species and sex of captured insect
visitors were determined, as were the total number of pollinaria on each
insect, the position of each pollinarium attached to the insect, and num-
ber of pollinaria attached indirectly by attachment to another pollin-
arium. Each pollinarium was scored as intact, missing one pollinium, or
missing both pollinia. Insects of the same species visiting flowers of other
plant species in the community were examined, but no attached pollin-

aria were observed.

Mature fruits were allowed to dry and dehisce at room temperature.

_ Average numbers of seeds per pod were estimated from thirty or more

166 MADRONO [Vol. 24

follicles for each year data were recorded. Average seed weights were
established from 200 seeds randomly selected from 30 mature fruits.
Ranges in seed weights were each established from 50 individual seeds in
the average seed weight sample.

OBSERVATIONS AND DISCUSSION

Plant Productivity: In comparison with most milkweed species,
A. solanoana produces a small biomass and averages fewer than four
slender stems and 15 pairs of leaves per plant. The most robust plant
observed produced 16 stems from its rootstock and had 34 inflorescences.
Several plants at each site produced only a single stem and many had
no flowers (of 118 plants at Tedoc Mtn., e.g., 35 were non—flowering).
At the Stanton Ck. site no first-year plants produced flowers and only
four of the eight produced flowers in the second year. Of these four
plants, two set no fruit, one produced a single mature fruit, and one pro-
duced four. All plants observed in the third year of growth produced
inflorescences. At Tedoc Mtn. in 1975 a late freeze damaged many stems
and inflorescences. However, an area encompassing 83 plants in 1973
contained 107 in 1975; 16 new plants were found and eight established
plants had failed to emerge. Of the 16 new plants nine flowered, produc-
ing eight fruit on four plants.

Table 1 summarizes vegetative and reproductive growth of the 35
marked plants at each site. There is a direct relationship between num-
ber of stems and number of inflorescences per plant, as well as between
number of stems per plant and number of fruits matured. At each site,
inflorescences matured in less than one month, the majority opening
within a two-week period. The few late—maturing inflorescences in each
population were often significantly smaller and had fewer flowers than
those maturing earlier.

The number and average weight of seeds per fruit did not vary signifi-
cantly from year to year or between sites. In 1973 the Tedoc Mtn. site
produced .021 fruits per flower for a 2.1% fruit-set level and an esti-
mated ratio of .742 seeds per flower. Fruit—set levels for the Stanton Ck.
site include an estimated 2.7% for 1973, 2.4% for 1974, and 1.8% for
1975. The overall fruit—set level for the Stanton Ck. site was 2.3% and
the seeds per flower ratio was .719, both remarkably close to the esti-
mates for Tedoc Mtn.

Flowers and Pollinia: A total of 1000 flowers was examined for
the number and position of pollinia pulled and inserted (Table 2). Data
for the Stanton Ck. population in 1973 and 1975 are based on counts of
100 flowers picked randomly, ten each from ten different plants within
the population. The 1974 data at Stanton Ck. and the 1973 data from
Tedoc Mtn. are based on an examination of 400 flowers per site taken
from 20 inflorescences picked randomly from 20 different plants in the

LYNCH: ASCLEPIAS 167

1977]

uv IO} UdY¥e} 19M S}UNOD

‘oaunf SULINp sseUWep.9za01f 0} INP UIye} JOU VIM S}UNOD -
‘syur[d ¢g SuIpNpoUr vole

UIeJUNOT IOpay JW “Youd uojuejys ye sjueld ¢¢ jo uonejndod sigue sy} 10} Usye] IaM SY}eeIp PUP SSUT[P9es JO S}UNOD

= = (8-0) am a (9-1) (asuey )

8 OI a = Sy ae = 61°72 IDPIIAY
SL6I

(0'6Z-9'ST) (9b-1Z) (¢72-0) (79-0¢) (21-1) C=) (93Uv y )

—_ = Oc LT OVE 1G Sy CLS eve IDRIIAY
eL6l
“NIW JOGHUL

_ ae oa ad (+I-0) Ci) (o3uey)

0 Or oe oa = = Bs'e 6r'e ISPIIAY
9261

Chve=C Cl) EV -1G) (61-0) (85-8) Gael ©) (SI-T) (a5uey )

0 v 76 91 Cte COP e9o¢ CCL I'v IBRIIAY
SZ6l

= = (82-0) (S9-T¢) (62-0) Cats) (osuey )

0 p a = 616 w Lb 80°8 igs aSRIDAY
vol

ay. a (OS?Vs) (7b—-0Z) (62-0) (¢9-ZT) (ve-0) (91-1) (o3Uey )

ase ae not PLE ET, 68e Lvl 99° IDVIIAY
oLO
“MO NOLNVIS
:Syyevp pspueyd 3 UL JINAY yd aod LOYUL pueyd jueyd IBIA
jo MOU JU JYSIOM tod syinay 40d od tod pur
JaqunN Jeaquiny PaaS Sp99G JIN} YA SLIMO[ JOyuy SWS JUS

NIVINNOJ OOdGAL GNV MAID NOINVLS LV VNVONVIOS SvIdaIOSY AO ALIAILONGOUNg INVIG ‘[ adv

168 MADRONO [Vol.24

TABLE 2. FREQUENCIES OF POLLINARIA REMOVED AND POLLINIA INSERTED IN
ASCLEPIAS SOLANOANA FLOWERS.

Site No.of No.of Pollinaria Extract Pollinia Slits Propof
and inflor flowers removed acts per inserted inserted pollinia in
year examined examined per flr flower! perflr per flr intact slits?
STANTON CK.

1973 TOTAL 10 100 3:85 4230 1.61 1.30 230
1974 TOTAL 20 400 4.43 10.86 1.81 1.66 151
Peripheral 20 200 4.34 10,12 1.79 1.65 .182
Central 20 200 4.52 11.72 1.83 1.67 A120
1975 TOTAL 10 100 3.95 7.80 155 1 S35) .194
SITE TOTAL 40 600 4.25 O:7 7 173 1755 70
TEDOC MTN.

1973 TOTAL 20 400 2.91 4.36 123 112 484
Peripheral 20 200 3.03 4.66 1.26 1.14 478
Central 20 200 2.78 4.06 {24 i a fa 489

1 Mean number of extraction acts per flower which would yield an expected number
of extractions equal to the observed number of extractions.

2 Proportion of pollinia inserted into stigmatic chambers of slit regions from which
the flower’s own pollinarium had not been extracted.

population. From each inflorescence, ten flowers from the periphery and
ten from the central region were scored.

At both sites a rather high percentage of the pollinaria had been
removed: at Tedoc Mtn., 58% and at Stanton Ck., 85%. In the 1974
season at Stanton Ck. 1772 of the 2000 pollinaria from 400 flowers
(87%) had been extracted.

If we view the system of pollination of Asclepias flowers from the
standpoint of behavior necessary to extract a pollinarium, it is obvious
that the outcome of an event which might lead to extraction of a pollin-
arium from any given flower is not independent of the outcome of pre-
vious such events. If we define an extraction act as one which would
remove a pollinarium if that pollinarium were still intact, then it follows
that an initial extraction act on a given flower would have a 100%
chance of extracting a pollinarium, leaving only four intact on that
flower. A second extraction act on the same flower would have only an
80% chance of removing a pollinarium. The probability, therefore, that
three consecutive extraction acts would remove three pollinaria is .48
(1.0 x 0.8 x 0.6). If the second extraction act is performed on the slit
from which: the initial pollinarium was removed (a .2 probability) then
no removal would occur, and a third extraction act would still have an
80% chance of removing a pollinarium. The probability of two polli-
naria being removed through this sequence is .16 (1.0 x 0.2 x 0.8).

The differences between pollinaria removed per flower among the sets
of data are even greater when viewed in terms of insect behavior (extrac-

1977] LYNCH: ASCLEPIAS 169

tion acts) required to pull pollinaria. When an increasing number of
pollinaria is removed from flowers of a given population, an increasingly
greater number of extraction acts must be performed to extract addi-
tional pollinaria. By appropriate calculations we can establish the mean
number of extraction acts per flower which would yield an expected num-
ber of extractions. From an observed 2.91 pollinaria removed per flower
at Tedoc Min., and 4.25 pollinaria removed per flower at Stanton Ck., we
estimate it would take 4.36 and 9.77 extraction acts per flower, respec-
tively, to pull an equivalent number of pollinaria. Seemingly small differ-
ences in extractions between yearly samples may be rather large when
viewed from the standpoint of extraction acts. It would take, for example,
an estimated 3.06 additional extraction acts per flower to produce the
additional .58 extractions per flower observed for Stanton Ck. in 1974
as compared to 1975.

Unless otherwise stated, all correlations between pollinia and either
inflorescences, flowers, or insects were statistically examined for goodness
of fit using a y° test at a level of significance of .05. All statistical tests
followed the procedures and limitations outlined in Siegel (1956). Ex-
traction data were subjected to several statistical tests to assess random-
ness of pollinarium removal among and within inflorescences. Under the
assumption that any pollinarium is equally likely to have been removed,
regardless of its position in the flower or in the inflorescence, and using
actual data on the mean number of extractions per flower, we can calcu-
late the expected numbers of flowers with zero through five pollinaria
removed. No significant departure from randomness could be detected at
the 5% confidence level for any of the observations.

Based on a y? test, there are significant differences among inflores-
cences at each of the sites in terms of numbers of pollinia inserted per
flower. In spite of this, using the Kendall Rank Correlation Coefficient,
there was no significant correlation between number of flowers per inflo-
rescence and either numbers of pollinaria extracted or numbers of pol-
linia inserted per flower. Using the same test, however, a significant
direct correlation (P > 99.8%) can be demonstrated between the rate
of insertion and extraction per inflorescence (Fig. 9). It seems likely,
therefore, that inflorescences at each site differ in both insertion and
extraction activity per flower, but that this is not closely related to inflo-
rescence size.

A Fisher randomization test was used to examine the hypothesis that
the number of pollinaria extracted from flowers is independent of position
in the inflorescence. Extractions from peripheral flowers were compared
with those from central flowers using inflorescences from both sites, but
no significant departure from randomness was detected. Also, a binomial
test demonstrated no significant departure from randomness for inser-
tions as a function of position in an inflorescence.

170 MADRONO [Vol. 24

INSERTIONS PER FLOWER

1 2 3 4 )
POLLINARIA EXTRACTED PER FLOWER

Fic. 9. Observed relationship between insertions per flower and _ pollinaria
extracted per flower. Figures in outline each represent the mean values for ten
flowers from one inflorescence. Solid figures represent the mean values for all
flowers analyzed at a site for a given year. The yearly mean values were used to
estimate the slope of the line. The key to the shapes of the various figures is as
follows: = Stanton Ck., 1973; A = Stanton Ck., 1974- © = Stanton Ck, 1975.
® — Tedoc Mtn., 1973.

The number of pollinia inserted was significantly lower than the num-
ber extracted in all cases, averaging 1.73 per flower at Stanton Ck. and
1.23 per flower at Tedoc Mtn. The frequency distribution of rates of in-
sertion of pollinia on a per flower basis indicated a significant deviation
from expected frequencies for all but one set of observations, but no con-
sistent pattern could be detected to explain this departure from random-
ness. Although the number of slits with multiple insertions was rela-
tively low (10.1%) it is possible that the occasional insertion of two
pollinia during a single insertion event accounts for some of non—
randomness of the frequency distribution of insertions on a per flower
basis. To assess the possibility that simultaneous insertions in adjacent
slit regions might have occurred, flowers with two insertions in different
slits were examined to determine whether they were positioned in adja-
cent or opposite slit regions. No significant departure from randomness
could be detected, as an almost equal number occurred in both categories.

Perhaps more significant is the relationship between insertions and
extractions for any given slit region. Correlations ranged from slightly
negative, but not significant in all central flowers to significantly nega-
tive in all peripheral flowers. A disproportionately high number of inser-

1977] YLNCH: ASCLEPIAS Li

tions were in stigmatic chambers of slit regions with intact pollinaria.
This may indicate that pollinia inserted by insect visitors are more
likely to remain inserted in a given slit region when the flower’s own pol-
linarium is present and intact. This finding would at least partly concur
with Wyatt’s (1976) observations that, in A. tuberosa, the corpusculum
of the intact pollinarium often serves as the point at which the transla-
tor arm to an inserted pollinium is broken, thereby allowing the polli-
nium to remain lodged against the stigmatic surface. My observations
indicate that this may often be the case in A. solanoana flowers as well.
Therefore, while increased numbers of extractions reduce the efficiency
of additional extraction acts in producing additional extractions, they
may also reduce the likelihood of additional insertions. The net effect is

that the relationship between extractions and insertions is nearly linear
(Fig. 9).

Pollinator Behavior: At both sites observations of pollinator
activity were made through portions of each daylight hour. Days of
visitation were arranged to span the flowering periods at each site. At
Tedoc Mtn. the only floral visitor effective in extracting pollinia from
A. solanoana flowers was a large carpenter bee, Xvlocopa californica
(Hymenoptera: Xylocopidae). At Stanton Ck., X. californica was also
the principal floral visitor, but a bumble bee, Bombus vosnesenski (Hy-
menoptera: Apidae), and the honey bee, Apis mellifera (Hymenoptera:
Apidae) also visited flowers. No other floral visitors were observed to
extract or carry pollinaria from A. solanoana flowers.

X ylocopa bees typically flew into the study areas at high speed, ‘“‘buzz-
ing” the population and making angling flight patterns over the area
before sweeping in to hover above a plant. The bees sometimes repeated
the entire sequence several times before landing on an inflorescence. After
landing, they stayed on an inflorescence from one to 535 seconds (Fig.
6). Bees often flew from one inflorescence to another, hovering briefly
before landing. More frequently, however, they visited another inflores-
cence on the same plant by crawling short distances along a leaf, the
stem, or on the ground. Usually carpenter bees crawled over the surface
of an inflorescence probing for nectar in a large number of hoods and then
flew several feet from the plant before settling on another plant, re-
entering a “buzzing” pattern, or flying off. Wariness of the insects and
the terrain often prevented accurate determination of the sex of the
insects, but field observations and captured specimens indicated a ratio
of approximately 4:1 male: female visitations at both sites. No behavioral
differences were observed in male and female carpenter bees visiting 4.
solanoana.

Bombus vosnesenskii males also visited A. solanoana inflorescences at
the Stanton Ck. site, although at a consistently lower rate than that of
carpenter bees. No worker or queen Bombus was observed or captured.

172 MADRONO LVol. 24

Bumble bees were consistently more direct in visitation behavior, flying
into the study area and usually hovering briefly above a particular plant
before landing. On the inflorescence bumble bees displayed the same
behavior patterns as the carpenter bees. They flew from one inflorescence
to another rather than crawling along foliage. Duration of the visit to
each inflorescence varied greatly, ranging from two to 486 seconds.

Honey bee workers were observed as visitors only at the Stanton Ck.
population in 1974. Frequency of visitation by the honey bees was less
than 5% of the insect visits that year. Honey bees were quite direct in
their approach to an inflorescence, flying much more deliberately than
either Xylocopa or Bombus, and moving much more frequently and
freely from inflorescence to inflorescence and from plant to plant. Visita-
tion time varied greatly, ranging from a mere touchdown and take off to
a stay of 225 seconds. Honey bees also used the entire inflorescence as a
platform and crawled at random over the flowers searching for nectar
(Fig. 5).

Insects and Pollinaria: Table 3 summarizes the analysis of loca-
tion, condition, and attachment of 2706 pollinaria on insects captured on
blossoms of A. solanoana. Of the three species visiting flowers at the
Stanton Ck. site, Apis mellifera carried the greatest number of pollinaria
per insect (48.4). On Apis all the observed pollinaria were attached di-
rectly or indirectly to hairs or spines on the legs. Nearly half of these
were on the hind leg attached to the pollen—collecting apparatus of the
tibia and to the enlarged first tarsal segment.

Bombus had fewer pollinaria per insect (29.1) than A pis and slightly
fewer than the Xvlocopa (32.4) at Stanton Ck. As with A pis, the only
pollinaria attached to bumble bees were on the legs (Fig. 7), but these
were nearly evenly divided among the three leg pairs. Approximately
three fourths of all pollinaria on Bombus were attached to hairs and
spines of the tibia. Nearly all remaining were on various tarsal segments
or the claws.

At both sites Xylocopa males and females collected many pollinaria
not only on their legs but also on the fringe of hairs along the posterior
portion of the abdomen. The number of pollinaria per insect at Tedoc
Mtn. was significantly higher (163.0 for females and 74.2 for males)
than for the Xyocopa at Stanton Ck. Greater numbers of pollinaria on
females are attributable to the additional pollinaria attached to the
longer and more plentiful hairs and spines of the hind tibiae. The num-
ber of pollinaria on the abdominal brush of females was not significantly
different from that of males. Ratios of pollinaria found on each leg pair
of male carpenter bees from either site and bumble bees from Stanton Ck.
were not significantly different. Both Xylocopa females and Apis work-
ers, however, had a significantly greater percentage of pollinaria on their
modified hind legs than did male pollinators.

LYNCH: ASCLEPIAS 173

1977]

‘s8a] PUIY IY} JO JUSWIBas [eSIe] ISIY posre[Us IY} 0} payoe}}e 919M VSO} JO %9' Ch =
‘syedyynow uo wnueuryjod ouo pue sirey ofoes0y} uo elieUTjod OM} Sepnpouy

oS OCS VIc 6 $C [eas EOS ace GAS Sec 6°61 206 c'6V 9047 eh) TIVLOL
DIWMAO {11D
tac 9 TP 6 VC Woe vt Or 6¢ 965 cl Ly OFOT Cvs $99 6 pr0goIK X
pItUs0 {2]DI
60 COL cut vcr vt VIL SV Vez Lvv ctl L8t O cor CS9 v pI0gojn x &
“NII OOCGHL
DAafipjaue
Ve One a 2 O' br 00 ess GE AY. 00 3 8Pr LoVe VOC V8 CVC 5 S1dV 6
11ySUaSIUsSOa
Oe C9 OsI iat C0 SCL CLG 00 ete C9¢ Oe 162 S6v LT snquiog PP
DIMAO {IDI
oS OY L&¢ LO0¢ of O'8r OTT 68e CSI 2 0c [Rae Ove 1019 SI 0090] x P
DI1UAO{1]D)I
00 vos Ll Of¢ See =o SCL | rs VST ¢ Or OLS S6r 6¢ C ng 020k x &
“MO NOLNVILS
CNI ©) ad V WHt dil aVL ddv TH IW Td IdV dL IL sx1ozeur[od

‘a pue
elieuryjod jo ssa] UO Apoq uo ays Apnys

uoI}IPUuOD UOI}ISOg UOT}ISOd

‘paAdasgo eBlieur[od [e101 Jo JUadsad sev peat CNY Ysnoiy) Jy ‘eueurjod payorye Apyaatpur Jo uonsodoig = GNI ‘Vel elulyjod ou y WAN
eueuljog = 9 {umntuyjod auo yyWM eleUl[og = g ‘ eIuTod om} yyM eeUTOg = Vy :e1oway = WHA :ePIqIL = AIL ‘S41eL = UVL : UaWOpqy
= Gay {32[ 3ut = TH {321 PIPPI = IW {32] WoI = Jy {yesur sad eleurjod jo raquinu osei9aAW = [dV ‘eeurjod jo Jaquinu [ejoL — dL
‘sy9suT JO JaquINU [e}O], = [], ‘VNVONVIOS SVIdATOSY ONILISIA GILOITIOD SLOASN] NO VINVNITIOd JO NOILVOOT GNV AONANOAIT “¢ ATAVE

174 MADRONO [Vol. 24

Pollinaria were examined to determine whether the “‘slit” or corpus-
culum was attached directly to the insect’s body or indirectly to another
pollinarium carried by the insect. The proportion of indirect attach-
ments was low, ranging from 13.3% in Xylocopa males from Tedoc Mtn.
to .9% for females of the same species at the same site. In all, only 5.6%
or 152 of 2706 pollinaria were attached indirectly, 58% of these on the
male carpenter bees at Tedoc Mtn.

The condition of pollinaria, intact (condition A), one pollinium re-
moved (condition B), or both pollinia removed leaving only the corpus-
culum (condition C), were also examined. On A pis workers and X vlocopa
males the ratios among A, B, and C pollinaria were roughly similar, honey
bees tending to have slightly more A pollinaria, and the male carpenter
bees more C pollinaria. Both Bombus males and X vlocopa females, how-
ever, had significantly higher percentages of C pollinaria. In Bombus
these C pollinaria were evenly divided among the three leg pairs, while in
the female carpenter bees more C pollinaria were on the pollen—collecting
hairs of the hind legs. This may indicate that the insects actively remove
pollinia by their cleaning behavior in addition to inserting them into
stigmatic chambers. Also of note was the existence of a small number of
A pollinaria with pollinaria attached to them. This indicates that the
introduction of a pollinium into a slit or the attachment of the translator
arm of that pollinium to the corpusculum of the flower’s pollinarium does
not necessarily result in a successful insertion.

Plants and Insects: Accounts of observations of pollinators on
other species of Asclepias (Robertson, 1929; Woodson, 1954; Macior,
1965; Willson and Rathcke 1974), as well as personal observations in
California, Arizona, and Mexico, have shown that the majority of Ascle-
pias species attract numerous kinds of floral visitors, many of which are
effective in extracting pollinaria. With A. solanoana, however, only two
native insect species were observed to pull pollinia from the flowers.

The inflorescences of A. solanoana are rather compact, the individual
flowers being tightly clustered into a hemispherical head. Each inflores-
cence is erect on its peduncle, providing a dome-shaped platform for
visiting insects. The gynostegium of each flower is situated well above the
hoods so that, upon landing, the insects utilize the gynostegia as foot-
holds as they crawl about the flowers gathering nectar. Situated as they
are, the slit regions are easily accessible to hairs and spines on the legs
of insects. The large, hairy carpenter bees seem particularly well suited
for the removal of pollinaria since they also drag their abdominal brush
across the exposed slit regions (Fig. 6). Bumble bees did not make appre-
ciable body contact with the exposed gynostegia and collected no polli-
naria on their abdomens. The bodies of honey bees, nearly devoid of hairs
or spines, also collected no pollinaria.

Significantly fewer pollinaria were pulled per flower at Tedoc Mtn.

1977] LYNCH: ASCLEPIAS 175

than at Stanton Ck. Therefore, it would be expected that the Tedoc Mtn.
pollinators would tend to pick up more pollinaria per visit than those at
Stanton Ck. The Xylocopa at Tedoc Mtn. did, in fact, carry more polli-
naria per insect than those at Stanton Ck.

The presence of Apis at Stanton Ck. in 1974 may be important in the
explanation of the slightly but significantly greater number of extrac-
tions per flower that year than in previous or following years. Since,
theoretically, three additional extraction acts per flower would be neces-
sary to extract the additional pollinia, the introduction of an additional
source of pollinators (Apis) could account for this difference. Another
consideration is that the number of available pollinators at the Tedoc
Mtn. site may have been smaller in proportion to the number of plants
available at Stanton Ck. The population size at Tedoc Mtn. was much
larger, numbering over 150 plants, as compared to an initial total of
about 35 plants at the Stanton Ck. site. Also, while two other populations
of A. solanoana, totaling an additional 100 or more plants, were located
within three miles of the Tedoc Mtn. study site, extensive ex olorations
near the Stanton Ck. site revealed fewer than 15 additional plants.

Additional sources of nectar and pollen may be significant in limiting
pollinator activity on 4. solanoana. Female Xvlocopa, for example, car-
ried pollen of other plant species. Male and female carpenter bees visited
Arctosiaphyvlos viscida flowers early in the spring at the beginning of
flowering of 4. solanoana at Stanton Ck., and a few X. californica and
several X. tabaniformis ssp. orpifex visited Penstemon breviflorus sso.
glabrisetalis near the study site. No pollinaria, however, were found on
the three X. californica captured and examined. Carpenter bees at the
Tedoc Mtn. site were not observed visiting other plant species in the
vicinity of the study site.

Only males of Bombus vosnescnskii were observed on A. solanoana at
Stanton Ck., but numerous other individuals of the same species, both
males and workers, and several Xylocofa tabaniformis sso. orpifex ob-
tained nectar and pollen from Phacelia imbricata less than 100 m from
the study site. However, there were no pollinaria on any insects collected
on Phacclia. These observations suggest that the floral constancy of the
insects may limit the number of visitors to A. solanoana blossoms. For
example, while X. tabaniformis was present near the Stanton Ck. study
site and could have been an effective pollinator of A. solanoana, no indi-
viduals were observed visiting any Asclepias plants.

Field observations of insect visitation indicated that plants within each
study population were not visited equally, but that microrelief, number
of open inflorescences, and general position of particular plants within
the population seemed to affect pollinator preference. Analysis of extrac-
tion and insertion data for the sites, however, indicated that frequency
of insect visitation and subsequent pollen transfer did not limit mature
fruit production. In all inflorescences examined, the number of insertions

176 MADRONO [Vol. 24

far exceeded the number of fruits produced per inflorescence. Further-
more, it appeared that no more than one pollinium per flower was neces-
sary to effect fruit set and full seed production. In an examination of
flower parts from tips of healthy, rapidly maturing follicles at Tedoc
Mtn., three of ten contained only a single pollinium in one of the five
stigmatic chambers, a percentage similar to that found in flowers which
did not set fruit.

Insertion of pollinia into flowers of the same plant from which they
were produced might partially account for the high number of pollinia
inserted for each fruit produced. If A. solanoana is largely or completely
self sterile, a condition assumed for most milkweed species (Woodson,
1954; Wyatt, 1976), then pollinia inserted in self pollinations should be
discounted in any estimate of successful pollen transfers. However, exces-
sive numbers of pollinia inserted in comparison with numbers of fruit
set and the presence of many abortive fruit argues strongly against the
hypothesis that insufficient pollination is the cause of relatively low fruit
set in both populations. Galil and Zeroni (1969) observed that, in A.
curassavica, pollen tube germination occurred only where the convex
margin of the pollinium contacted the stigmatic surface. Nearly all the
inserted pollinia of A. solanoana examined, however, were oriented with
the convex margin in contact with the stigmatic surface. Partially in-
serted pollinia were not counted in this study.

In experimenting with populations of A. syriaca in Illinois, Willson
and Rathcke (1974) hypothesized that selection favored a balance be-
tween energy allocated to production of pods and seeds and energy allo-
cated to production of surplus flowers which increase the output of pol-
linia, though at the expense of decreasing efficiency of seed production
per flower. This model may well be applicable to A. syriaca. In A. solano-
ana however, the lack of correlation between the number of flowers in an
inflorescence and rates of extraction or insertion, the relatively narrow
range of variation in inflorescence size, and the seeming “overkill” in
terms of numbers of pollinia inserted per inflorescence, make this hy-
pothesis untenable. The relatively large number of small flowers pro-
duced may be important in attracting insects and in providing sufficient
quantities of nectar to maintain a constant rather than occasional rela-
tionship with them. The majority of Asclepias species I have studied
show a distinct seasonal variation in number of flowers per inflorescence
with larger inflorescences with more flowers produced early in the season.
The size of inflorescences may be correlated with vigor and rate of growth
of the whole plant which may affect the size of floral apices which pro-
duce inflorescences. If this were so, numbers of flowers per inflorescence
would not only vary throughout the year, but from year to year. Fruit
production might also vary from year to year, not as a direct consequence
of inflorescence size, but rather in response to greater allocation of energy
for reproduction during periods of vigorous growth.

1977] OLIVIERI & JAIN: HELIANTHUS 177

ACKNOWLEDGMENTS
I thank Drs. H. G. Baker, David C. Randall, Robbin W. Thorp, and Grady L.
Webster for their advice, technical assistance, and critical review of this manuscript.

LITERATURE CITED

Farcri, K. and L. vAN bER Pir. 1971. The principles of pollination ecology. Per-
gamon Press, New York.

GatiL, J. and M. Zeroni. 1969. On the organization of the pollinium in Ascelpias
curassivica. Bot. Gaz. 130:1-4.

Grant, V. 1949. Pollination systems as isolating mechanisms in angiosperms. Evo-
lution 3:82-97.

KRUCKEBERG, A. R. 1954. The ecology of serpentine soils. III. Plant species in rela-
tion to serpentine soils. Ecology 35:267-274.

. 1969. Soil diversity and the distribution of plants, with examples from

western North America. Madrono 20:129-154.

Macior, L. W. 1965. Insect adaptation and behavior in Asclepias pollination. Bull.
Torrey Bot. Club 92:114-126.

Proctor, M. and P. Yeo. 1972. The pollination of flowers. Taplinger Pub. Co.,
New York.

RopBerTson, C. 1929. Flowers and insects. Published by Author, Carlinville, Ill.

SIEGEL, S. 1956. Non parametric statistics. McGraw Hill, New York.

STEBBINS, G. L. 1970. Adaptive radiation of reproductive characteristics in angio-
sperms. I. Pollination mechanisms. Ann. Rev. Ecol. Syst. 1:307-326.

. 1974. Flowering plants: evolution above the species level. Belknap Press,

Harvard Univ., Cambridge, Mass.

Waker, R. B. 1954. The ecology of serpentine soils. II. Factors affecting plant
growth on serpentine soils. Ecology 35:259-266.

WuitTAKeEr, R. H. 1954. The ecology of serpentine soils. IV. The vegetational
response to serpentine soils. Ecology 35:275-288.

Witson, M. F. and B. J. Rathcke. 1974. Adaptive design of the floral display of
Asclepias syriaca L. Amer. Middl. Nat. 92:47-57.

Woonson, R. E., Jr. 1941. The North American Asclepiadaceae. I. Perspective of
the genera. Ann. Missouri Bot. Gard. 28:193-244.

—. 1954. The North American species of Asclepias, Ann. Missouri Bot. Gard.
41:1-211.

Wyatt, R. 1976. Pollination and fruit set in Asclepias: A reappraisal. Amer. J. Bot.
63:845-851.

VARIATION IN THE HELIANTHUS EXILIS-BOLANDERI
COMPLEX: A REEXAMINATION

A. M. Otivierr* and S.K. JAIN

Department of Agronomy and Range Science
University of California, Davis 95616

In his classic monograph on the evolution of Helianthus annuus L. and
H. bolanderi Gray, Heiser (1949) analyzed variation in natural popula-
tions as well as artificial hybrids that led him to the following conclu-

* Present address: Istituto di Agronomia, Universita di Padova, Padova, Italy.

178 MADRONO [Vol. 24

sions: “Of the entities comprising H. bolanderi, the one Gray designated
H. exilis appears to be confined almost exclusively to areas of serpentine
outcrops in many of the foothills regions of California, whereas true ZH.
Bolanderi occurs in the valleys as a weed. The two races of H. Bolanderi
conform to the definition of the ‘‘ecotype” of the experimental taxono-
mist.”’ Heiser (1949) discussed the role of introgressive hybridization
between H. annuus, an introduction into California by the American In-
dian in recent times, and H. bolanderi (native foothills race). Accord-
ingly, he provided a taxonomic key for the two races of H. bolandert,
annuus x bolandert hybrid swarms, and for the western races of H.
annuus. A hybrid swarm between H. annuus and H. bolanderi was exten-
sively studied by Stebbins and Daly (1961) in which a hybrid index
based on six morphological traits was used to analyze population changes
over an eight—year period. Although Heiser (1949) raised issues about
the definition of species boundaries and the tentative nature of his con-
clusions regarding the role of introgression, and later (Heiser, 1973)
concluded that “no additional evidence has appeared to support or to
reject the hypothesis of the origin of the weedy race of H. bolanderi
through introgression,” the sunflower story is frequently cited as an out-
standing example of introgression (e.g. Grant, 1971; Briggs and Walters,
1969).

Here we shall briefly reexamine evidence from population studies of
two races of H. bolanderi (the ex:lis—bolanderi complex) and weedy H.
annuus in relation to certain systematic, genetic, and statistical aspects
of introgression. Morphological variation for a wide range of characters
and in three new collections of H. exilis provided the basis of a multi-
variate analysis of these taxa. Preliminary data from the electrophoretic
assays of allozyme variation are presented as a test of specific gene trans-
fers through introgression. Two other lines of study, namely, the cyto-
genetics of hybrid swarms and the ecological tests of the adaptive role of
gene exchange, will be discussed in another paper. It should be noted that
introgression resembles the common backcross method of plant breeding,
and therefore detailed analyses of introgresive hybridization are of wide
interest in both evolution and crop genetics.

MATERIALS AND METHODS

Fourteen populations were sampled during the summers of 1973, 1974
and 1975 by harvesting 50-100 individuals along two or three linear tran-
sects per population (Table 1). Twenty to 75 plants were scored for
plant height*, branching index (1 = none to 5 = extensive branching),
stem diameter, stem pubescence*, stem color (1 = green to 4 = red),
number of opposite leaf pairs, number of unpaired leaves’, leaf length”,
and width*, leaf shape index* (1 = linear, 5 = cordate), leaf margin

* Characters marked by an asterisk were also scored by Heiser (1949).

1977] OLIVIERI & JAIN: HELIANTHUS 179

TABLE 1. SPECIES AND PROVENENCE OF THE POPULATIONS STUDIED,

No. of
plants
Species Locality Code studied
1. H. annuus Yolo Bypass; 0.5 mi. S of Ist Sisal 24
is bridge of I-80 from Davis to
= Sacramento.
S
S 2. H. annuus and
H. annuus x 40 meters N from site DVS DVS =2 8:
H. Bolanderi 3, across the railways.
/ 3.H. Bolanderi About 3 mi. NW of Knoxville KNX 1 20
(located N of Lake Berryessa) ,
off Berryessa-Knoxville Road
2 and along Cedar Creek Road.
© \ 4. H. Bolanderi 1.5 mi. W of KNX 1, KNX 2 42
near Hunting Creek.
5. H. Bolanderi Patch located near the KNX 2a 20
Campground, about 50 meters
from KNX 2.
S 6. H. Bolanderi 4 mi.S of Williams WLS Ey!
3 Colusa Co., along Rd. 15.
3
S$ ( 7.H. Bolanderi 4 mi. W of West Butte, WBT 46
S Sutter Co., along Pass Road.
oO
S 8. H. Bolanderi 2.5 mi. NW of Sutter, STR 39
Sutter Co., along Pass Road.
9.H.Bolanderiand 5.4 mi. E of Davis, Yolo Co., DVS 3 to
H. Bolanderi x in ditch area between railways
H. annuus and Frontage Road along 180.
|
10. H. Bolanderi About 7 mi. NW of Knoxville, KNX 5 36
off Morgan Valley Road.
S | 11. H. Bolanderi About 5 mi. NW of Knoxville, © KNX 7 a6
S along Morgan Valley Road.
"S
> ( 12. H. Bolanderi 3 mi. NE of Middletown, MTW 1 26
Lake Co., along $29.
s 13. H. Bolanderi 4 mi. NE of Middletown, MTW 2 28
Lake Co., along $29. .
14. H. Bolanderi ca. % mi. N of KNX 2, KNX 3 22

along Hunting Creek.

dentation (1 = none, 3 = extensive), head diameter*, disk diameter”,
number of ray flowers*, ray width, ray flower shape, floret length, floret
tube length, basal floret shape index* (1 = not swollen to 3 = largely
swollen), floret color (1 = yellow, 2 = light red, 3 = dark red), stigma
apex color (1 = yellow, 2 = orange, 3 = red), lateral and central palea

180 MADRONO [Vol. 24

cusp length, palea apex color, involucral bract number, bract length* and
width*, bract pubescence*, achene length* and width*. Numerical taxo-
nomic analyses of vegetative and floral characters were carried out with a
BMD principal component analysis program.

Samples of populations KNX 1, KNX 2, DVS 1, and DVS 2 were
grown during 1975 summer in the greenhouse in 18 cm pots and UC soil
mix, and scored for a subset of 15 characters to study population differ-
ances under a common environment. To measure the genetic similarities
among different groups of populations, electrophoresis for isozyme varia-
tion was carried out using standard horizontal starch gel techniques de-
scribed by Shaw and Prasad (1970), with minor changes to adapt to our
materials. Three— to four-week-—old seedlings were used for sample ex-
tracts. Three enzyme systems (leucine aminopeptidase, phosphogluco-
mutase, phosphoglucose isomerase) were scored. Data on the allelic com-
position at two alcohol dehydrogease loci were kindly supplied by Dr. A.
Torres of the University of Kansas on a small set of samples. For present
purposes, a phenotypic analysis of variation is presented in terms of the
presence vs. absence of well-developed and consistent bands within popu-
lations, rather than gene frequencies, since genetic analyses using prog-
eny tests are not yet completed for most of the loci. Accordingly, data are
summarized in terms of the number of different alleles present in several
populations, and estimates of similarity are based on Jaccard’s index,

C
J = ———_.., where c = number of common alleles between popula-
Riga aa Oe a ek ©
tions A and B: a and b are total numbers of alleles in populations
A and B respectively.

RESULTS AND DISCUSSION

Means were determined for 15 characters and ten selected populations
based on the field samples (Table 2). Heiser (1949) had noted that
important taxonomic distinctions were based on the shape of involucral
bracts, the nature of the palea or chaff (length, texture and angle of awn
of the palea or chaff), and the overall size of the plants. Leaf shape, disk
diameter, ray number, and achene length were included in his ‘“‘taxonomic
key features” (Heiser’s Table 1). Our data confirm his observations on
these characters for describing the two H. bolanderi “ecotypes” and
weedy H. annuus (Table 2). However, it should be noted that our H.
exilis populations are a distinct group based on the same characters and
in fact represent an extreme below the following ranges for the foothills
H. bolanderi:

181

HELIANTHUS

OLIVIERI & JAIN:

1977]

(uruZ) YPM paeg
(wur) YyySua] poss

xXopul
guaosueqnd yoRIg

(WW) YVSUII RIG

Xopul
IO[OO eUIsT}s [eddy

Xopul
advys Jo10p [eseg

(WIUL) Y}SUZ] J10[4
Joquinu Avy

(WU) JoJUTRIP ASIC
(WW) JoJaUIeIp peofyT
xopul odeys fray
(wUr) YIBUIT Jeo'T
Xopul sulpouesg
(wd) JYSIOY JU

19}B18Y,)

€1 OT ST oT al O' ec i v7 OG
ac 8° re (Ene O~ Sar ov Tv [ee vs
97 Ce OL 67 82 oC Le 6 | vt
8 38 Vcr ct Vil 9°6 6 TT ecl oe e cl
oe OT TT OT OT LT O'¢ VC ise 6¢
cal oT on vi vl 6T ST O° Lac Oe
oP res 19 8S Ls 8S C9 C9 69 SL
Le 39 9°6 LOT 8'OT £6 vit fl 6 eT p85
VE LL vit oC! 9'¢T v6 Sel gst 98ST |
192 L162 VeVv a Lv Liss OLY O'6r Lv9 6°79
OL se Oe 67 mi fe Le 8°C ve cv LY
O'Or O82 OCs 9'8P 0'r9 0°92 9°6P O°'0S $88 0°68
Ge hve 97 ara LY pa | O% Le ee oC
Lr¢ bv V6¢ 90S eke) en Oke os £09 S80l S28
‘TXNY IXNM SXNY 7MINW IMIW SAG MES da SAC —1SAC
SI[IXO “FY S][Tq}00} AoT[ea snnuue “H
Wopue[og "H
SNOILVINdOg GILOITAS OT YOA SNVAJ AALOVUVHD AO SALVWIISH “7 AAV LE

uorRI0] pue ;
satads

182 MADRONO [Vol. 24

plant disk bract ray achene
height diameter width number length
serpentine,
uae 30-100cm 15-20mm 30-40mm 10-13 3.0-4.0mm
race (ala
Heiser )

All three populations from new locations for what we have designated as
H. exilis group are homogeneous, different, and fall below the ranges
given for the foothills bolanderi populations (cf. Jain et al., 1977 for
agronomic traits in H. exilis). Apparently, these localities were not visited
by Heiser; the access road to these populations on the Bureau of Land
Management land was completed ten years ago. Several of the localities
described on the University of California, Berkeley herbarium speci-
mens for H. exilis were revisited in 1974. A majority have very small
populations and showed a variation pattern in agreement with that de-
scribed by Heiser (1949) for the H. bolanderi group.

Analyses of intrapopulation variation are of particular interest. A
nested analysis of variance showed that (1) the H. annuus and H. bolan-
deri populations are significantly different (P < .05) for 20 out of 21
characters tested; and (2) populations of each of the two species are also
significantly different for 18 characters at P = .01 and two characters at
P = .05 levels. Estimates of the coefficients of variations (CV) for a
majority of characters were higher in the valley H. bolanderi and weedy
annuus groups than for the foothills H. bolanderi or the H. exilis popu-
lations.

Figure 1 compares the means and ranges (as well as the estimates of
standard deviations) for H. exilis vs. the two H. bolanderi groups pooled
together. The ranges overlap considerably but the H. exilis means were
consistently different from those of H. bolanderi. Diagnostic keys in such
cases would clearly require a numerical taxonomic approach as well as a
further search for certain qualitative genetic traits.

Principal component analysis for vegetative and floral characters re-
spectively reveals that populations 3, 4 and 5 representing H. exilis form
a separate group, DVS 1 and DVS 2 representing H. annuus as a second
distinct group, and the remainder H. bolanderi populations form a con-
tinuum between them (Figs. 2,3). Populations 10-14 represent the foot-
hills race of H. bolanderi which are separated from the valley race (rep-
resented by the populations WLS, WBT, STR). Population 9 (DVS
3) represents a hybrid swarm studied by Stebbins and Daly (1961)
which was recently fragmented by cultivation and road construction. It
seems to be differentiated from all others in its vegetative characters.
Overall, the multivariate analysis confirms the observations of Heiser
and ours on the four overlapping groups identified in these collections
(Table 1). Moreover, data from greenhouse materials of KNX 1, KNX 2

1977] OLIVIERI & JAIN: HELIANTHUS 183

MM 3845 60 78

APICAL STIGMA
COLOR INDEX

BRACT NUMBER

BRACT WIOTH
MM

BRACT SHAPE INDEX
APEX BRACT LENGTH
MM

BRACT MARGINAL
HAIR INDEX

SEED LENGTH
MM

SEED WIDTH Saas as Rea
MM 10 E B 20 3!

Fic. 1. Range of variation in H. exilis (E, stippled area) vs. H. bolanderi (B,
nonstippled area). Bars near E or B indicate the means and the two bars placed on
each side of the means indicate standard deviations. Note the distinctness of H.
exilis on the basis of range as well as the lower extremes for several characters.

vs. BGS vs. DVS 1 and DVS 2 gave significant intergroup differences in a
majority of the same characters. Thus, populations as grouped here seem
to represent genetically differentiated clusters. Introgression, on the other
hand, is neither supported directly nor ruled out by these observations.
Valley H. bolanderi and weedy H. annuus are most likely connected by a

184 MADRONO [Vol. 24

MWK OO “I

es
art
4p 10F 13 2
_3 (A
-A ee ae

COMPONENT I
Qe

=4 S| =2.73 50 ol 2-3. a5

COMPONENT I

Fic. 2. Principal component analysis of 14 OTUs (Operational Taxonomic Units)
using stem, leaf, and seed characters. The codes of locations (1 to 14) are given in
Table 1. Dotted boundaries around the four groups of OTUs are drawn simply to
match with the four “taxa” as noted in Table 1. No. 9 is an outlier (see text).

series of gene exchange events, and H. bolanderi and H. exilis in the foot-
hills probably represent another series of populations connected by a
two-way introgression underlying the origin of variation in H. bolanderi.
On the other hand, native H. exilis might have been highly variable and
could have colonized some disturbed areas on its own. A complete descrip-
tion of parental forms uncontaminated by hybridization is a prerequisite
to the “proof” for introgression, and as noted by Heiser (1973) in a
recent review, reliance on simply circumstantial evidence is not sufficient.

The so-called hybrid index method is often used to establish intro-
gression (e.g. Keeley, 1976). In this method a set of diagnostic charac-

1977] OLIVIERI & JAIN: HELIANTHUS 185

ea5,

( j4™

COMPONENT I

me eo oes

COMPONENT I

Fic. 3. Same as in Fig. 2 except for the use of floral characters. Note the same
four groups, in particular, the separation of H. annuus (1,2) and H. exilis (3,4,5)
from each other.

ters are often selected to pictorialize the variation in putative hybrid
swarms. If they vary in a way to show a large amount of overlap with
one of the parents but with a few characters from the other parent, intro-
gression is assumed. Many of the examples perhaps show this con-

186 MADRONO [Vol. 24

vincingly, but as noted recently by Namkoong (1966), several statistical
and genetic assumptions are rarely stated, much less tested. For example,
the number of populations required and the statistical differentiation of
hybrid index scores are fairly stringent conditions. Additive genetic vari-
ation, independence of different traits, etc. are also assumed. We devel-
oped hybrid indices (or what we prefer to call “‘species identity scores”
(SIS) since hybridity need not be presumed) using nine characters
chosen to include the six characters (Fig. 4) used by Heiser (1949) and
Stebbins and Daly (1961) in their studies and to represent a highly
heritable and more or less correlated set of nine traits (Fig. 5) in which
case the distributions of DVS 1 and DVS 2 become wider and so also for
the WBT population of H. bolanderi. The overall gradient across four
groups still remains consistent. Use of t-test for significance showed SIS
means to be significantly different among various groups, with a larger
difference when we used SIS based on the nine correlated traits (Fig. 5).
Further weighting by their respective heritability estimates obtained
from a greenhouse study (Olivieri, 1976) confirmed that differentiation
is at least partly genetic. Use of more characters with known genetic
control should improve the interpretation of the SIS scores.

Electrophoretic assays provided a series of isozyme markers (presence
vs. absence of bands on gels). Zymograms are drawn to derive the allelic
designations for various phenotypes. For alcohol dehydrogenase, surpris-
ingly, H. exilis and two cultivated varieties of sunflower had the same
alleles whereas H. bolanderi showed two unique alleles (A. M. Torres,
pers. comm.). For the other three enzyme systems our data are summar-
ized in Table 3. H. exilis (KNX 1 and KNX 2) have unique alleles at
four out of the eight loci whereas H. bolanderi populations have a great
deal of variability in both foothills and valley populations but with fewer
“unique” alleles. The estimates of Jaccard’s index based on the shared
alleles between groups taken pairwise are as follows: (a) H. exilis vs.
foothills H. bolanderi: 0.63; (b) AH. exilis vs. valley H. bolanderi: 0.57;
(c) valley vs. foothills H. bolanderi: 0.69. These estimates are based on
very small samples (20 plants per population, two populations per group)
and should be considered preliminary. However, so far there is no con-
vincing evidence to reject or accept the hypothesis that the foothills race
of H. bolanderi is more similar to H. exilis due to gene exchange. With
further genetic work on allozyme loci and extensive enzyme assays of our
collections, this method might yield a crucial test for the postulated gene
exchanges between different taxa.

For genetic variation analyses to be useful in a specific test of intro-
gression hypothesis, the key criteria, as outlined by Heiser (1973), in-
clude increased hybridity and genetic variation, frequently through the
occurrence of a few alleles characteristic of species A in the populations
of species B living in the areas of sympatry and habitat changes in recent
past. Experimental hybrids and backcrosses could provide some clues to

1977] OLIVIERI & JAIN: HELIANTHUS 187

KNX |

NUMBER OF INDIVIDUALS

10 14 18 22 26 30 34 33
SCORE
Fic. 4. Histograms showing the frequency distributions of “species identity
scores” (= hybrid indices) based on the six characters used by Heiser (1949) in his

earlier studies.

the potential for gene exchange as well as to the genetics of species differ-
ences. For example, Rick (1969) developed a test for controlled intro-
gression in tomato (Lycopersicon—Solanum.) crosses using recombination
data in the marked regions of three chromosomes. Wall and Wall (1975)
developed an experimental test in Phaseolus species based on allozyme

188 MADRONO [Vol. 24

=~ wonwn © =

—- won Oo

NUMBER OF INDIVIDUALS
=a lon

—-wurno©

(4)

KNX 3

—- W

0 2 4 6 8 10 l2 14 16
SCORE

Fic. 5. Same at in Fig. 4 except for the use of nine most highly correlated traits,
namely, plant height, leaf shape, stem diameter, diameter of head and disk, ray
number, bract number, bract length and bract width. Note that the pattern of
distribution is slightly changed toward a wider range and less overlap between the
scores for H. annuus (DVS1, DVS2) and H. exilis (KNX1, KNX2).

TABLE 3. GENETIC VARIATION AT ALLOZYME LOCI.

Number of alleles

foothills valley
Enzyme exilis bolanderi bolanderi
locus total “unique” total “unique” total “unique”
*Pgm-1 2 1 1 0 1 0
Pgm-2 2 0 2 0 2 0
Pgm-3 3 1 2 0 2 0)
*Lap-1 1 3 0 3 1
Lap -2 1 0 1 ¢) 2 1
Lap -3 3 1 3 1 4
*Pgi - 1 1 0 1 1

rU
)
(

dO
—
ae)
bo
—_
bo
_

:

1977] OLIVIERI & JAIN: HELIANTHUS 189

surveys. Several other recent reports (e.g. Sorghum—Saccharum ,; maize—
teosinte) have recently appeared in the crop science literature. The
results of this study show that evidence for introgression needs to be
examined in relation to morphology, Mendelian loci, quantitative genetics
of distinguishing characters, and appropriate statistical tests of differ-
ences among various taxa. Hopefully, the sunflowers will provide some
very exciting materials for population studies on the role of hybridization
in plant evolution.

LITERATURE CITED

Briccs, D. and S. M. Watters. 1969. Plant variation and evolution. McGraw Hill,
New York.

GranT, V. 1971. Plant speciation. Columbia Univ. Press, New York.

HEIsEr, C. B., Jr. 1949. Study in the evolution of the sunflower species Helianthus
annuus and H. bolanderi. Univ. Calif. Publ. Bot. 23:157-208.

—. 1973. Introgression reexamined. Bot. Rev. 39:347-366.

Jain, S. K., A. M. Onivierr and J. FERNANDEZ-MartTINEZ. 1977. Serpentine sun-
flower, Helianthus exilis, as a genetic resource. Crop Sci. (in press).

KEELEY, J. E. 1976. Morphological evidence of hybridization between Arctostaphylos
glauca and A. pungens (Ericaeae). Madrono 23:427-438.

NAMKOONG, G. 1966. Statistical analysis of introgression. Biometrics 22:488—502.

OxtvierI, A. M. 1976. Role of introgression in two Helianthus species. M.S. Thesis,
Univ. Calif., Davis.

Rick, C. M. 1969. Controlled introgression of chromosomes of Solanum pennellii
into Lycopersicon esculentum: Segregation and reccmbination. Genetics 26:
735-768.

SHAw, C. R. and R. Prasap. 1970. Starch gel electrophoresis: a compilation of
recipes. Biochem. Genet. 4:297-320.

STEBBINS, G. L. 1965. Colonizing species of native California flora. Jn H. G. Baker
and G. L. Stebbins (eds.), The genetics of colonizing species, p. 173-195, Aca-
demic Press, N.Y.

, and K. Daty. 1961. Changes in the variation pattern of a hybrid popu-
lation of Helianthus over an eight-year period. Evolution 15:60-71.

Torres, A. M. 1974. Genetics of sunflower alcohol dehydrogenase: Adh», nonlinkage
to Adh, and Adh,; early alleles. Biochem. Genet. 12:385-392.

Wat., J. R. and S. W. WALL. 1975. Isozyme polymorphisms in the study of evolu-

tion. In: Isozymes IV. Genetics and evolution, pp. 287-305. Academic Press,
New York.

190 MADRONO [Vol. 24

A NEW COMBINATION IN CYMOPHORA
(COMPOSITAE: HELIANTHEAE: GALINSOGINAE)

Jup1ItH M. CANNE
Department of Botany and Genetics, University of Guelph
Guelph, Ontario, Canada N1G 2W1

Specimens of 7vidax L. and Cymophora B. L. Robins. were examined
by the author during studies of the generic and specific relationships of
Galinsoga Ruiz & Pavon (Canne, in press a, b). A recent, additional
study of several specimens of the relatively poorly known Tridax vene-
zuelensis Arist. & Cuatr. indicate that this species falls within the concept
of Cymophora as discussed by Turner and Powell (1977). The transfer
of 7. venezuelensis to Cymophora is made here and comments are in-
cluded concerning interspecific relationships in Cymophora.

Tridax venezuelensis shares features of the pappus, achenes, and capit-
ulescence with 7. dubia Rose and Cymophora accedens (S. F. Blake)
Turner & Powell while resembling Galinsoga in several vegetative and
floral features (Aristeguieta, 1964; Powell, 1966). However, both T.
venezuelensis and C. accedens possess additional features that do not
occur, or occur only rarely, in 77idax and Galinsoga. These are the white
to creamy yellow corolla color; paniculate capitulescence; angular disc
achenes; and the cylindrical to subcampanulate involucre that charac-
terize Cymophora. Tridax venezuclensis differs from Tvidax proper in
achenes glabrous to strigose, not densely long villous or pilose; pappus of
fimbriate scales rather than plumose bristles; and heads less than 8 mm
diameter. The transfer of T. venezuelensis to Cymophora is made on the
basis of these morphological comparisons.

Cymophora venezuelensis (Arist. & Cuatr.) Canne, comb. nov.

Tridax venezuelensis Aristeguieta & Cuatrecasas, Flora de Venezuela
10:694. 1964. TYPE: VENEZUELA: Mriranpa, La Providencia, Sep
1936, H. Pittier 13754 CHOLOTYPE, VEN; ISOTYPES,. F! Usd:
PARATYPES, V. M. Badillo 271, 772, VEN; H. Eggers 13508, US!; 2.
Pittter 11152, GH! NY! P! US!).

Additional specimens examined: VENEZUELA: Distrito FEpeERAL, between
Naiguata and Hacienda Cocuizal, 7 Oct 1966, J. Stevermark 97465 (F); above
Chichiriviche, 1 Jul 1966, J. Steyermark & L. Aristeguieta 122 (NY, US). State
and locality unknown, 9 Aug 1891, H. Eggers 13568 (US) ; 1865, Moritz s.n. (BM).

Cymophora venezuelensis is distinguished from the other three species
in the genus by its pistillate ray florets with corollas having short inner
lobelets. The peripheral florets of other species are perfect and have
inconspicuously ligulate corollas. Cymophora venezuelensis differs in
distribution as well, being known only from northern Venezuela, whereas
other species of Cyvmophora occur in south central Mexico.

1977] CANNE: CYMPHORA 191

The four species of Cymophora fall into two rather well-defined
morphological groups that presumably reflect closeness of relationship.
Cymophora venezuclensis and C. hintonit Turner & Powell have ovate
to trullate leaves with coarsely serrate margins, petioles 2-5 cm long,
phyllaries with 15-20 veins, and, in C. venezuelensis, pales with 8-16
veins but pales absent in C. hintonii. In contrast, C. accedens and C.
pringlei B. L. Robins. have ovate to ovate—lanceolate leaves with sub-
entire to serrate margins, petioles to 2 cm long, phyllaries with 4—9 veins,
and pales with 3-5 veins. Nomenclatural recognition of these species
groups seems unnecessary in such a small genus until warranted by addi-
tional knowledge of the biology of the species.

ACKNOWLEDGMENTS
Appreciation is offered to B. L. Turner for discussing and reading the manu-
script, to a reviewer for helpful comments, and to the curators and directors of the
following herbaria for loan of specimens: BM, F, GH, NY, P, UC, US.

LITERATURE CITED

ARISTEGUIETA, L. 1964. Flora de Venezuela 10:694-696.

CanneE, J. M. In press,a. A revision of the genus Galinsoga (Compositae: Helian-
theae). Rhodora.

————. In press,b. Circumscription and generic relationships of Galinsoga (Com-
positae: Heliantheae). Madrono.

Powe tl, A. M. 1966. Tridax venezuelensis, comments on a newly described species.
Brittonia 18:284.

Turner, B. L. and A. M. PowE Lt. 1977. Taxonomy of the genus Cymophora (Aster-
aceae-Heliantheae). Madrono 24:1-6.

REVIEW

California Mushrooms, a Field Guide to the Boletes. By Harry D. Tutrers. vii +
261, 4 figs., colored microfiche with 54 figs. Hafner Press, New York. 1975. $15.95.

Californians interested in the boletes now have a treatise prepared by an inter-
nationally recognized student of this taxonomically difficult group. Previous com-
prehensive treatments of this group in North America have emphasized eastern
species. Various pcpular mushroom books have included a few bolete species, but
the identification of all but the most common and unique California bolete species
was essentially impossible, except for the specialists. It is perhaps a bit optimistic to
expect that everyone, even with the aid of Professor Thiers’ book, will be able to
identify bolete species without a good compound microscope and a good deal of
effort and experience. Nevertheless, this book provides a means, regardless of pre-
vious training, for anyone to learn the California bolete flora.

Recent taxonomic studies, including this work, in the Boletaceae and of other
groups of fungi strongly suggest that there is a greater diversity of fungal species
than previously thought. Thus, regional taxonomic treatments rather than world-
wide, or even continental, monographs will probably become the objective of the
fungal taxonomist, especially those working with the taxa of the saprobic basidio-
mycetes. Professor Thiers’ publication cn the Boletaceae of California places this
group among the few fungal taxa that are relatively well known in any region in
North America. Even so, as the authcr emphasizes, this publication probably repre-
sents only a firm starting point for an understanding of the bolete species in
California.

The author has wisely based his species descriptions on those specimens housed

192 MADRONO [Vol. 24

in his herbarium. The vast majority of these specimens were collected by the author
or his students.

In the introduction (19 pp.) Professor Thiers briefly describes the somatic, or
vegetative, phase of the boletes and discusses in more detail the bolete fruiting struc-
tures, especially those features of the basidiocarp that are of taxonomic importance.
He introduces the necessary terminology in such a manner that a serious student
will be able to use the book. Four figures are provided to supplement the explana-
tions given in the text. Techniques of study of the basidiocarps and methods of
collecting boletes are subjects that are also covered. The author describes the prob-
ably mycorrhizal association of 36 species in table form. This emphasizes the
necessity of noting the trees associated with the various bolete species when making
collections and of the probable importance of the Boletaceae as mycorrhizal formers
with forest trees. In California all bclete species occur only in forested habitats;
therefore, they are probably all mycorrhizal formers.

Professor Thiers also provides a brief history of bolete taxonomy and points out
the unique features of the species of boletes in California; i.e., the relatively high
percentage of endemic species and the absence of representative species of the genera
Strobilomyces and Boletellus.

After a synopsis of the species and subspecies of Boletaceae known from Cali-
fornia, a formal description of the family is provided followed by a key to the
genera (7) represented in California. Each genus description is followed by a key
to the species within each section.

Each species description lists the accepted name of the species as well as several
other synonyms and a list of illustrations to the species in other publications. Fifty-
four of the eighty California taxa are also illustrated in the colored microfiche. The
species or subspecies descriptions consist of a discussion of the macroscopic features
of the pileus and stipe, especially the color and texture. The context of both the
stipe and pileus are described as are the microscopic features of the tube trama,
basidia, and spores. Included here also is the spore print color, still an important
taxonomic character in the Agaricales, followed by a description of the reactions of
various parts of the basidiocarp to the several chemical tests outlined earlier in the
introduction. An appealing feature here, which is often overlooked in other taxo-
nomic treatments of the Agaricales, is that the author cautions the reader that the
nature and significance of these chemical tests are essentially unknown.

The author concludes the taxonomic description of the basidiocarps with a dis-
cussion of the habitat and distribution of each species in California. He then lists
the specimens studied and comments on certain unique features of the species,
including its known edibility properties.

The book closes with a field key to the genera, species, and subspecies treated, a
list of references, and an index to the various taxa.

This publication is designed for both the professional and amateur mycologist. It
is reasonable to expect, in my opinion, that a serious amateur, equipped with a
good compound microscope, would be able to identify bolete specimens using this
text. While field keys are useful for both the amateur and specialist, mycological
taxonomy has long since reached the stage at which microscopic examinations are
essential for correct determinations. Among the Agaricales this is probably espe-
cially true for the Boletaceae, which are taxonomically notoriously difficult.

There is little doubt but that this publication is an important contribution to the
taxonomy of the Agaricales. It will certainly be an essential acquisition for all those
interested, for whatever reason, in the mushrooms of California and for all myco-
logical taxonomists. It would, in my opinion, have been desirable to present the 54
color photographs as colored plates rather than in a microfiche; however, this would
no doubt have drastically increased the cost, which by today’s standards is reason-
able. In any case, I, and many others, now eagerly await Professor Thiers’ publi-
cations on the other families of the Agaricales——KrnnetH WELLS, Department of
Botany, University of California, Davis 95616.

Membership in the California Botanical Society is open to individuals ($12.00 per
year, regular; $8.00 per year, student). Members of the Society receive MapRoNO
free. Institutional subscriptions to MapRONoO are available ($14.00 per year).

Back issues of Madrofio are available at the following rates:

Vol. 1 (1916-1929) and Vol. 2 (1930-1934), each consisting of 17 numbers: $17.00
per volume, $1.00 per issue.

Vol. 3 (1935-1936) through Vol. 20 (1969-1970), each biennial, consisting of 8
numbers: $16.00 per volume, $2.00 per issue.

Vol. 21 (1971-1972) through Vol. 23 (1975-1976), each biennial, consisting of 8
numbers: $24.00 per volume, $3.00 per issue.

Vol. 24 (1977) et seq., one volume per year, each consisting of 4 numbers: $14.00
per volume, $3.50 per issue.

Applications for membership (including dues), orders for subscriptions, requests
for back issues, changes of address, and undelivered copies of MaproNo should be
sent to the California Botanical Society, Inc., Department of Botany, University of
California, Berkeley 94720.

INFORMATION FOR CONTRIBUTORS

Manuscripts submitted for publication in MApRONo should be sent to the Editor.
Membership in the California Botanical Society is requisite for publication in
Maprono.

Manuscripts and accompanying illustrative materials must be submitted in dupli-
cate and should follow the format used in recent issues of MaproNo. Original illus-
trations should not be submitted until paper is accepted for publication. All manu-
scripts MUST BE DOUBLE SPACED THROUGHOUT, including title, text, tables,
captions, lists, literature cited, etc. Footnotes (which should be avoided wherever
possible), captions, and tables must be typed on sheets separate from the text.
Presentation of nomenclatural matter (accepted names, synonyms, typification)
should follow the format used for Rhus integrifolia in MApRoNOo 22:288. 1974. All
measurements should be given in S. I. (metric) units. Where appropriate, scales
should be included on figures rather than in captions as estimates of ratio of repro-
duction such as X %, X 1620, etc. Institutional abbreviations in specimen citations
should follow Holmgren and Keuken’s list (Index herbariorum, Part 1, The herbaria
of the world. Sixth edition. 1974. Regnum Veg. vol. 92). Abbreviations of names of
journals should be those in Botanico-Periodicum-Huntianum (Lawrence, G. H. M.
et al. 1968. Hunt Botanical Library, Pittsburgh). If the correct abbreviation cannot
be determined, the full title of the journal should be used. Titles of books should be

given in full, together with the place and date of publication, names of publisher,

and an indication of the edition, if other than the first.

Short articles such as range extensions and other brief notes are published in con-
densed form under the heading “Notes and News”. Authors of such articles should

follow the format used in recent issues of MaproNo.

Authors are allowed up to 10 pages per year without page charges; charge for

additional pages is $30.00 per page. Subject to approval by the Editors, articles may
_ be published ahead of schedule, as additional pages of an issue, provided the author

!

|

assumes complete costs of publication.

SPECIAL ANNOUNCEMENT
CALIFORNIA BOTANICAL SOCIETY GRADUATE STUDENT MEETINGS

The California Botanical Society Graduate Student Meetings will be held at the —
University of California at Berkeley on November 12-13, 1977. The meetings will —
take the form of a series of short research papers, or reports on work in progress, |

contributed by graduate students in all botanical fields. Members and non-members
of the Society are invited to participate. If you wish to receive further announce-
ments concerning these meetings, please send your name and address to Nancy
Morin, Botany Department, University of California, Berkeley, CA 94720.

VOLUME 24, NUMBER 4 OCTOBER 1977

Contents

HYBRIDIZATION OF FOXTAIL AND BRISTLECONE PINES,
William B,, Critchfield 193

A New GYPSOPHILOUS SPECIES OF PHACELIA (HyYDROPHYLLACEAE)
FROM CoAHuILA, Mexico, NV. Duane Atwood and
Donald J. Pinkava 212

Root:SHoot BromMAss RATIOS IN SHRUBS IN
SOUTHERN CALIFORNIA AND CENTRAL CHILE,
Philip C. Miller and Edward Ng 215

A NEw SPECIES OF IVESIA (ROSACEAE) FROM SOUTHEASTERN
OreEcoN, Barbara J. Ertier and James L. Reveal 224

MISCELLANEOUS CHROMOSOME COUNTS OF WESTERN
AMERICAN PLants—IV, James L. Reveal and Reid Moran 227

Four NEw SPECIES OF CENTAURIUM (GENTIANACEAE) FROM
Mexico, C. Rose Broome 235

SPECIAL ANNOUNCEMENT Cover 4

A WEST AMERICAN JOURNAL OF BOTANY

JU LISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY

MaproNo is published quarterly by the California Botanical Society, Inc., and is
issued from the office of the Society, Herbarium, Life Sciences Building, University
of California, Berkeley. Established 1916. Second-class postage paid at Berkeley.
Return requested.

Editor — BARBARA D. WEBSTER
Department of Agronomy and Range Science
University of California, Davis 95616

Associate Editor — Gravy L. WEBSTER
Department of Botany, University of California, Davis 95616

Board of Editors
Class of:

1977—Wi111AM Louis CuLBerson, Duke University, Durham, North Carolina

Date M. Smita, University of California, Santa Barbara
1978—SHERWIN CarRLQuIsT, Claremont Graduate School

LestiE D. GoTtTLies, University of California, Davis

Dennis R. PARNELL, California State University, Hayward
1979—Puiip W. RuNDEL, University of California, Irvine

ISsABELLE TAVARES, University of California, Berkeley
1980—JameEs R. GriFFin, University of California, Hastings Reservation

FRANK A. Lanc, Southern Oregon College, Ashland
1981—DaAnIEL J. CRAWFORD, University of Wyoming, Laramie

James Hewnricxson, California State University, Los Angeles
1982—Dean W. Tayvtor, University of California, Davis

RICHARD VOGL, California State University, Los Angeles

CALIFORNIA BOTANICAL SOCIETY, INC.

OFFICERS FOR 1977

President: Dr. Winstow R. Briccs, Carnegie Institution of Washington,
Stanford, California 94305

First Vice President: Dr. AtvA Day, Department of Botany, California Academy of
Sciences, Golden Gate Park, San Francisco, California 94118

Second Vice President: Dr. Jop Ku1jt, Department of Biological Sciences,
University of Lethbridge, Lethbridge, Alberta, Canada

Recording Secretary: Dr. CHARLES F. QUIBELL, Department of Biological Sciences,
Sonoma State College, Rohnert Park, California 94928

Corresponding Secretary: Dr. RupoLF ScHmip, Department of Botany,
University of California, Berkeley, California 94720

Treasurer: Dr. G. Douctas BarBeE, California Department of Food and Agriculture,
1220 N Street, Sacramento, California 95814

The Council of the California Botanical Society consists of the officers listed above
plus the immediate Past-President, DENNIs R. PARNELL, Department of Biological —
Sciences, California State University, Hayward, CA 94542; the Editors of Madrono;
and three elected Council members: L. R. HecKArpD, Jepson Herbarium, Department
of Botany, University of California, Berkeley, CA 94720 (1975-1977); James R.
GriFFIN, Hastings Reservation, Star Route 80, Carmel Valley, CA 93924 (1976- —
1978) ; Harry D. Turers, Department of Ecology and Systematic Biology, Califor-
nia State University, San Francisco, CA 94132 (1977-1979).

HYBRIDIZATION OF FOXTAIL AND BRISTLECONE PINES

WILLIAM B. CRITCHFIELD
Pacific Southwest Forest and Range Experiment Station
Forest Service, U.S. Department of Agriculture, Berkeley CA 94701

The pines have been more successful than most of their coniferous
relatives in occupying marginal habitats at the upper and lower edges of
the forest zone in western North America. Among the groups restricted
to such habitats is subsection Balfourianae of Pinus, comprising the fox-
tail and bristlecone pines. These pines characteristically grow on cold
dry sites at high elevations, and in most places have few tree associates.
Perhaps because of their inaccessibility and limited economic impor-
tance, not much was known about them until E. Schulman’s discovery in
the mid—1950’s that some bristlecone pines reach greater ages than other
higher organisms (Ferguson, 1968). Since then, much has been learned
about the Balfourianae, and investigations of natural variation have gen-
erated two taxonomic proposals in the group. Bailey (1970) named the
western populations of bristlecone pine P. longacva, restricting the older
name, P. aristata Engelm., to the eastern populations. Mastrogiuseppe
(1972) proposed the subdivision of foxtail pine (P. balfouriana Grev. &
Balf.) into two subspecies.

The first attempts to cross bristlecone and foxtail pines were made in
1940 by the U.S. Forest Service’s Institute of Forest Genetics (IFG) at
Placerville, California, but most of our exploratory crossing of the Bal-
fourianae pines with each other and with other pines was carried out
between 1963 and 1971. This paper summarizes the results of these
crosses, examines variation in several characteristics of the Balfourianae,
and relates this information to recent investigations of the systematics
and evolution of the group.

THe Taxa oF Balfourianae

Subsection Balfourianae is a morphologically and geographically co-
herent group, with no close affinities to any other group of white pines in
subgenus Strobus (Haploxylon). The needles mostly number five per
fascicle, but this common feature of white pines is accompanied by the
absence of marginal teeth on the needles, cone scales with dorsal umbos
terminating in a mucro or spine, and seeds with long detachable wings—
a combination of characters that readily differentiates the Balfourianae
from other subsections of Strobus.

Among the named taxa in the Balfourianae, foxtail pine is the most
limited in distribution. This Californian endemic grows only in the Klam-
ath Mountains of northwestern California and a portion of the southern

Madrofio, Vol. 24, No. 4, pp. 193-244. October 7, 1977.
193

194 MADRONO [Vol. 24

Sierra Nevada centered on Sequoia National Park. The northern and
southern stands are separated by more than 500 km. They differ in tur-
pentine composition (Haagen—Smit, Wang and Mirov, 1950) and in
morphology, particularly quantitative characteristics of cones and seeds
(Mastrogiuseppe, 1972; Mastrogiuseppe and Mastrogiuseppe, 1975).
Mastrogiuseppe (1972) summarized the evidence supporting their recog-
nition as separate subspecies.

The bristlecone pines are much more widely distributed, ranging from
California to the southern Rocky Mountains. The Colorado River sepa-
rates eastern P. aristata from western P. longaeva. Eastern bristlecone
pine grows in the mountains of Colorado and New Mexico, with a dis-
junct population more than 500 km west in the San Francisco Peaks of
northern Arizona. Western bristlecone pine is widely distributed in the
mountains of Utah, Nevada, and eastern California. At its western limits
in the White and Inyo Mountains of California, it grows only about 25
km from the nearest foxtail pine stands on the east slope of the Sierra
Nevada.

The eastern and western bristlecone pines differ in morphology
(Bailey, 1970) and turpentine composition (Zavarin and Snajberk, 1973;
Zavarin, Snajberk and Bailey, 1976). Bailey noted differences in cone
morphology and color, resin odor, needle retention, and other features,
but his separation of eastern and western bristlecone pines was docu-
mented primarily by characteristics of the needle resin canals. He found
that most needles of western bristlecone and foxtail pines had two exter-
nally visible resin canals and ungrooved needle surfaces. In the needles
of eastern bristlecone pine, the resin canals were located beneath shallow
grooves in the needle surface, and Bailey estimated the number and dis-
tribution of resin canals by counting grooves. About half of the needles
had only one groove. The rest had two or more, but in more than half of
these needles all but one groove terminated below the apical quarter of
the needle.

The resin canals of the eastern trees are also much smaller and closer
to the needle surface. They often burst, exuding resin onto the surface.
The dried resin forms a white fleck, and the eastern and western trees
differ greatly in the incidence of these flecks. Only 5 percent of the
needles Bailey sampled in western stands had flecks. In the east, flecks
were present on 92—94 percent of the needles from Colorado and New
Mexico and 47—63 percent of the needles from Arizona (Bailey, 1970;
Zavarin et al., 1976).

Bailey also observed differences between eastern and western bristle-
cone pines in the shape of the cone base, stoutness and length of the cone
bristles, and color of ripe pollen—cones and maturing seed—cones. Meas-
urements of the length of bristles on cones in the IFG herbarium sub-
stantiate his statement that eastern bristlecone pines often have longer
bristles. Cones from eastern trees averaged about 5 mm in bristle length,

1977 | CRITCHFIELD: PINES 195

and ranged from 3—8 mm. Cones from western trees had bristles averag-
ing about 3 mm long (range 1.5—6 mm).

Polymorphism in the color of maturing reproductive structures is one of
the most distinctive characteristics of western bristlecone pine, differen-
tiating it from all other elements of the Balfourianae. The seed—cones of
most western trees are dark purple during their second season of develop-
ment, and the pollen—cones of the same trees are nearly always deep red
at maturity. A minority of trees have green seed—cones and yellow (some-
times pink-tipped) pollen—-cones. The green—cone trees lack an antho-
cyanin pigment that is present in the others (Mastrogiuseppe, 1976).
This color variant is more common in some western stands than Bailey’s
(1970) estimate of less than 1%. We found that about 20% of the trees
in the White Mountains had green cones (Johnson and Critchfield,
1974), and on Telescope Peak in the Panamint Mountains about 4% had
green cones (Unpub. data, IFG). Eastern bristlecone pine and foxtail
pine are not known to be polymorphic in cone color. Both have purple or
greenish—purple seed—cones and yellow pollen—cones. Thus nearly all
western bristlecone pines differ from the other Balfourianae pines in the
color of either the ripening seed—cones (green) or the mature pollen—
cones (deep red).

Bailey’s classification of the bristlecone pines was supported by the
results of the first detailed investigation of turpentine composition in the
group (Zavarin and Snajberk, 1973). With a single exception in each
region, turpentine of 37 eastern trees was mostly 3-carene (63-92%);
that of 30 western trees was almost entirely a—pinene (95-97% ).

A more complex picture has emerged from further work in which Za-
varin et al. (1976) used larger samples and analyzed needle resin as well
as the turpentine fraction of wood resin. Between 10 and 26% of the
trees sampled in Arizona and the White and Inyo Mountains of Califor-
nia were chemically deviant in turpentine composition. The most remark-
able finding of these authors was the turpentine composition of a Cali-
fornia stand of P. /ongaeva recently discovered on Sentinel Peak in the
Panamint Mountains (Johnson, 1976). All ten trees sampled were higher
in 3—carene content than most eastern bristlecone pines. Needle resin
composition showed three main groups: Arizona, Colorado—New Mexico,
and west of the Colorado River. Arizona trees were closer to Colorado—
New Mexico trees in a quantitative measure of chemical similarity, but
in several constituents they were intermediate or resembled western trees.
Foliage from the single Sentinel Peak tree sampled was most like Colo-
rado—New Mexico trees in resin composition, but in some constituents it
was apparently outside the range of other bristlecone pines. Thus there
are four chemically identifiable groups of bristlecone pines: Colorado—

New Mexico, Arizona, Utah—Nevada, and the highly variable California
stands.

196 MADRONO [Vol. 24

METHODS

Parent trees. Access is a problem in the controlled pollination of foxtail
and western bristlecone pines, and all parent trees were in a few rela-
tively accessible native stands. No arboretum trees of these taxa were
used; none has survived to reproductive age in the warm foothill climate
of Placerville.

The first crosses on western bristlecone pine were made in 1940 and
1948 at Telescope Peak in the Panamint Range (plot location: 36°10’N,
117°05’W, 3330 m). In the same years, crosses were made on foxtail pine
at Onion Valley, on the east slope of the southern Sierra Nevada
(36°46’N, 118°20’W, 2790 m). The fate of the seed harvested from these
early crosses is not fully documented and they are not included in the
summarized data. Two 1940 crosses of P. monticola females and Tele-
scope Peak pollen are included in the data summary (Table 1).

In 1963 and 1971 crosses were made on six and five bristlecone pines
about 1.5 km north of Schulman Grove in the White Mountains (plot
location: 37°24’N, 118°11’W, 3095 m). Five of the trees were used as
females in both years. Pollen was collected from other stands within 1.5
km of Schulman Grove at elevations of 2990-3200 m. All crosses with
foxtail pine as female parent were made in 1965 on four trees at Onion
Valley. Pollen was collected in several seasons within 0.5 km of Onion
Valley at elevations of 2775-2955 m. Pollen from northern foxtail pine
was collected in 1961 and 1964 on North Yolla Bolly Mountain at ele-
vations of 2195-2285 m (40°12’N, 122°59’W). Pollen of eastern bristle-
cone pine was collected in 1970 by D. K. Bailey in three stands west of
Denver, Colorado. The stands were 13—27 km apart, and ranged in eleva-
tion from 2990-3290 m (39°40’—52’N, 105°34’-44’W).

Crosses of western bristlecone and foxtail pines with other species
utilized trees in natural stands and the IFG arboretum. Female parents
of P. bungeana and P. flexilis were old arboretum trees, mostly of un-
known origin. Pinus monticola and P. monophylla females grew in native
stands in the central Sierra Nevada, P. monticola at 2130 m and P.
monophylla on the east side at 1615 m. Arboretum trees supplied all
pollen except P. flexilis, which was collected near Schulman Grove in the
White Mountains.

Techniques and Terminology. Standard breeding and seed—processing
techniques were used (Critchfield, 1963). Pollen was collected during the
season before pollination and deep—frozen, unless otherwise noted. The
data summaries (Tables 1—4) include all crosses for which we have com-
plete information: numbers of strobili pollinated, cones harvested, and
sound and hollow seeds.

An attempt is the pollination during a single season of a female parent
with pollen from a single male parent (Tables 1-4) or with a mixture of
pollen from more than one male parent (Table 1 only). Crossability is

) 1977]

CRITCHFIELD: PINES

197

TABLE 1. Crosses oF Balfourianae WITH OTHER WHITE PINES.

Female parent

Male parent

| monticola longaeva
2) 2 Z(2)/ 0
monticola 79: 43 38:87
76.8/ 94.4 O/ 92.2
flexilis balfouriana
2 (2)/ 2 2 (2)/ 0 ay
flexilis 17: 59 18:0
38.0/ 65.4
<a 3 ae 3°(5)7 3
balfouriana 18: 78 44: 93
0.2/ 42.8 46.4/ 55.4
bungeana balfouriana
| 7 s/, 1 GL) 720
bungeana 8: 100 3s 507
4.0/ 6.0 0/ 0.5
3° (3)7 0 SUK
balfouriana 12 73 47: 85
0/ 13.0 53.6/ 62.3
monophylla balfouriana
2 (4)/ 2 2 (6)/ O
monophylla 25: 44 34: 29
9.4/ 11.6 0/ 8.5
‘nd pollination ee ere ne: Number of tree x tree
jllen frozen for 2 years F. . combinations
‘f pollination Pease tae Number of attempts
! per akeee producing sound seeds
Bpcrsof attempts. ..........cc.c00-Bpcceecceccocecesseeeecceee dee ) (0) On
uber of female strobili....... Pens An te aad hh eA lee On: Oe ee Percent of strobili

¢ number of sound seeds per cone...............|_-0.

}

producing cones

Ae! arta eer eels Mean total number of

seeds per cone

the mean yield of sound seeds per cone from crosses between two taxa,
expressed as a percent of the yield from control crosses within the seed—
parent taxon. All control crosses summarized in Tables 1—4 were made on
the same seed parent in the same season as crosses between taxa.

The following abbreviations are used: northern foxtail pine—NFO,
southern foxtail pine—SFO, western bristlecone pine (P. longaeva)—
WBR, and eastern bristlecone pine (P. aristata)—-EBR.

198 MADRONO [Vol. 24

Seedling Observations. Seedlings of the Balfourianae are slow-growing,
and do not regularly produce secondary leaves (needles in fascicles) until
the third growing season. Needles were sampled at the end of that season.
Comparisons of needle characteristics were based on five fascicles from
each of five seedlings from each of these groups: WBR (White Moun-
tains); EBR (Colorado); EBR (Arizona); WBRxEBR, and SFO
(Onion Valley). The seedlings were as diverse as possible in origin and
parentage.

Counts of grooves on the needle surface did not provide reliable data
on the number and distribution of resin canals in the needles of EBR
seedlings. All needles were hand-sectioned at one-fourth of the distance
from the tip to the base of the needle, and about half were also sectioned
at one-half and three-fourths of this distance. All resin canals were
external, and with a few exceptions were adjacent to the abaxial face of
the needle. Resin canal diameters were measured perpendicular to the
abaxial surface. Measurements were made between the outer walls of the
epithelial cells, since the cells in this layer were highly variable in degree
of flattening.

Other observations. Several unnoticed or poorly documented character-
istics of the Balfourianae pines were observed in California stands of fox-

tail and western bristlecone pines or in herbarium collections and stored —
seeds at the IFG. The materials available limited comparisons to major |

groups within the Balfourianae. Cone specific gravity was determined

from volume, measured by water displacement, and oven—dry weight. |

Measurements were made of single cones from ten trees in each of the |
following groups: NFO (trees scattered throughout range) ; SFO (all but |

two trees in Onion Valley area); WBR (three trees in Utah, one in |

Nevada, two in Panamint Range, and four in White and Inyo Moun-

tains); EBR (Arizona); and EBR (Colorado/New Mexico: two in New |
Mexico, eight in four Colorado stands). Seed—wing length was measured |
on four seeds from each of ten trees or mass collections in these groups: ©

SFO and NFO combined (two SFO, eight NFO); WBR (three Utah,

two Nevada, one Panamint Range, four White and Inyo Mountains);
and EBR (seven Arizona, two Colorado, one New Mexico). De—winged,
air-dried seeds in cold storage were weighed to the nearest milligram. |

Five seeds were measured from each of eight trees or collections in these
groups: SFO and NFO combined (five SFO, three NFO); WBR (two

Utah, six White Mountains): and EBR (Arizona); and four trees or |
collections of EBR (Colorado/New Mexico: three Colorado, one New

Mexico). Tree means were analyzed for seed weight and wing length.
Stratification requirements were determined for seed samples of EBR

(Kenosha Pass, Colorado); WBR (White Mountains) ; and SFO (Onion.

Valley). Samples of 100 seeds received one of three treatments (stratifi- |
cation for 45 or 90 days, or no stratification). All seeds were placed in

1977 | CRITCHFIELD: PINES 199

germinators at the same time, and germination was observed for four
weeks.
RESULTS

Reproductive Phenology and Capacity. Bristlecone and foxtail pines
flower later in the season than any other North American pines. Flower-
ing (pollen shedding and/or maximum opening of ovulate strobili) was
observed in natural stands of these pines during 12 seasons between 1940
and 1974. All observations fall within a narrow range of calendar dates:
a three-week period beginning in mid—July and peaking during the last
week in July. Western bristlecone pine flowers later than P. flexilis in the
White Mountains, and foxtail pine is later than P. flexilis or P. albicaulis
at Onion Valley. In Tilden Park, near Berkeley, California, planted west-
ern bristlecone pines were observed in two seasons to flower in mid—June,
about six weeks earlier than in their native habitat.

The Balfourianae pines flower at almost the same time. In four years
of observations on successive days in stands of foxtail pine at Onion Val-
ley and bristlecone pine in the White Mountains, we found no differences
in flowering time, nor are there any indications of differences between
eastern and western bristlecone pines. Pollen was collected in three Colo-
rado stands between July 25 and August 2, 1970. In an Arizona stand at
2925 m, pollen was shed on July 22-27, 1969 (Schubert and Rietveld,
1970), and on July 31, 1975 (Rietveld, pers. comm., 1975). More gener-
alized dates for pollen shedding in Arizona at 3050 m, based on three
seasons of observations, were July 20 to August 20 (Pearson, 1931).

Northern foxtail pine has been observed to flower later than southern
stands. In 1963 and 1964, observations were made 3—4 days apart at
Onion Valley and North Yolla Bolly Mountain. In 1963 the North Yolla
Bolly stand was estimated to be at least a week behind the Onion Valley
stand, but in 1964 flowering was nearly simultaneous. Western bristle-
cone pine stands at different elevations have also been observed to peak
a few days apart. With these exceptions, there appear to be no substantial
differences in flowering time in the Balfourianae.

These pines are also uniform in the timing of cone—opening. Cones of
foxtail pine at Onion Valley and western bristlecone pine in the White
Mountains open in a two—week period starting about September 20. Ari-
zona cones mature at about the same time. Schubert and Rietveld (1970)
found that seeds of Arizona trees matured (were germinable) between
September 24 and October 2, 1969, and the cones opened between Sep-
tember 27 and October 10. They concluded that cone maturation is not
complete until just before the cones open, and our experience with west-
ern bristlecone pines confirms this. Some control—pollinated cones of
White Mountains female parents, still closed when they were collected
at the end of September, failed to open and had to be discarded.

In crosses where genetic barriers were not encountered, controlled pol-

200 MADRONO [Vol. 24

lination of western bristlecone pine produced more sound seeds per cone
than wind pollination. Wind—pollinated cones from the bristlecone female
parents yielded a mean of 38 sound seeds per cone (range 25—65)—close
to the 36 sound seed per open—pollinated cone reported for Arizona trees
(Schubert and Rietveld, 1970). Cones from controlled pollination of the
White Mountains trees averaged about 30% more sound seed than wind-
pollinated cones, and the most successful controlled crosses on individual
parents averaged 70 sound seed per cone. A potential source of bias in
these comparisons is the possibility that seed extraction was more com-
plete for control—pollinated cones.

The reproductive capacity of bristlecone and foxtail cones can be con-
servatively estimated from these seed yields, together with available data
on numbers of cone scales. Bailey (1970) estimated that western bristle-
cone pine averages 117 scales per cone. Since the most successful con-
trolled crosses produced 70 sound seeds, a minimum of 30% of the scales
can bear two sound seeds. Foxtail pine, averaging only 81 scales per cone
(Bailey, 1970; Mastrogiuseppe, 1972), had a mean of 58 sound seed per
cone in the most productive controlled crosses. Thus at least 35 or 36%
of its cone scales can bear two sound seeds. Wind pollination utilizes only
a fraction of this potential reproductive capacity—54% in our bristle-
cone pine female parents. Comparable data are not available for the fox-
tail female parents, but a few estimates of sound seed per cone in bulk
collections suggest that about 40% of the cone’s reproductive capacity is
exploited by wind pollination.

Crossing the Balfourianae with other Pines. Like most other subsec-
tions of Pinus (Critchfield, 1975), the Balfourianae appear to be isolated
from the rest of the genus by genetic barriers. With one doubtful excep-
tion, our limited attempts to cross them with members of other white pine
groups failed (Table 1). Representatives of other groups include P. mon-
ticola and P. flexilis (subsection Strobi), P. monophylla (Cembroides),
and P. bungeana (Gerardianae).

In two instances there were indications of strong barriers acting early |
in the reproductive process. Pinus balfouriana x P. bungeana and the
species—reciprocal cross yielded very few hollow seeds in either direction
(Table 1). Since seed coats form early in the second season of cone devel-
opment, at about the time of fertilization, this drastic reduction in hollow
seeds indicates a developmental breakdown before fertilization. Another
suggestion of early barriers is the abortion of all P. flexilis strobili polli-
nated with P. balfouriana pollen, but here a genetic interpretation is con-
founded by pollen age (Table 1).

Of the crosses summarized in Table 1, only P. balfouriana x P. flexilis —
produced any sound seed. Two P. balfouriana parents crossed with the |
same P. flexilis pollen parent produced a total of three sound seeds with
normal embryos (determined by X-ray radiographs). Only one of the —

1977 | CRITCHFIELD: PINES 201

three seeds germinated. In its germination time and three-year height
the seedling was similar to sibling P. balfouriana seedlings in adjacent
nursery rows, but before its non—hybrid identity could be firmly estab-
lished the seedling was accidentally destroyed.

Crossing Northern and Southern Foxtail Pines. Northern and southern
stands of foxtail pine were almost fully crossable (Table 2). Onion Val-
ley females pollinated with North Yolla Bolly pollen produced 84% as
much sound seed as they did in combination with Onion Valley males.
The reduction in seed yield was not statistically significant (four pairs,
0.20 > p > 0.10) but it was consistent among female parents. Individual
trees averaged 5—27% less sound seed in hybrid combinations than in
control crosses.

The hybrid seedlings did not differ appreciably in size or other features
from their non-hybrid siblings in the nursery, and it is doubtful whether
they will be morphologically identifiable as hybrids until they reach
reproductive maturity.

Crossing Foxtail and Western Bristlecone Pines. White Mountains
bristlecone pine and Onion Valley foxtail pine were fully compatible in
both directions (Table 3). Foxtail pine females produced nearly identical

TABLE 2. CROSSES BETWEEN SOUTHERN AND NORTHERN FOXTAIL PINES.
(See Table 1 for legend.)

Male Parent
Female parent balfouriana balfouriana
(northern) (southern)
balfouriana Ne) CT!
99: 84 63: 87
(southern) 43.4/ 55.4 51.6/ 60.6

TABLE 3. CROSSES BETWEEN FOXTAIL AND WESTERN BRISTLECONE PINES.
(See Table 1 for legend.)

Male parent

Female parent

balfouriana balfouriana longaeva
(northern) (southern)
11) / b 1/ 23 (2397723 13 (16 )/ 162-27,
longaeva 72.43 191: 54 96: 65
bray ol 52.4/ 68.5 36.9/ 47.3
: VOGT S33 yO
"a 63: 87 31: 90
ay 51.6/ 60.6 52.9/ 65.4

1/_ pollen frozen for 2 years
2/ fresh pollen used in 1971

202 MADRONO [Vol. 24

amounts of seed in combination with bristlecone pines and with other
foxtail pines (52.9 and 51.6 sound seeds per cone). And bristlecone fe-
males produced more seed when foxtail pine was the pollen parent than
they did in control crosses with other bristlecone pines (52.4 and 36.9
sound seeds per cone). Single-tree combinations ranged from 40-66
sound seeds per cone in SFO x WBR crosses, and 4-116 in WBR x SFO
crosses.

Although the reduced seed yield of crosses within bristlecone pine was
not significant (eight comparisons, 0.10 > p > 0.05), it was consistent
in both breeding seasons. Within—bristlecone crosses averaged 40.6 and
30.9 sound seeds per cone in 1963 and 1971, compared to 49.1 and 57.8
seeds from crosses with foxtail pine. The 1971 reduction may have been
due to the use of slightly abnormal bristlecone pine pollen, which formed
enlarged pollen tubes when it was germinated in the laboratory, but the
1963 difference is unaccounted for. This anomalous reduction cannot be
attributed to inbreeding between closely related neighbors: in both sea-
sons the bristlecone pollen was collected in stands 1-2 km from the
female parents.

A single cross between northern foxtail pine and western bristlecone
pine yielded fewer sound seeds (5.3 per cone) than other crosses on the
same bristlecone female (x foxtail pines: 34.8 seeds; x other bristlecone
pines: 25.8). Although the foxtail pollen had been frozen for two years
and its viability may have been slightly reduced, this cross provides the
only suggestion in our data that the crossing behavior of northern and
southern foxtail pines may differ in combination with western bristlecone
pine.

By the end of three growing seasons most of the hybrids could be dis-
tinguished from their western bristlecone siblings. At this age most
bristlecone pines (both western and eastern) resembled cushion plants,
lacking an emergent leader. If a leader was present it usually lacked a
definite terminal bud, and the stem was mostly concealed by branches,
fascicles, and closely appressed primary leaves. The slightly taller hybrids
and much taller foxtail pines of the same age nearly all had emergent
leaders, and much of the stem was visible. This was due partly to fewer
primary leaves on the third—season stems and partly to slightly longer
internodes—1.2—1.5 mm compared to 1.0 mm for bristlecone seedlings.
Many hybrids and nearly all foxtail pines had well-developed terminal
buds at this stage.

Crossing Eastern and Western Bristlecone Pines. Unlike the combina-
tions described above, crosses between California and Colorado bristle-
cone pines were relatively unsuccessful. Although 19 or 20 combinations
produced filled seeds, the mean of all 20 crosses was only 6.1 filled seeds
per cone, and the maximum for a single cross was 18.7 seeds. The filled
seeds were routinely X—rayed before most of them were planted in the

1977 | CRITCHFIELD: PINES 203

nursery. The rest were later germinated in petri dishes. Germination was
low in the nursery and only slightly higher in the laboratory. A total of
69 seeds from 11 crosses germinated. The mean number of germinated
seeds per cone for all hybrid crosses was 1.6, compared to 19.8 for non—
hybrid combinations (Table 4)—a crossability of 8%. This could be an
overestimate because of the possibly abnormal pollen used in the non—
hybrid crosses (see above); 4-6% may be a more realistic estimate of
| crossability.

X-ray radiographs of the seeds provided an explanation of the low
germination. Most “filled” seeds contained fully developed or slightly
_ shrunken female gametophytes, with well-defined embryo cavities ex-
tending almost the full length of the gametophyte. Nearly one-fifth of
the filled seeds had empty cavities, another one-fifth contained embryo-—
like objects too small to identify with certainty, and the others contained
identifiable embryos. Most of these embryos were smaller than the em-
bryos of germinable seeds from control crosses, which ranged in size from
about two-thirds to the full length of the embryo cavity. Although the
_ hybrid seeds were not handled individually, a few that germinated must
_ have had embryos as small as 35-40% of the length of the cavity. But
| the great majority of germinable seeds from hybrid combinations had
embryos from 45-65% of the cavity length. Embryos of this size showed
about 80% germination.

Nine of the 20 hybrid combinations failed to produce any germinable
seeds. Ail but one produced filled seeds, but the embryos were mostly
vestigial, with a few ranging in length up to a third of the embryo cavity.
_ Only two of the nine failed crosses (both on the same female parent)
_ produced ungerminable embryos in the 45-65% size range.
| Third—year hybrid seedlings resembled unrelated Colorado and Ari-
_ zona seedlings more closely than their western bristlecone siblings, al-
though all of the seedlings were similar in size and appearance. The

|
|

TABLE 4. CROSSES BETWEEN WE STERN AND EASTERN BRISTLECONE PINES.

i : s Sees

Male parent
Female parent ae ee aera

cues longaeva
(20) / 11 SUE Gay) ye
| longaeva ae 37 14: 79

1.6/ 49.6 19:3/ 35.1

Rerear Number of tree x trec

re combinations

5 At oe ee Number of attempts pro-
G ducing germinable seed

Hrnbe of female Sec anali Se eee eee ee eee eee en ee PS Fae Oe eee eee Percent of strobili

Jan number of germinable seeds per cone................J-. 0.0 / 0. 0 producing cones

Se Mean total number of
seeds per cone

204 MADRONO [Vol. 24

needles of the hybrids were more like those of eastern seedlings in having
grooves overlying the resin canals and in the small size of the canals. The
needles of western seedlings, like those of older western trees, lacked
erooves. The mean diameter of resin canals in western, eastern, and
hybrid seedlings was 0.138, 0.085, and 0.105 mm.

The expression of other needle characteristics was quite different in the
seedlings than in older trees, as it often is in other pines. Two of the
needle characteristics that Bailey (1970) found most useful in distin-
guishing P. aristata and P. longaeva—resin flecks and number of resin
canals—were much more similar in seedlings than in the older trees he
observed in natural stands.

Seedlings from (a) Colorado, (b) Arizona, and (c) the White Moun-
tains had resin flecks on (a) 61, (b) 58, and (c) 27% of their needles.
Corresponding percentages in the older trees that Bailey sampled were
(a) 92-94, (b) 47-62, and (c) 3. Only 9% of the hybrid needles had
flecks—fewer than either parental group of seedlings.

Eastern and western bristlecone pine seedlings were also more similar
in resin canal number than the older trees that Bailey sampled in natural
stands. Only 3% of the needles of eastern (and hybrid) seedlings had
single canals extending the length of the needle, compared to 49% in
older trees. Arizona and Colorado seedlings had single resin canals in the
apical quarter of only 20 and 8‘% of their needles, compared to 84 and
87 in older trees (Bailey, 1970).

The presence of three or four resin canals per needle was fairly com-
mon in eastern bristlecone seedlings and occasional in hybrids, but none
of the needles of western bristlecone or foxtail pine seedlings had more
than two canals. Only 3 and 5% of Colorado and Arizona seedling needles
had three canals near the tip, but near the base the frequency of three or
more canals was 30 and 48%. The mean number of canals in Colorado
and Arizona seedlings increased from 2.0 and 1.8 per needle near the tip
to 2.3 and 2.6 near the base. Four resin canals—the highest number ob-
served—were present in 15% of the needles of eastern seedlings. Where
three or four canals were present, they were crowded together and often
shared grooves. In seedlings, at least, the number of resin canals would
be underestimated by counts of the grooves.

Variation in Mor phological and Other Characteristics. Closer attention
to the Balfourianae pines in recent years has turned up an array of differ-
ences that merit more detailed study:

Tree form—Most old foxtail pines have single, erect stems even under
severe timberline conditions (Arno, 1966), although exceptions have
been noted on dry sites (Bailey, 1970). Old bristlecone pines in the desert
ranges of California have what Arno described as an ‘“‘ungainly, weedy”
form, usually twisted and multi-stemmed. To what extent this multi—
stemmed form is characteristic of old bristlecone pines farther east is
uncertain, although it is present in Colorado (Bailey, 1970; Fig. 1).

1977 | CRITCHFIELD: PINES 205

Bark—Young trees of eastern bristlecone pine differ from the other
Balfourianae pines in having the smooth, blistered cortical bark and
delayed onset of periderm formation that is characteristic of the firs and
many white pines (Zavarin and Snajberk, 1973; Zavarin et al., 1976).
Old foxtail pines in southern stands have thick, reddish-brown bark in
squarish plates, a combination that has not been observed in other Bal-
fourianae pines (Bailey, 1970; Mastrogiuseppe, 1972).

Foliage—A vegetative difference noted by LeRoy C. Johnson (pers.
comm., 1974) distinguishes living trees, including seedlings. The needles
on a foxtail pine twig are painfully sharp to the touch; those of eastern
and western bristlecone pines are not. This difference, difficult to quan-
tify but useful in identification, appears to be due partly to sharper points
but primarily to the greater stiffness of foxtail pine needles.

Cones—E astern and western bristlecone pine cones project abruptly
from the branch, and their straight axes form right or acute angles with
the distal part of the branch. The cones of foxtail pine are pendent to
varying degrees. This visually striking difference in orientation is due
primarily to a difference in peduncle length. The peduncles of mature
bristlecone pine cones in the [FG herbarium were usually flush with the
basal cone scales, but occasionally extended beyond them a maximum
of 4-5 mm. Intact peduncles of foxtail pine cones in the herbarium were
7-16 mm long, and several trees growing at Onion Valley had cone pe-
duncles 14-16 mm long. An associated feature is the curved axis of
21-23% of foxtail pine cones (Mastrogiuseppe, 1972).

The cones of foxtail pine have somewhat “fleshy” apophyses (Bailey,
1970; Mastrogiuseppe, 1972), and are softer and more fragile than cones
of the bristlecone pines. This difference can be expressed quantitatively in
terms of cone specific gravity. Northern and southern foxtail pine cones
did not differ significantly in specific gravity, nor did the western, Colo-
rado—-New Mexico, and Arizona samples of bristlecone pine. But the
combined data showed a significant difference (0.05 > p > 0.01) be-
tween the less dense cones of foxtail pine (mean specific gravity 0.45)
and the cones of the bristlecone pines (mean 0.49).

Seeds—Our data confirm Uyeki’s (1927) observation that foxtail pine
seeds have longer wings than seeds of bristlecone pine. His measurements
of a few seeds of unspecified origins showed ranges of 18-20 mm and
8-11 mm for the two pines. In our samples, wing length was almost iden-
tical in eastern and western bristlecone pines: the means were 10.7 and
10.0 mm, and ranges were 7-15 mm in both samples. Seeds of foxtail pine
have much longer wings (mean 17.2 mm), and the differences between it
and the bristlecone pines were highly significant (p = < 0.01). Foxtail
pine’s range in wing length (13-24 mm) just overlapped that of the
bristlecone pines. In disagreement with these data is Mastrogiuseppe’s
(1972) observation that southern foxtail pine has rather short seed
wings. His large samples of seeds from single localities of northern (Lake

206 MADRONO [Vol. 24

Mountain) and southern (Timber Gap) foxtail pines had mean wing
lengths of 17.4 and 11.5 mm, a highly significant difference. Timber Gap
may be poorly representative of southern stands, however. The two
southern foxtail pines in our sample, both collected by Mastrogiuseppe
(XIV-—7, Onion Valley, and XI-5, Silliman Crest), had average (17.5
mm) or long (23.3 mm) seed wings—the latter the longest wings of any
tree in the sample.

The Balfourianae pines were much more variable in seed weight. Fox-
tail and Arizona bristlecone pines had the heaviest seeds, with means and
ranges of 25.4 mg (13-39) and 26.4 mg (18-38). The small sample of
Colorado—New Mexico trees had somewhat lighter seeds (mean 20.8 mg,
range 11-25), the difference between it and the Arizona sample ap-
proaching statistical significance (0.10 > p > 0.05). Western bristlecone
pine seeds were by far the lightest—less than half the weight of the oth-
ers (mean 8.8 mg, range 6-15). The differences between this and the
other samples were highly significant (p = < 0.01).

A difference in seed color has been noted by Zavarin et al. (1976),
eastern bristlecone pines having darker seed coats than western trees.

Stratification requirement—Bristlecone pine seeds germinate promptly
without stratification, but the germination of foxtail pine seed—like that
of most other white pines—is slow and incomplete without this pretreat-
ment (Forest Service, 1974). Unstratified seed of Arizona bristlecone
pine showed 75°0 germination within eight days (Schubert and Rietveld,
1970); untreated Colorado seed germinated 75-80% in 4-10 days
(Reid, 1972); and untreated White Mountains seed showed 90% germi-
nation within six days (Wright, 1963). In foxtail pine, Mastrogiuseppe
(1972) obtained only 29-63 and 50-55% germination in three-month
tests of unstratified seed from northern and southern stands. In germina-
tion tests at the IFG, 45 or 90 days of stratification reduced mean germi-
nation time of Colorado and White Mountains seed by 2-3 days (un-
stratified: 11.3 and 7.9 days; stratified: 8.3-8.4 and 5.5—5.8 days), but
did not increase the amount of germination. The same pretreatments
increased the germination of Onion Valley foxtail pine seed from 41 to
100% and reduced mean germination time by 8—9 days (unstratified:
11.9 days; stratified: 2.6-3.5 days).

DISCUSSION

The ability of pines to hybridize is generally restricted to taxa that are
considered to be related on other grounds. Within groups linked by the
ability to hybridize, however, the degree of crossability is sometimes
highly discordant with other evidence of relationship. Among the pines,
P. muricata (bishop pine) provides the closest parallel to the Balfouri-
ana? in this respect. The three races of this coastal Californian species—
northern, central, and southern—exhibit little correspondence between
crossability and other indicators of relationship (Critchfield, 1967). The

1977 | CRITCHFIELD: PINES 207

northern race differs morphologically from the other two, and all three
are chemically distinct. These distinctions are nearly absolute except in
the narrow zone where the northern and central races meet. Crossability
is complete between northern and central races (the only two in contact),
low between central and southern races, and close to zero between north-
ern and southern races.

Among the Balfourianae, the crossing behavior of northern and south-
ern foxtail pines is an exception. Their crossability of 84% is fairly typi-
cal of segments of a species that have diverged sufficiently to warrant
taxonomic recognition. In P. ponderosa, the western (var. ponderosa)
and eastern (var. scopulorum) races had an average crossability of 52%
in small-scale reciprocal tests (Krugman, 1970). The coastal and Sierra
Nevada races of P. contorta, usually given subspecific or varietal status
but still occasionally considered two species (P. contorta, P. murrayana),
has a crossability of 93% in an extensive series of reciprocal crosses (un-
published data, IFG).

No macrofossils of foxtail pine have been described from the Tertiary,
although Axelrod (1976) noted that a fossil pine in the Thunder Moun-
tain, Idaho, flora (Eocene age) resembles this species. Pollen identified
as P. balfouriana was present in several pollen floras in and near the
southern Sierra Nevada during the first recorded Pleistocene glaciation
(Axelrod and Ting, 1961), but one problem with this identification is the
assumption that the dimensions of the pollen have remained constant
since then. Foxtail pine cones have recently been found near Clear Lake,
California, in late Pleistocene (probably Ilinoian) deposits (J. Wolfe,
pers. comm., 1977). At this site, more than 100 km south of and more
than 1000 m below the present northern distribution of foxtail pine, the
abundance of cones suggests that this species was a major component of
a high—elevation mixed—conifer forest.

Mirov (1967), Bailey (1970), and Mastrogiuseppe (1972) have
speculated—apparently on the basis of geological history—that the
northern and southern stands of foxtail pine have been separated since
the end of the Tertiary or the early Pleistocene. If so, the accumulation
of genes influencing crossing ability has been very slow in the two or
three million years these groups have been isolated from each other.
However, the recent Clear Lake find raises the possibility of contact be-
tween them as recently as one of the last major glacial episodes of the
Pleistocene.

The most striking instance of discordance between genetic and other
evidence of relationships in the Balfourianae is the complete crossability
of western bristlecone and southern foxtail pines. This combination is
also by far the best supported by crossing data; it is the only combination
made in both directions and in more than one season. This level of cross-
ability is remarkably high for pine species, although it is approached by
the 69-85% crossability of the California closed—cone pines P. attenuata

208 MADRONO [Vol. 24

and P. radiata (Critchfield, 1967) and a few other less fully investigated
combinations.

Although genetic barriers to interbreeding are nonexistent in the White
Mountains and Onion Valley stands, the morphological and other differ-
ences between foxtail and western bristlecone pines are of the same mag-
nitude as those distinguishing most other closely related pine species.
Two cone characteristics have traditionally been emphasized: foxtail
cones have minute mucros but lack conspicuous bristles, and have fewer
scales—about 80-81, compared to 117 for western bristlecone pine. But
they also differ in peduncle length and the complex of associated charac-
ters (cone orientation, curvature), cone density, seed weight, seed—wing
length, stratification requirement, foliage stiffness, and probably tree
form.

The morphological distinctions between eastern and western bristle-
cone pines are less conspicuous, with the notable exception of the resin
flecks on the needles. The principal differences between the two are in the
complex of characters associated with the needle resin canals and in resin
composition, but they also differ in cone color, bristle length, seed weight
and color, bark, and other features noted by Bailey (1970). These taxa
are also far less crossable than foxtail and western bristlecone pines, with
a crossability of only a few percent, estimated from crosses made in one
direction in a single season. A smaller number of crosses made by Bailey
in the opposite direction in the same season were even less successful
(Zavarin et al., 1976).

The reproductive barriers between the two bristlecone pines resemble
those of many other white pine combinations in acting mainly or entirely
after fertilization (Kriebel, 1975). The seed coats, which form at about
the time of fertilization, showed no reduction in numbers in western x
eastern combinations (Table 4). Post-fertilization barriers are also indi-
cated by the many seeds with fully developed female gametophytes and
embryo cavities, but lacking embryos or with very small embryos. Em-
bryo cavities form after fertilization (Sarvas, 1962), and the female
gametophyte usually degenerates rapidly after the death of the last devel-
oping embryo (Sarvas, 1962; Plym Forshell, 1974). Mature or nearly
mature seeds with empty embryo cavities have been reported in only a
few instances: in P. sylvestris by Ehrenberg et al. (1955), Sarvas
(1962), and Plym Forshell (1974); and in the cross P. jeffreyvt x P.
coultert by Krugman (1970). These authors interpreted such seeds as
indicators of breakdown at a relatively late stage of embryo development.
Another unusual feature of the cross between western and eastern bristle-
cone pines was the germination of seeds with embryos less than half the
length of the embryo cavity. This has also been reported in P. sylvestris
(Ehrenberg et al., 1955).

The Balfourianae group is an old lineage in western North America.
Macrofossils resembling contemporary bristlecone pines have been found

1977 | CRITCHFIELD: PINES 209

in Eocene and Oligocene deposits in Nevada, New Mexico, Colorado, and
Montana. Most of these fossils have been described under the name P.
crossii Knowlton.

The oldest P. crossi fossils were part of the Eocene Copper Basin flora
of northern Nevada (Axelrod, 1966), which has an estimated age of
about 40 million years. Copper Basin is 120 km north of the present
northern limits of western bristlecone pine. The five—needled fascicle
illustrated by Axelrod (Plate 6, Fig. 6) has needles only 13 mm long—
much shorter than most needles of contemporary Balfourianae. Needles
this short were not observed on specimens of eastern bristlecone pine and
were very rare on foxtail pine, but bristlecone pines in the White Moun-
tains occasionally produce fascicles of this size on slow—growing branches.
I have examined the Copper Basin fascicle (University of California
Museum of Paleontology hypotype 8873), and one of the needles shows
two linear impressions suggestive of resin canals extending the length of
the needle.

According to Axelrod (1976), a mid—Oligocene flora at Hillsboro, New
Mexico, was dominated by a pine that closely resembles contemporary
P. aristata in its cones and fascicles. The nearby Hermosa flora, also
mid—Oligocene, consists mostly of a “‘five-needled pine allied (distantly )
to P. aristata.”

A pine resembling P. aristata is also abundantly represented in a fossil
flora deposited near Creede, Colorado, near the end of the Oligocene,
about 27 million years ago. This pine (P. crossii) is represented by a
conelet, seed wing, and many needle impressions (Fig. 34-37, Bailey,
1970). Bailey and earlier authors consider it nearly identical with P.
aristata, which grows nearby today.

A single fascicle that can probably be assigned to P. crossi was found
in late Oligocene deposits in the Ruby Basin of southwestern Montana
(Becker, 1961). It consists of three needles and the base of a fourth.
These needles are also shorter (16-17 mm) than those of contemporary
Balfourianae pines.

Impressed by the characteristics shared by foxtail and western bristle-
cone pines and by the similarity of the Creede fossils to eastern bristle-
cone pine, Bailey (1970) proposed a polyphyletic origin for the bristle-
cone pines. According to this scheme, two lineages arose in northwestern
North America at the start of the Tertiary or earlier, and migrated south
into the Rocky Mountain and Pacific regions. The eastern line—substan-
tially unchanged since at least the late Oligocene (Creede flora)—is
modern P. aristata, and the Arizona population is a Pleistocene offshoot.
The western line resembled P. balfouriana. It gave rise to P. longaeva at
the end of the Tertiary and beginning of the Pleistocene as the major
uplift of the Sierra Nevada produced increasing aridity in the Great
Basin.

A principal reason for rejecting Bailey’s polyphyletic hypothesis is the

210 MADRONO [Vol. 24

burden it places on convergent evolution, which must account for all of
the diverse ways in which eastern and western bristlecone pines resemble
each other and differ from foxtail pine. These include, in addition to
cone-scale number and bristles, several characteristics noted above—
peduncle; orientation and density of the cone; seed wings; absence of a
stratification requirement; foliage stiffness; and perhaps tree form. Nor
does recent chemical evidence (Zavarin et al., 1976)—-particularly the
chemical intermediacy of Arizona stands and the similarity of Sentinel
Peak to Colorado-New Mexico trees—support the view that western
bristlecone pine evolved from a foxtail—pine—like ancestor.

The tendency of western bristlecone pine to resemble foxtail pine in
some respects (needle resin canals, chemistry) and eastern bristlecone
pine in others can be explained more simply in other ways. Western
bristlecone pine could be the product of ancient hybridization between
eastern (P. aristata) and western (P. balfouriana) ancestral lines. One
difficulty with this hypothesis is that western bristlecone pine is not inter-
mediate in most respects (bristle length is an exception). Instead, it
closely resembles either eastern bristlecone pine or foxtail pine. And in a
few characteristics it resembles neither (cone—color polymorphism, seed
size, great longevity). A more critical objection is the asymmetrical cross-
ing behavior of western bristlecone pine in combination with its presumed
parents. The hypothesis of a hybrid origin is difficult to reconcile with the
the complete absence of crossing barriers on the one hand (P. balfourt-
ana) and the presence of strong barriers to interbreeding on the other
(P. aristata).

An alternative—and more acceptable—hypothesis is that all of the
elements of the Balfourianae are segregates of a single ancestral line most
closely resembling western bristlecone pine (Eocene Copper Basin fossil).
By the middle or late Oligocene a Rocky Mountain lineage split off and
developed some of the distinctive attributes of contemporary eastern
bristlecone pine (Hillsboro and Creede floras). Arizona bristlecone pine
could have arisen as a very early offshoot of this lineage, or it could be
(as Zavarin et al., 1976, suggested) the product of later contact between
eastern and western bristlecone pines. It is uncertain when foxtail pine
originated from the western bristlecone pine lineage, although separation
must have been complete before the late Pleistocene (Clear Lake fossils).
Foxtail pine has undergone pervasive changes in morphological and other
characteristics, but divergence has apparently been too recent for the
accumulation of genes influencing its ability to cross with the closest
modern equivalent of its ancestor.

The crossing data summarized in this paper are incomplete in many
respects. The exploratory hybridization of the Balfourianae should be
extended to include Arizona P. aristata, P. longaeva of the Panamint
Range, the untried combination of P. balfouriana and P. aristata, and
more combinations of P. dalfouriana and P. longaeva. Crosses between

I
|

1977 | CRITCHFIELD: PINES 2h

P. aristata and P. longaeva also need to be repeated and extended, but
our preliminary finding of very low crossability between these taxa pro-
vides tentative support for Bailey’s (1970) proposal to recognize them
as species.

LITERATURE CITED

Arno, S. F. 1966. Interpreting the timberline. M.F. Thesis: Univ. Montana, Mis-
soula, 206 p.

AxELrop, D. I. 1966. The Eocene Copper Basin flora of northeastern Nevada. Univ.
Calif. Publ. Geol. Sci. 59: 83 p.

. 1976. History of the coniferous forests, California and Nevada. Univ.
Calif. Publ. Bot. 70: 62 p., illus.

aa , and W. S. Tinc. 1961. Early Pleistocene floras from the Chagoopa sur-
face, southern Sierra Nevada. Univ. Calif. Publ. Geol. Sci. 39:119-194.

Baitey, D. K. 1970. Phytogeography and taxonomy of Pinus subsection Balfourt-
anae. Ann. Missouri Bot. Gard. 57:210-249.

Becker, H. F. 1961. Oligocene plants from the upper Ruby River Basin, south-
western Montana. Geol. Soc. Amer. Mem. 82: 127 p.

CrITCHFIELD, W. B. 1963. Hybridization of the southern pines in California. South.
Forest Tree Improv. Comm. Publ. 22:40-48.

1967. Crossability and relationships of the closed-cone pines. Silvae
Genet. 16:89-97.

. 1975. Interspecific hybridization in Pinus: a summary review. Proc. 14th
Meeting, Canad. Tree Improv. Assoc., Part 2:99-105.

EHRENBERG, C., A. GustArsson, C. PryM ForsHert, and M. Srmaxk. 1955. Seed
quality and the principles of forest genetics. Hereditas 41:291-366.

Fercuson, C. W. 1968. Bristlecone pine: science and esthetics. Science 159:839-846.

Forest SERVICE. 1974. Seeds of woody plants in the United States. USDA Agric.
Handb. 450: 883 p.

Haacen-Smit, A. J., T-H. Wanc, and N. T. Mrrov. 1950. Composition of gum tur-
pentines of Pinus aristata, P. balfouriana, P. flexilis, and P. parviflora. J. Amer.
Pharm. Assoc. 39:254-259,

Jounson, L. C. 1976. Range extensions of three conifers and a dwarf mistletoe
in the Panamint Mountains, Death Valley National Monument. Madronfio
23:402-403.

—, and W. B. Critcurierp. 1974. A white-pollen variant of bristlecone pine.
J. Hered. 65:123.

KrieBEL, H. B. 1975. Interspecific incompatibility and inviability problems in forest
trees. Proc. 14th Meeting, Canad. Tree Improv. Assoc., Part 2:67—79.

KrucmMan, S. L. 1970. Incompatibility and inviability systems among some western
North American pines. Proc. Symp., Sexual Reproduction of Forest Trees. Vol.
2. IUFRO Sec. 22 Work. Group, 13 p.

Mastroctuseprr, J. D. 1976. Flavonoids of Pinus subsection Balfourianae. Bot. Soc.
Amer. Ann. Meeting, Abstracts of Papers:57—58.

MAstTrROGIUSEPPE, R. J. 1972. Geographic variation in foxtail pine, Pinus balfouriana
Grev. & Balf. M.S. Thesis: Calif. State Univ., Humboldt, 103 p.

, and J. D. MAstrociusEpre. 1975. Geographic variation in Pinus balfouri-
ana Grev. & Balf. Bot. Soc. Amer. Ann. Meeting, Abstracts of Papers:57.

Mirov, N. T. 1967. The genus Pinus. 602 p. Ronald Press, New York.

Pearson, G. A. 1931. Forest types in the Southwest as determined by climate and
soil. USDA Tech. Bull. 247:-143 p.

Ptym ForsHett, C. 1974. Seed development after self-pollination and _ cross-
pollination of Scots pine, Pinus sylvestris L. Stud. Forest. Suecica 118: 37 p.
Rew, W. H. 1972. Germination of Pinus aristata Engelm. Great Basin Nat. 32:

235-237.

210 MADRONO [Vol. 24

SARVAS, R. 1962. Investigations on the flowering and seed crop of Pinus silvestris.
Comm. Inst. Forest. Fenn. 53(4): 198 p.

SCHUBERT, G. H., and W. J. RrETVELD. 1970. Bristlecone pine—its phenology, cone
maturity and seed production in the San Francisco Peaks, Arizona. USDA
Forest Serv. Res. Note RM-180, 7 p. Rocky Mountain Forest and Range Exp.
Stn.

Uvex, H. 1927. The seeds of the genus Pzmus, as an aid to the identification of
species. Bull. Agric. and Forest. Coll., Suigen, Korea, No. 2: 129 p.

WricHT, R. D. 1963. Some ecological studies on bristlecone pine in the White
Mountains of California. Ph.D. Thesis: Univ. Calif., Los Angeles, 118 p.

ZAVARIN, E., and K. SNAJBERK. 1973. Variability of the wood monoterpenoids from
Pinus aristata. Biochem. Syst. 1:39-44.

, K. SnNayBerkK, and D. Battery. 1976. Variability in the essential oils of
wood and foliage of Pinus aristata and Pinus longaeva. Biochem. Syst. 4:81-92.

A NEW GYPSOPHILOUS SPECIES OF PHACELIA
(HYDROPHYLLACEAE) FROM COAHUILA, MEXICO

N. DUANE ATWoopD
Bureau of Land Management, Cedar City, Utah 84720
DoNALD J. PINKAVA
Department of Botany and Microbiology
Arizona State University, Tempe 85281

Floristic studies of the Cuatro Ciénegas Basin in central Coahuila,
Mexico, have revealed a new species of Phacelia growing on gypsum
dunes and flats. This taxon is apparently restricted to the basin, adding
to the list of gypsophiles only known from there: Dyssodia gypsophila
Turner (1972a), Gaillardia gypsophila Turner (1972b), Haploesthes
robusta I. M. Johnston (1941), Machaeranthera gypsophila Turner
(1973a), and M. restiformis Turner (1973b).

Phacelia marshall-johnstonii Atwood and Pinkava, sp. nov. Plantae
perennes 1.5—2.5 dm altae, caudicibus ligneis usque ad 1.5 cm diametro;
caules 1—plures e basi erecti vel ascendentes ramificantes supra saepe
viscidi et dense canescentes, pilis patulis 1-2 mm longis et pilis brevior
et mollior 0.3-0.8 mm longis; folia aggregata diminuta sursum, petiolis
0.2—2.0 cm longis, laminis ovatis ad elliptica 0.8—5.5 cm longis 0.5—2.5 cm
latis, margine grosse crenatis ad duple crenatis vel leviter lobatis, apicibus
obtusis, basibus rotundatis ad subcordatis, utrinque dense _hirsutis—
viscidis; inflorescentiae terminales in axibus principalibus et ramis mag-
nis Cymarum compositarum scorpioidarum, cymae usque ad 13 cm longae
in fructum; flores numerosi subsessiles; sepala elliptica ad oblanceolata
usque ad 4.5 mm longa et usque ad 2 mm lata in fructum dense et grosse
hirsuta—viscida; corollae caesiae albidae ad extremum 5 mm longae infun-
dibuliformia, tubis 4 mm longis glabris, lobis 1 mm longis hirsutis extis
subtiliteris; antherae globosae ca 0.5 mm diametros exsertae—longae, fila-
mentis purpureis ca 1 cm longis glabris 0.8 mm supra basim corolla inser-
tis, appendicibus basalibus auriculiformibus 0.7 mm longis; styli longo—

Zl

ATWOOD & PINKAVA: PHACELIA

1977]

Phacelia marshall-johnstoni. a, habit; b, stem segment enlarged; c, co-

Fic. 1.
rolla and androecium; d, calyx and gynoecium (at flowering); e, fruit in calyx;

f, seed, dorsal view; g, seed, ventral view. Illustration based on holotype. Scale

line = 5 mm.

214 MADRONO [Vol. 24

exserti stamina equantes purpurei subti pubescenti grosse pilis 0.2—0.5
mm longis; capsulae ovoidae ad subglobosae 3.0-3.5 mm longae 2.5—3.0
mm latae apice subtiliter puberulentae; semina 4 elliptica 2.0-2.5 mm
longa 1.1-1.3 mm lata fusca irridescentia leviter, pagina dorsali foveo-
lata, pagina ventralis excavata reticulata, marginibus et crista centralie
ventralis corrugatis leviter.

Type: MEXICO: CoaAnuita: Cuatro Ciénegas Basin, along new
road to gypsum dunes, 0.3 miles S of Poso de la Becerra, gypsum flats
with Atriplex, Nama, and Sporobolus, 15 Aug 1975, Reeves & Pinkava
P13100 (Holotype: ASU; Isotypes: BRY, ENCB, GH, MEXU, NY,
LL-TEX, UC, US).

Additional materials examined: MEXICO: Coanutta: Poso de Escobeda, grassy
banks, 17 Aug 1967, Cole, Minckley & Pinkava P4086 (ASU); Julio’s Canal, 4.5
miles SSW of Cuatro Ciénegas, along roadside, 15 Aug 1967, Cole, Minckley &
Pinkava P3812 (ASU); stabilized dunes W of headwaters of El Chiqueros, 15 Aug
1967, Cole, Minckley & Pinkava P3953 (ASU); stabilized dunes S of Laguna
Grande, 8 Jun 1968, Lehto, Keil & Pinkava P5024 (ASU); gypsum dunes, 16 km $
of Cuatro Ciénegas, ca. 2 km SW of Poso y Balneario La Becerra, 26° 52’ N, 102°
09’ W, 770 m, 22 Mar 1973, Johnston, Wendt & Chiang 10334 (LL-TEX); 19 km
SW of Cuatro Ciénegas, gypsum dunes, with Petalonyx, Sporobolus, Dasylirion, 26°
52’ N, 102° 09’ W, 700 m, 11 Jun 1972, Chiang, Wendt & Johnston 7649 (LL-TEX).

Phacelia marshall-johnstonii (Fig. 1) is apparently closely related to
P. pallida 1. M. Johnston of the Phacelia crenulatae group (Atwood,
1975), but is distinguished from that taxon by the densely canescent,
harsh spreading hairs, only sparsely glandular pubescence and smaller
seeds.

This species is named in honor of Professor Marshall C. Johnston, Uni-
versity of Texas, Austin, devoted scholar of the floras of Texas and the
Chihuahuan Desert.

ACKNOWLEDGMENTS
We thank Dr. David Keil for critically reviewing the manuscript and Wendy
Hodgson for the illustration. Field work was supported in part by National Science
Foundation grants GB2461 and GB6477X awarded Dr. W. L. Minckley, Department
of Zoology, Arizona State University.

LITERATURE CITED
Atwoop, N. D. 1975. A revision of the Phacelia crenulatae group (Hydrophyllaceae)
for North America. Great Basin Nat. 35:127-190.
Jounsron, I. M. 1941. New phanerogams from Mexico, IV. J. Arnold Arb, 22:110-
124.
Turner, B. L. 1972a. A new species of Dyssodia (Compositae) from north central
Mexico. Madrono 21:421-422.
_ 1972b. Two new gypsophilous species of Gaillardia (Compositae) from
north central Mexico. Southw. Nat. 17:181-190. |
_ 1973a. Two new gypsophilous species of Machaeranthera (Asteraceae- —
Astereae) from north-central Mexico. Phytologia 26:116—120.
_ 1973b. Machaeranthera restiformis (Asteraceae), a bizarre new gypso-
phile from north-central Mexico. Amer. J. Bot. 60:836-838.

ROOT:SHOOT BIOMASS RATIOS IN SHRUBS IN
SOUTHERN CALIFORNIA AND CENTRAL CHILE

Puitie C. MILLER AND Epwarp Nc!

Department of Biology
San Diego State University, San Diego, CA 92182

One characteristic of vegetation in mediterranean-type climates is an
abundance of deep-rooted shrubs (Hellmers et al., 1955; Sacchori et al.,
1967). The deep rooting patterns correspond with the cool, wet winter
and hot, dry summer climatic pattern in which water is available deep in
the soil when surface layers are dry and temperatures are favorable for
photosynthesis and growth (Miller and Mooney, 1974). Where ever-
green, sclerophyllous shrubs predominate, annual precipitation is about
400-650 mm (Aschmann, 1973; Miller et al., 1977). Deep rooting pat-
terns imply an investment of carbon resources by the plants in the root
system, but there are few data on the root:shoot biomass ratios for
shrubs in mediterranean regions, probably because of the difficulty in
obtaining these figures (Kummerow et al., pers. comm.). At the begin-
ning of this study we expected to find that root-shoot biomass ratios of
shrubs in the mediterranean climatic regions of southern California and
central Chile would be greater than one and that the biomass of absorb-
ing roots, identified by small root size, would be correlated with the
transpiring leaf area of the shrub. Our field study tested these hypotheses
on selected shrubs in the 2 regions. We excavated 14 individual shrubs of
6 species: 8 in southern California and 6 in Chile. The main comparison
involved 2 Californian species: Adenostoma fasciculatum (4 shrubs exca-
vated) and Ceanothus greggii (2 shrubs excavated); and 2 species of
similar areal stature in Chile, Satureja gilliesii (4 shrubs) and Colliguava
odorifera (2 shrubs). In addition, one plant of Heteromeles arbutifolia
and one of Arctostaphylos glauca were excavated in California.

DESCRIPTION OF STUDY SITES

In California roots of all plants except the Heteromeles and one Ade-
nostoma (Adenostoma 4) were excavated at Echo Valley in San Diego
County (32°54”N, 116°39”’W). These 2 plants were excavated near the
Viejas Road, 15 km southwest of the Echo Valley site. In Chile all the
plants were obtained by excavating back from a road cut at the Fundo
Santa Laura s'te (33°04”S, 71°00’W). We used hydraulic excavation
to remove roots from the soil. This requires large volumes of water, a
high pressure pump, and suitable drainage away from the root washing
site. Rapid drainage of water and soil was necessary during excavation

'Present address: Department of Agronomy, University of California, Davis 95616

215

216 MADRONO LVol. 24

to keep the root system exposed and allow its extraction. Therefore,
plants near road cuts or on steep banks were chosen. However, the center
of the excavated plant was always more than 1 m from the road cut.
Roots were excavated at comparable sites in California and Chile: ap-
proximately 55 km from the coast at about 1,000 m. Annual precipita-
tion in both regions is about 550 mm, with about 60% during the winter
6 months in California and 83% in the analogous months in Chile (Mil-
ler et al., 1976). The annual mean temperatures at the research sites are
about 17C in California and 15C in Chile. The Californian site burned
in 1950. The Chilean site was subjected to wood cutting prior to 1959,
although the species measured are not used for firewood (Aschmann and
Bahre, 1977). Root excavations in both countries were made during the
respective summer when surface layers of soil were dry but subsurface
layers were still moist. Roots were excavated in June, July, and August,
1971 in California and in February, 1972 in Chile. Precipitation through
the winter preceding the excavations was about 250 mm at the site in
California. A complete precipitation record prior to excavations is not
available in Chile; however, in California there was no precipitation
during the 2 months preceding excavations, and in Chile none during the
4 months preceding excavations. Average soil moisture at 0.3-1.2 m
depth decreased from 0.25 to 0.09 cm® water per cm® soil during the ex-
cavations in California; thus, the soil was presumed to be dry at excava-
tion in both countries.

Shrubs excavated from the Echo Valley site were growing near recent
road cuts; those from the Viejas site were adjacent to recent excavations
for land fill. Slope inclinations were 10°—15° for each site. Slope exposure
of the plants at Echo Valley was north; Adenostoma 4 was on a south—
facing, and Heteromeles on an east-facing exposure. The soils at both
excavation sites are sandy loams, underlain by decomposed granite from
weathered Bonsall Tonalite, without much vertical stratification and,
although of variable depth, more than 0.5 m deep. The soils at the Viejas
site were much rockier than those at Echo Valley. Ceanothus and Ade-
nostoma were excavated in a mixed stand of these species. Adenostoma
4 was in a pure Adenostoma stand. Arctostaphylos was associated with
Quercus agrifolia and Ouercus englemanii, and Heteromeles was associ-
ated with QO. agrifolia and Salvia apiana. No herbaceous cover occurred
on the plots excavated.

The excavation site in Chile was predominantly northwest facing. The
slope inclination was highly variable but was 10°-25° where excavations
were made. The surrounding vegetation was dominated by Satureja
gilliesiti and Colliguaya odorifera, with occasional Lithraea caustica,
Trevoa trinervis and Cryptocarya alba. The site had a definite herbace-
ous layer. Soils were generally coarse textured but not underlain by
decomposed granite. The absence of decomposed granite facilitated the
excavation.

1977] MILLER & NG: ROOT:SHOOT BIOMASS Ail?

METHODS

For hydraulic excavations in California a 400-gal tank truck and a
Barnes impeller type pump were used. The pump was capable of putting
out 15 gal min™ at 100 lbs in® pressure. A similar pump configuration
was assembled in Chile where two 50—gal drums provided water storage.
The relatively low outflow rate was important because the water supply
was limited and sites were remote from the source. A variable nozzle reg-
ulated the pressure of water from a narrow, high pressure stream to a fine
spray. The fine spray minimized loss of root material during excavation,
but the narrow high pressure stream was necessary to remove the soil.

For selected plants, maximum height and radial extent of the crown
were measured, and the foliage was tied. Three metal reference stakes
were placed in the ground, two in the upper and lower edges of the plot
0.5-1.0 m from the plant, and the third as close as possible to the plant,
without piercing its burl (lignotuber). The stakes served as depth gauges
while washing the soil and to support the plant as it was undercut. As
soil was washed away, exposed roots were marked according to depth.
Depth was recorded every 0.2 m in California and every 0.1 m in Chile.
Excavation stopped when a impenetrable layer of soil or rocks was
reached, when the drainage became insufficient because the level of the
road was reached, or when the root system was exposed as completely
as possible. Maximum radial extent of the root system was measured. The
entire plant was then brought into the laboratory and separated by level
for oven drying for 24 hr at 85C.

In California, unbiased estimates of the biomasses of different sized
roots, total root length, and root area could not be made because of the
nature of the excavation. The decomposed granite washed away in 0.5—
1.0 cm*® chucks, which probably contained most of the fine roots. There-
fore, only dry weights of all roots in each stratum were measured. In
Chile the soil particles disintegrated more readily with the application of
water and fewer fine roots were lost. The retention of small roots allowed
those of each plant to be divided into size classes of different diameters
within different soil depths. The dry weights of roots in each class were
measured. Roots were randomly subsampled from similar size classes and
their total length and weight were measured to obtain a length: weight
ratio for each size class. The length: weight ratio times the weight and
diameter were used to estimate the area of each size class of roots.

RESULTS AND DISCUSSION
Even with the variable size of shrubs of the same species, the measured
root:shoot biomass ratios were less than one for all individuals (Table 1).
The largest root:shoot ratio of an individual was 0.93 and the smallest,
0.25. Ceanothus greggii, a shallow—rooted species, had the smallest root:
shoot ratios (0.25 and 0.39), while Adenostoma fasciculatum, a deep—
rooted species, had root:shoot ratios of 0.49-0.69. Heteromeles, which is

*,Ul $0 X PZ ‘BUOTGO sem Bay ,

"[anq jo jY4SIoM SapNpouy x»

+
= 68°0 09°0 640 6.0." 220. Se 0 L8°0 vL'0 6r'0 790 090: 050 6¢°0 $70 OILVa
S SSVWOId LOOHS: LOO
O7E OSL O02 O62 OLI9 OO8h O'rrrl 0968 O9SL O€F2 092 O82 Op OTe (3) SSWWOIG IVLOL
OO OTL 30'S0r Ore “0102 0751 = = = eee gee sa = (3) sseuorg
Jett
(¢-U 3)
= yaa 069 Ors O8tT O86I O'OLET 0°000¢ OSS 0.69 — 0 00F O'06Z OOF, Atsuap ssvULOIq Uva
OS O82 006 O29 O09 O8CT xx0'0L9 OVS VOSC™ 00 OTL rs OTT 0O'¢9 (3) sseulorg
ee Oct 80 88T ¢£9°0 670 $800 cov eVO00 =~ ECO0 800 s80°0 (UI) OUINJOA
ee NS ee US CS lic. evo vO'T 820 cov 7 Oa S00. oa 0) 870 (,W) valy
O a as vt a ROG GI cl 90 VC VOW Se pac, 0 v0 9°0 (U1) JoyoUTRIC,
(Z 90 720 £80 £0 90 v0 56 @) ae) OT OT 80 €0 2.0 rane) (wi) yydaq
2 100y
A
= = = ae ee O'sTz 0°64 = i shee OZE O26 (3) sseurorq year]
a 6S eee a Sie +x0'OSS oe92 = ee Bee OST OTST (3) ssBuLorq Wa1g
OL OV OLLIE ~OZ9 “OLS. OSS? O'ST2 O'CbvE *x0'90S OPT OST OZT oers 062 (3) sseulorg
¢00;0 $00 GiOe SCO SOLE. = 9600 c8'0 080 29 07 ~ $0010. =~ .c10'0 L0°0 1a xe) (gW) oUINJOA
sto0 800 OF0 S8¢c0 2¢1 60 8£°0 sv'0 LOO. SIO) B= “S00 0} Me) eT0 (;W1) Roly
vO ceo 03:0) ROVOS Sol. FOL 0Oo'T s£°0 ¢6°0 SLO: Sa USC 0 Sz 0 00) (UL) JoyoureIg
070 090 s80 060 S60 O0T Sol 641 S60. <'S0 “= 070 140 = OTT (ul) 4310H
y00YS
@ I b c Z I I I v exe I Z I qied quel
DAaf14o0 po us aiyis pDlaaAnjoy DINDIS po f7qnqav WMNIDINISY{ DUWOISOUIP PL 135943 SNYOUDIAD
DADNSIII0 D sojxydpjs sajawosajaH
-0JIAY
- [eNpPIAIpur Jo Jaquinu pure sateds
nN

0} jou ‘sua}sAS JOOI PUL JOOYS JY} JO }U9}KO [VII JY} O} AdJoI QUINOA pu ‘vaIe ‘AAJOULIC( “SINTWAINSVAY AZIS ANUHS JO AUVINIWAS

‘JOOI 10 JOOYS [eNprIAIpur ue

‘l AITAVL

1977] MILLER & NG: ROOT:SHOOT BIOMASS 219

considered deep-rooted, and Arctostaphylos glauca, Satureja, and Colli-
guaya, which are all considered shallow-rooted, had similar root:shoot
ratios, usually 0.60 to 0.93.

Except for Heteromeles, the ground surface area underlain by the
roots of an individual was larger than the vertical projection of the
crown (Table 1) (Hellmers et al., 1955). The ratios of the ground sur-
face underlain by roots to the vertical projection of the crown were:
Ceanothus, 1.9-2.0; Adenostoma, 6.7—7.2 in two deeper-rooted plants
and 1.4 in the shallow-rooted plant; Arctostaphylos, 2.1; and Satureja,
2.6—4.4.

Roots were concentrated in the upper 0.3 m of soil (Table 2). In Cea-
nothus, Heteromeles, Arctostaphylos, and one Adenostoma all of the root
biomass was in the upper 0.3 m. In the other Adenostoma, 50-55% of
the biomass was in the upper 0.3 m and over 94% in the upper 0.8 m.
Roots of Adenostoma extended to at least 1 m on large plants. In Sature-
ja 83-100% of the root biomass was in the upper 0.3 m, with maximum
root depths between 0.4 and 0.84 m. In Colliguaya over 84% of the root
biomass was in the upper 0.3 m, with roots of one individual extending to
0.6 m. Rooting depth appeared independent of root weight both within a
species and with all species combined, but was directly related to shoot
height in Satureja and Colliguaya. Root biomass densities of 100-1400 g
dry weight m™ in the 0O-0.3 m depth (Table 3) compare with values for
crop plants of 300 g m™“* (Penning de Vries, pers. comm.), and tundra
plants of 1,000 g m™® (Dennis and Tieszen, 1972). Root biomass densi-
ties at the surface were less with deeper—rooted individuals, indicating a
trade-off between high exploitation of the surface and exploitation of
the deeper soil.

Mean root diameters decreased with depth in Satureja and Colliguaya,
the only species for which root weights could be measured by diameter
class (Table 4). At each depth the length of the roots in the smallest
diameter class was larger than that for the other diameter classes. Roots
smaller than 0.5 mm diameter comprised 10-25% of the total root bio-
mass and 40-50% of the total root length. The weight of the small roots
decreased with both shoot and root biomass. Percentage of the total
root biomass which was small roots decreased with increasing root bio-
mass because of the increasing diameter of roots with age and did not
correlate with the shoot biomass in the measurements.

Our excavation underestimated the root biomass because of the loss of
fine roots in excavation (Caldwell and Fernandez, 1975) and because of
the death and sloughing of fine roots as the soil dried in early summer,
but a correction for this loss does not increase root:shoot ratios greatly.
The correction is based on concepts of absorbing root densities required
for the absorption of phosphorus, nitrogen, and water. Chaparral soils
are generally considered to be nitrogen and phosphorus deficient (Hell-
mers et al., 1955; Christensen and Muller, 1975) and are dry during the

[Vol. 24

MADRONO

—N

“yoIy} W 7°C Ie S[aAg[ suljdwies ,,,

"We ¢°0-0'0 [eA] Suljdwues ,,

°°%6'°O Uk} Sso] SI 88e}UIIIEd Sa}eOIpUT ,

O°I-6'0
*k | : | 6'0-8'0
c 8°0-L'0
a Sr ed | L0-9'0
v (b I I 90a s0
¢ z ‘ z LT LC SO-F'0
L 6 y v xk y'0-¢'0
I z 2 es iz 7 a eam [ | | €°0-7'0
val o¢ v2 LI 6€ 82 OoT oor i ve} oor{ oor OOr Z'0-1'0
Ss 89 8¢ Is 6b Sr Te l l [ 1'0-0'0
j00"g
OOT iS 8Y Tv 19 LV 62 LI xk he Be 94 4 Ol c'0-0'0
O¢ €? ue LT 92 ce er 62 cv vs ve ve IT v'0-c'0
Sil 1Z LT ei eA 61 LT 14 GE 8 a4 Or 9°0-7'0
L el! 6 5 81 cl Be : 92 8°0-9'°0
I © I I Or 6 at 82 O'T-8°0
I Or ST et-O
s Viel
Or 9 I-VI
9 sist
7c
c it v £ c I I v a g I 6 I
DAa{1dOPO 18 anp]03 pInD}s = pyoftjnquD WNIDINIVS Df 133048
DKkDNS1]0D Dl aAnqDS sopkygvjs sajamosaja yy DULOISOUIP PF snyjouvay
-OJIAY
/ ; [eNprIArput jo raquinu pur satseds yied urd

‘AOVAUNS TIOS AHL MOTAG AGNV AAOUV STAAAT LNAMAIAIAG NI SSVWOId LOOU JO GNV SSVWOId LOOHS AO SHOVINGONAG ‘¢ ATV

2

MILLER & NG: ROOT:SHOOT BIOMASS

1977]

“Ul 7°Q AB S[IAI[ BUTIOOY yyy

"We ¢°0-0'0 [PAF] SUNOOY x»
"SUIa}S WO} pajeiedas jou 319M SOARIT ,

£0'0 00 90'0 00 70-00
CTO cO'0 vv'0 exe) b'0-c'0
Os'0 £0°0 891 £0°0 9°0-¥'0
CLT EGO OCC O8'P 8°0-9'0
vO’ 8720 OO'T O'T-0'T
£9°C TZ°0 99'0 Colt a
0s'0 VA-cL
8O'T OT bay a
O1}BI SSPULOIG W19}s: Jeo]
O'I-6'0
! 07 | oe
cl I SOs 0
ZI 9 I Be ad /'0-9'0
cl MS SI 025 0
bs 62 of Z ey Ee | S0-F'0
Tcl Sol hy CSI 26 ZL yO-¢£'0
SET 88 S7e STZ i | | S0-20
O77 $9¢ clr c9¢ OL¢I OOO¢ Z01 ost OOr 062 Ors CO-T'0
i [ 1'0-0'0
(g-W 3) A}ISUaP SSPUTOIG JOOY
v o e ¢ I T T v S T C T
4218 14]13 pInns 0110 {27nQAav 4WNIDINIIIS Df 133943 (Ww) JY4SIIFy
pDlaAnqDsy SOJKYGDISOJIAP SaJamodsaja DWUWOISOUIP F snyjounay

fenptarput jo Joquinu pue satzads

‘psa{1d0po *D pure Z WNIDINIISD{ ‘Pf UO AapeUL JOU I1aM S}UIWIINSea “AOVANAS
T1IOS AHL MOTAG GNV TAOSV SIAATI INAATIAIG NI SOILVA SSVINOIA WALS!AVAT GNV ‘ALISNAG SSVWOId LOOM ‘NOILNAINLSIG SSVINOIA AVAT ‘¢ ATAVE

222 MADRONO [Vol. 24

TABLE 4. FRACTION OF THE TOTAL ROOT LENGTH IN DIFFERENT SOIL DEPTHS AND
SIZE CLASSES. The mean total root length for four Satureja plants 49.2 m and for two
Colliguaya plants 8.3 m.

Root diameter class (mm)

Soil depth
(m) O-0.5 0.5-1.0 1-2 2-4 4—6 6-8 8-10 > 10
Satureja

0-0.1 0.265 0.064 0.031 0.029 0.017 0.011 0.008 0.005
0.1-0.2 0.105 0.048 0.050 0.021 0.010 0.003 6) 0
0.2-0.3 0.093 0.050 0.052 0.019 0.006 6) 0 O
0.3—0.4 0.025 0.023 0.016 0.006 0.001 6) 0)
0.4—-0.5 0.009 0.007 0.007 0.001 0 6) 0 0
0.5—-0.6 0.007 0.005 0.001 6) 0) 6) 0 O
0.6—0.7 0.002 0.002 0.002 0 0) 0) 6) 6)

Total 0.506 0.199 0.159 0.076 0.034 0.014 0.008 0.005

Colliguaya

0-0.1 0.203 0.122 0.087 0.052 0.023 0.006 oO 0
0.1-0.2 O51 0.110 0.070 0.046 0) 0) 0) ¢)
0.2-0.3 0.035 0.041 0.036 0) 0 0 0 0
0.3-0.4 0.006 0.006 0.006 6) 0 0) 6) 0)

Total 0.395 0.279 0.199 0.098 0.023 0.006

summer, so soil systems can be expected to be organized for the efficient
uptake of these minerals. To absorb phosphorus efficiently a root system
should have roots or mycorrhizae about 0.5 cm apart because of low
mobility of phosphorus in the soil (Bieleski, 1973). For efficient absorb-
tion of nitrate, roots should be about 4 cm apart (Van Keulen et al.,
1975) and for water, about 8 cm apart (Lambert and Penning de Vries,
1975). If a biomass of fine roots adequate to exploit the soil nitrate is
added to our measured values, root:shoot ratios from 0.26 to 0.93 are
calculated; to exploit the soil phosphate, root:shoot ratios from 0.34 to
3.35 are obtained. Ratios above 1.0 are always associated with rooting
systems deeper than 0.5 m and small calculated root biomass densities.
It is unrealistic to add the full biomass density to these sparse root sys-
tems, because the larger roots are not present to support the development
of the fine roots. Thus, we conclude that the root-shoot ratios of chap-
arral shrubs, even though some are considered deep-rooted, have root:
shoot ratios between 0.3 and 1.0 and have root biomasses concentrated
near the soil surface.

ACKNOWLEDGMENTS
This work was supported by NSF grant DEB75-19491 as part of the IBP-Origin
of Ecosystem Structures project. We thank Mr. Ernesto Hajek, Universidad Catolica
de Chile, for his support in Chile and Dr. Jochen Kummerow and Mr. Wayne Stoner,
San Diego State University, for their critical discussions and Ms. Martha Poole and
Patsy Miller for their help in preparing the manuscript.

1977 | MILLER & NG: ROOT:SHOOT BIOMASS 206

LITERATURE CITED

ASCHMANN, H. 1973. Distribution and peculiarity of Mediterranean ecosystems,
p. 11-19. In F. diCastri and H. A. Mooney (eds.) Mediterranean type ecosys-
tems: Origin and structure. Springer-Verlag, New York.

, and C. Bawre. 1977. Man’s impact on the wild landscape. In H. A. Moon-
ey (ed.) Convergent evolution in California and Chile. Dowden, Hutchinson,
and Ross, Stroudsburg, Pa. (in press).

Brerski, R. L. 1973. Phosphate pools, phosphate transport, and phosphate avail-
ability. Ann. Rev. Plant Physiol. 24:225-252.

CaLpwELL, M. M. and O. A. FERNANDEZ. 1975. Dynamics of Great Basin shrub root
systems, p. 38-51. In N. F. Hadley (ed.) Environmental physiology of desert
organisms. Dowden, Hutchinson, and Ross, Stroudsburg, Pa.

CHRISTENSEN, N. L. and C. H. Mutter. 1975. Effects of fire on factors controlling
plant growth in Adenostoma chaparral. Ecol. Monogr. 45:29-55.

Dennis, J. G. and L. L. Trrszen. 1972. Seasonal course of dry matter and chloro-
phyll by species at Barrow, Alaska, p. 16-21. Im S. Bowen (ed.) Proceedings
1972 Tundra Biome Symposium, Lake Wilderness Center, University of Wash-
ington, 3—5 April 1972. U.S. International Biological Program, U.S. Arctic Re-
search Program, U.S. Tundra Biome, Hanover, New Hampshire.

Heitimers, H., J. S. Horton, G. JUHREN, and J. O’KEEre. 1955. Root systems of
some chaparral plants in southern California. Ecology 36:667-678.

KuMMERow, J., D. Krause, and W. Jow. 1977. Root systems of chaparral shrubs.
(pers. comm.).

LameertT, J. R.and F. W. T. PENNtNG de Vries. 1975. Dynamics of water in the soil-
plant-atmosphere system: A model named Troika, p. 257-273. Im A. Hadas
(ed.) Ecological Studies. Analysis and Synthesis, vol. 4. Springer-Verlag, Berlin.

Mittrer, P. C. and H. A. Mooney. 1974. The origin and structure of American arid-
zone ecosystems. The producers: Interactions between environment, form and
function, p. 201-209. Jn Proceedings First International Congress of Ecology,
The Hague, Netherlands.

————, D. Brappury, N. THrower, V. La Marcue, and E. Hajek. 1977. Macro-
climate. Jn H. A. Mooney (ed.) Convergent evolution in California and Chile.
Dowden, Hutchinson, and Ross, Stroudsburg, Pa. (in press).

SHACHORI, A., D. RosEnNzweic, and A. PoLyAKorr-Mayser. 1967. Effect of Mediter-
ranean vegetation on the moisture regime, p. 291-311. In W. E. Sopper and H.
W. Lull (eds.) International Symposium on Forest Hydrology. Pergamon
Press, New York.

Van KeuLeN, A., N. G. SELIGMAN, and J. GoupriAn. 1975. Availability of anions
in the growth medium of roots of an actively growing plant. Neth. J. Agris. Sci.
23:131-138.

A NEW SPECIES OF IVESIA (ROSACEAE)
FROM SOUTHEASTERN OREGON

BARBARA J. ERTTER! and JAMES L. REVEAL

Department of Botany, University of Maryland, College Park 20742

The geology and relative isolation of southeastern Oregon make the
area an excellent location for discovering new and unusual plant species.
Leslie Gulch in Malheur County is one site in particular whose botanical
treasures have only recently been discovered (Glad, 1975; Barkley, in
press). Approximately ten percent of the species found there are rare or
endangered (Packard, pers. comm.), and several are endemic to the
barren ash and tuff slopes lining the canyons in the Succor Creek and
Owyhee River drainage basin. In this paper we present a new species of
Ivesia (Rosaceae: Potentilleae) from Leslie Gulch. This species was ini-
tially discovered by the senior author while a student at the College of
Idaho, Caldwell.

Ivesia rhypara Ertter & Reveal, sp. nov. Species insignis stylo soli-
taria, petalis brevibus albis, ramis prostratis et inflorescentiis apertis cy-
mis, a speciebus nobis notis bene distincta. Fig. 1.

Low spreading herbaceous perennial from a branched caudex atop an
extensive, woody root system, this often clothed with old, reddish—pubes-
cent leaf—bases; herbage grayish— or greenish—white, villous to canescent,
eglandular; stems erect before anthesis, becoming prostrate and trailing
as the inflorescence develops and lengthens, 5-15 cm long; leaves essen-
tially basal, canescent, 3-8 cm long, with 5—15 pairs of closely overlap-
ping leaflets; leaflets divided to near the base into 3-5 segment, these
ovate to rounded, 1.8—3 (—4) mm long; inflorescence a more or less open
cyme, 3—20 cm long; bracts leafy below and often appearing near the
middle of the stem due to the reduction and loss of the first branch of
the inflorescence, 3—7 mm long, 1.5—-4 mm long above in the inflorescence,
mostly ovate; hypanthium shallowly cupulate, 2—2.3 (-—2.5) mm wide,
yellowish to golden within, the receptacle densely covered with long,
white, silky—villous hairs; bractlets ovate, about two-thirds the length
and one-half the width of the sepals; sepals 5, triangular, 1.8-—2.5 mm
long; petals 5, white, 0.8—-1.5 mm long, 0.2—0.5 mm wide, narrowly spatu-
late to oblanceolate, inconspicuous; stamens 5, inserted well away from
the margin of the receptacle, the filaments linear, 1.5—-1.8 mm long, gla-
brous, the anthers yellow except for the magenta—colored marginal su-
tures, 0.4—-0.5 mm long; pistil solitary, 1-2 mm long, glabrous; achenes
smooth, brown, 1—1.3 mm long. Flowering from May to October.

1 This paper has been submitted to the Department of Botany, University of
Maryland, as partial fulfillment of two credits of Special Problems given during the
Fall Semester of 1976.

224

1977] ERTTER & REVEAL: IVESIA

bo
bo
[wae

Yas. y
nets alah
et UY ory
a ay
=a + ‘A a \ od
Ds 2 wr ty, de
ae \ aw Dae
A y, ty
(} { i
¥\ X
\ f
\
AG
% Rue a
\)
Be)

Fic. 1. Illustration of Jvesia rhypara. A, habit (x 1/5); B, details of the leaves
and inflorescence, the latter shown in an upright position for convenience; see A for
actual location (x 1); C, detail of flower (x 6); D, mature achene (x10);

226 MADRONO [Vol. 24

Type: OREGON, Malheur Co., along the rim above Leslie Gulch E
of Owyhee Reservoir, 0.2 km down Leslie Gulch road from the entrance
to the canyon, on red to yellow volcanic ash in sparse vegetation, at ca
1430 m, sec. 9, T.26S., R.45E., 13 Jul 1975, Reveal & Ertter 3894
(Holotype: US, isotypes: BRY, CAS, CIC, GH, ID, K, MARY, MO,
NY, OSC, RM, RSA, TEX, UC, UTC, WTU, and elsewhere).

ADDITIONAL SPECIMENS EXAMINED: OREGON, Malheur Co., Leslie
Gulch, 30 Sep 1973, Packard 73-293 (CIC); Leslie Gulch, 26 May
1974, fivtter 47/4 (CIC).

Ivesia rhypara is not likely to be confused with any other species of
the genus. In Keck’s (1938) revision, /. rkyvpara would key out to J.
shocklesi S. Wats., an alpine plant of the Sierra Nevada of California
and west-central Nevada. Our new species may be quickly distinguished
from that species by the denser, non-glandular pubescence, the consis-
tently solitary style, and the small, white petals. In southeastern Oregon,
only J. baileyi S. Wats. is regularly encountered, and then only in the
higher mountains. It differs from /. rhypara in having an erect, glandu-
lar—pubescent stem, larger leaflets, and three to seven pistils. The inter-
mountain species of /vesta which are found in the desert all have more
than five stamens, while those species in the region with five stamens all
occur at higher elevations in more mesic sites. In addition, all these other
species are glandular, less densely pubescent, and typically with yellow
petals.

The new species is apparently confined to an area of less than one
mile square on either side of the road entrance to Leslie Gulch, and has
not been found in any adjacent areas in the Succor Creek drainage
(Packard, pers. comm.). It has been found on soils ranging from a red-
dish tuff to a loose, yellowish volcanic ash. Near the road on the north
side of the canyon, it is associated with Poa sandbergi, Agropyron sbica-
tum, Eriogonum strictum, Physaria chambersii, Astragalus sterilis, Linum |
perenne, Penstemon acuminatus, Eriophyllum lanatum, and an unusual |
form of Monardella odoratissima. On the south side of the canyon, it
grows on the bare ground among Juniperus osteosperma and Purshia —
tridentata. It is very local and relatively infrequent in all known loca-
tions, and should be considered an endangered species. Leslie Gulch is
now open to recreational traffic, and off-road vehicle activities in the area |
could easily destroy the fragile slopes to which the species is restricted.

The species epithet is based on the Greek “rhyparos” meaning “dirty” |
or “grimy’’, not only in reference to the dirty or dusty appearance of the —
species, but also here applied to honor James W. Grimes, a fellow student |
and collector with the senior author of the Leslie Gulch flora. Grimes is
currently attending Utah State University and working on the flora of ©
the Leslie Gulch area as part of his Master’s program.

1977 | REVEAL & MORAN: CHROMOSOME COUNTS 221

ACKNOWLEDGMENTS
We wish to thank Dr. Patricia L. Packard, Professor of Biology at the College
of Idaho, who aided us in field studies of this new plant, and for making her
herbarium material and observations available. The assistance of C. Rose Broome
and Arthur Cronquist in reviewing the manuscript is appreciated. Field work in
1975 was supported by National Science Foundation grant BMS 75-13063 to the
junior author.

LITERATURE CITED
Bark ey, T. M. In press. “Senecio.” N. Amer. FI.
Grab, J. B. 1975. Taxonomy of Mentzelia mollis and allied species. Madrono
23:283-292.
Keck, D. D. 1938. Revision of Horkelia and Ivesia. Lloydia 1:75—-142.

MISCELLANEOUS CHROMOSOME COUNTS OF
WESTERN AMERICAN PLANTS—IV

JAMES L. REVEAL
Department of Botany, University of Maryland, College Park 20742
Rem Moran
San Diego Museum of Natural History, San Diego, CA 92112

This series of papers, of which this is the fourth, reports chromosome
counts of miscellaneous western American plants. Previous parts have
dealt with plants collected of various floristic projects of the senior
author (Reveal & Styer, 1973, 1974; Reveal & Spellenberg, 1976). The
present contribution covers counts of plants collected in Baja California,
Mexico, in February of 1973.

Flower buds were collected in developmental stages from plants grow-
ing in their native habitats, and were fixed in ethanol and glacial acetic
acid (3:1). Anthers were squashed in acetocarmine and camera lucida
drawings were prepared. The counts reported here were made by Reveal,
but all of the species determinations except those of Eriogonum, or as
indicated, were made by Moran. Voucher specimens are deposited at the
San Diego Museum of Natural History (SD).

Chromosome numbers are reported for 31 taxa; for 26 of these we
have seen no previous reports. Three counts represent first reports for the
genus: Chorizanthe (n = 20), Wislizenia (n = 20), and Errazurizia
ts — 14).

Garrya grisea Wiggins. n = 11. Voucher: M&R 20176 (Fig. 1), north
slope of Cerro Azufre, ca 1600 m, B.Cfa.Sur (near 27°30’N, 112°36’W),
17 Feb 1973.

This first reported number for the species is common in the genus
(Bolkhovskikh et al., 1969). The species has not been reported previ-
ously from south of Sierra San Pedro Martir.

MADRONO

[Vol. 24

ae
este?
eee.”

1977 | REVEAL & MORAN: CHROMOSOME COUNTS 229

Chorizanthe pulchella Brandegee. n = 20. Voucher: M&R 19625
(Figs. 2,3), 2 miS of El Crucero, ca 530 m, B.Cfa.Norte (near 29°14’N,
114°11’W), 1 Feb 1973.

This is the first published count for the genus Chorizanthe. Judging
from Reveal’s understanding of Eviogonum, the basic number of the
genus would seem to be x = 10. On labels for the voucher, the chromo-
some number was given incorrecily as 7 = 15.

Eriogonum elongatum Benth. var. areorivum Reveal. n = 17. Vouch-
ers: M&R 19913 (Fig. 4), bed of Arroyo San José de Castro, ca 340 m,
B.Cfa.Sur (near 27°33’N, 114°32’W), 7 Feb 1973; M&R 19930 (Fig.
5), bed of Arroyo Largo, 4 mi E of the mouth, ca 110 m, B.Cfa.Sur (near
27°36’N, 114°45’W), 8 Feb 1973.

This count agrees with that given by Stokes and Stebbins (1955) for
var. elongatum.

Eriogonum encelioides Reveal & Hanson. m = 20. Vouchers: M&R
19813 (Fig. 6), Arroyo del Portezuelo, 9.5 mi S of San José de Castro,
ca 275 m, B.Cfa.Sur (near 27°26’N, 114°27’W), 5 Feb 1973; M&R
19967 (Fig. 7), Arroyo de las Casitas, 2.5 mi above the mouth, ca 100 m,
B.Cfa.Sur (near 27°31’N, 114°36’W), 9 Feb 1973. |

Reveal and Hanson (1967) associated this species with Eriogonum
elongatum. However, examination of the plant in the field clearly shows
it to be a member of section Fasciculata Benth. and most closely related
to LE. molle Brandegee, of Cedros Island.

Eriogonum fasciculatum Benth. var. fasciculatum. n = 20. Voucher:
M&R 20289 (Fig. 8), Colnett Mesa, 0.5 mi N of Colnett, ca 80 m,
B.Cfa.Norte (near 31°05’N, 116°13’W), 25 Feb 1973.

This count agrees with previous reports for this variety (Stebbins,
1942; Stokes & Stebbins, 1955; Reveal, 1967).

Eriogonum fasciculatum Benth. var. flavovirde Munz & Johnston.
n = 20. Vouchers: M&R 19629 (Fig. 9), Arroyo Leon, 5 mi N of Punta
Prieta, ca 250 m, B.Cfa.Norte (near 29°O1’N, 114°12’W), 1 Feb 1973;
M&R 20163 (Figs. 12, 13), saddle on north side of Cerro Azufre, ca
250 1m, Cia.sur. (edn 27-30 N;112°36 WwW). 17 Keb 1973.

Fics. 1-29. 1. Garrya grisea, n = 11, diakinesis. 2. Chorizanthe pulchella, n = 20,
diakinesis; 3, metaphase 1. 4, 5. Eriogonum elongatum var. areorivum, n— 17,
diakinesis. 6. Eriogonum enceiloides, n = 20, telophase II; 7, anaphase I. 8. Eriogo-
num fasciculatum var. fasciculatum, n = 20, metaphase I. 9, 12. Eriogonum fascicu-
latum var. flavorirde, n = 20, diakinesis; 12, late diakinesis. 10. Eriogonum fascicu-
latum var. emphereium, n = 20, telophase II; 11, diakinesis. 14. Eriogonum fastigia-
tum, n = 20, metaphase I. 15. Eriogonum inflatum var. deflatum, n = 16, anaphase
I; 16, metaphase I. 17. Eviogonum intricatum, n — 16, diakinesis; 18, telophase I.
19. Eriogonum moranii, n = 20, anaphase I. 20. Eriogonum pilosum, n = 16, dia-
kinesis; 21, telophase II. 22, 24. Eriogonum pondii, n — 20, metaphase I; 23, dia-
kinesis. 25. Eriogonum preclarum, n — 20, metaphase 1; 26, diakinesis. 27. Eriogo-
num repens, n = 16, metaphase I; 28, diakinesis; 29, telophase IT.

230 MADRONO [Vol. 24

Shreve and Wiggins (1964) and Reveal and Munz (1968) considered
this variety endemic to southern California, but on the basis of pubes-
cence characters of leaves, involucres, and flowers, the plants from central
Baja California can be referred only to var. flavovirde.

Eriogonum fasciculatum Benth. var. emphereium Reveal. n = 20.
Vouchers: M&R 19660 (Fig. 10), Picachos de Santa Clara, ca 350 m,
B.Cfa.Sur (near 27°09’N, 113°40’W), 3 Feb 1973; MG&R 19690 (Fig.
11), north slope of SE peak, Picachos de Santa Clara, ca 475 m, B.Cfa.
Sur (near 27°07’N, 113°37’W), 3. Feb 1973.

This recently proposed variety (Reveal, 1976) differs in its large
flowers which may perhaps show past influence of gene flow by Evriogo-
num pondiu Greene. No chromosomal abormalities were noted in the sev-
eral buds examined.

Eriogonum fastigiatum Parry. n = 20. Voucher: M&R 20282 (Fig.
14), Colnett Mesa, 0.5 mi N of Colnett, ca 80 m, B.Cfa.Norte (near
31°05’N, 116°13’W), 25 Feb 1973.

Eriogonum inflatum Torr. & Frém. var. deflatum I, M. Johnston.
n = 16. Vouchers: M&R 19610 (Fig. 15), Arroyo San Francisquito,
4 mi NW of Las Arrastras, ca 350 m, B.Cfa.Norte (near 29°36’N,
114°26’W), 31 Jan 1973; M&R 20037 (Fig. 16), grade near Lucifer,
ca 110 m, B.Cfa.Sur (near 27°23’N, 112°24"W), 11 Feb 1973.

This count agrees with previously reported counts (Reveal, 1967).

Eriogonum intricatum Benth. n = 16. Vouchers: M&R 19755 (Fig.
17), 8 mi NW of Asuncién, ca 70 m, B.Cfa.Sur (near 27°13’N,
114°21’W), 4 Feb 1973; M&R 19894 (Fig. 18), Arroyo Malarrimo, 11
mi S of the mouth, ca 75 m, B.Cfa.Sur (near 27°39’N, 114°29’W), 6
Feb 1973.

This Baja California endemic belongs to the Eriogonum inflatum
complex and has the same chromosome number as F. inflatum.

Eriogonum moranti Reveal. n = 20. Voucher: M&R 19611 (Fig. 19),
Arroyo Calamajué, ca 330 m, B.Cfa.Norte (near 29°24’N, 114°14’W),
1 Feb 1973.

This count agrees with that published previously (Reveal, 1968).

Eriogonum pilosum S. Stokes. n = 16. Voucher: M&R 20214 (Figs.
20, 21), Arroyo de la Purificacién, ca 500 m, B.Cfa.Norte (near 28°10’N,
113°15’W), 19 Feb 1973.

Shreve and Wiggins (1964) included this species with Eriogonum
scalare S. Wats., and their description of E. scalare applies in large part
to E. pilosum, while the description of EL. pilosum applies mainly to E.
vepens (S. Stokes) Reveal (Reveal, 1976). The leaves of EF. scalare are
elliptic, nearly glabrous, 5-15 mm long, and 2-6 mm wide, whereas those
of FE. pilosum are oblanceolate to oblong, wavy-margined, pilose 0.8—3
cm long, and 4—8 mm wide.

Eriogonum pondi Greene. n = 20. Vouchers: M&R 19753 (Fig. 22),
8 mi NW of Asuncion, ca 70 m, B.Cfa.Sur (near 27°13’N, 114°21’W),

1977 | REVEAL & MORAN: CHROMOSOME COUNTS 251

4 Feb 1973; MGR 19932 (Figs. 23, 24), bed of Arroyo Largo, 6.7 mi E
of the mouth, ca 175 m, B.Cfa.Sur (near 27°36’N, 114°43’W), 8 Feb
1973.

In 1967, Reveal and Hanson proposed var. gentryi for the mainland
plants of this species, except for the strictly coastal populations, with var.
pondii restricted to Cedros and Natividad Islands and the Turtle Bay
area. After further study, the apparent differences upon which var.
gentryi were based now seem too trivial for varietal distinction, and
therefore it should be considered a synonym of £. pondit.

Eriogonum preclarum Reveal. n = 20. Vouchers: M&R 19814 (Fig.
25), Arroyo de Portezuelo, 9.5 mi S of San José de Castro, ca 275 m,
B.Cfa.Sur (near 27°26’N, 114°27’W), 5 Feb 1973; M&R 19955 (Fig.
26), bed of Arroyo de las Casitas, near the mouth, ca 10 m, B.Cfa.Sur
(near 27°29’N, 114°38’W), 9 Feb 1973.

The type of this newly proposed species (Reveal, 1976), M&R 19964,
was also counted and determined to be x = 20, but is not figured.

Eriogonum repens (S. Stokes) Reveal. 2 = 16. Vouchers: M&R
19640 (Fig. 27), 1 mi N of San Angel, ca 50 m, B.Cfa.Sur (near 27°15’N,
113°10’W), 2 Feb 1973; MG@&R 20135 (Figs. 28, 29), Valley de Tortuga,
ca 80 m, B.Cfa.Sur (near 26°35’N, 113°50’W), 15 Feb 1973.

Eriogonum thurberi Torr. n = 20. Voucher: M&R 20246 (Figs. 30-
32), 2 mi NW of Desengafo, ca 580 m, B.Cfa.Norte (near 29°0O8’N,
114°05’W), 23 Feb 1973.

Streptanthus arizonicus S. Wats. n = 14. Voucher: M&R 20159
(Figs. 33, 34), north slope of Cerro Azufre, ca 1250 m, B.Cfa.Sur (near
27°30’N, 112°36’W), 17 Feb 1973; identified by Reed Rollins.

This collection represents a substantial southern range extension for
the species, otherwise known from southern Arizona. M&R 20192
(2 = 14) was also counted but is not illustrated.

Wislizenia refracta Engelm. in Wisliz. n = 20. Voucher: M&R 20096
(Figs. 35, 36), Pozo de San Juanico, ca 25 m, B.Cfa.Sur (near 26°14’N,
112°28’W), 14 Feb 1973.

This seems to be the first count for the genus. The specimens are
atypical of the species in being perennials, with woody stems to 1.5 cm
thick at the base; but they seem not to differ otherwise from the more
typical annual phase of the species. We collected the perennial form
(M&R 20153) also at San Zacarias, where it is known as “Joaquito”;
tea made from the leaves is said to be used internally and externally for
scorpion stings.

Astragalus magdalenae Greene var. magdalenae. n = 11. Voucher:
M&R 19718 (Figs. 37-39), 4 mi NE of Abreojos, ca 5 m, B.Cfa.Sur
(near 26°45’N, 113°34’W), 4 Feb 1973; identification confirmed by
Rupert C. Barneby.

Dalea bicolor H.B.K. (unpublished new variety to be named by
Barneby; pers. comm.). 7 = 7. Voucher: M&R 20158 (Figs. 40, 41),

SEE Se
N
= aces"
o) ee
-
—
\
so 2&6?
*ae . ne
*
on
~o
ootte
was
quae?
© She ~
v7, ras 2
O
w ike
Q
=
8
SOS
e@
se,
Sovee
ereore
os

N

°%
Col enuf e®
440 .~ wee?
a Wiley, @ @ a
7 4 3 Y = =
o,te%e
wi385 sotusss,
pt ie at
v 6 vee Soo
@
ag
49 2,
eee
~ e 22% Fe
a
e
6¢— q' % =
a
a
ie Cz ' ae
DvD 2
wy
2
OR
On
Is i de
a. = ° ow

|

|
|

£9771. REVEAL & MORAN: CHROMOSOME COUNTS 233

northeast ridge of Cerro Azufre, ca 850 m, B.Cfa.Sur (near 27°31’/N,
113°34’W), 17 Feb 1973; identification confirmed by Rupert C. Barneby.

This new form of Dalea bicolor will be named in honor of C. R. Orcutt
in the near future.

Errazurizia benthami ( Brandegee) I. M. Johnston. m = 14. Voucher:
MG&R 19865 (Fig. 42), Arroyo Malarrimo, 9 mi S of the mouth, ca 60 m,
B.Cfa.Sur (near 27°41’N, 114°28’W), 6 Feb 1973.

This appears to be the first report for the genus.

Lupinus arizonicus (S. Wats.) S. Wats. var. barbatulus (Thornber )
I. M. Johnston. = 24. Voucher: M&R 19845 (Figs. 43-46), Malar-
rimo, ca 10 m, B.Cfa.Sur (near 27°48’N, 114°27’W), 6 Feb 1973; iden-
tification by David Dunn.

Castilleja aff. lanata A. Gray. Vouchers: M&R 20249 (n = 48, Fig.
47), north slope of volcanic hill at top of Jaraguay Grade, ca 875 m,
B.Cfa.Norte (near 29°37’N, 114°37’°W), 24 Feb 1973; M&R 20208
(n = 60, Fig. 48), north slope of Volcan las Tres Virgenes, ca 1325 m,
B.Cfa.Sur (near 27°29’N, 112°36’W), 18 Feb 1973; identifications by
Noel H. Holmgren.

Holmgren reports that these two collections approach Castilleja lanata
of southern Arizona eastward to Texas, but may be taxonomically dis-
tinct and deserve some formal recognition. In addition to the two ploidy
levels reported here (7 = 48, 60), Lawrence R. Heckard has found an
even lower ploidy level which he will report.

Orobanche cooperi (A. Gray) Heller. n = 12. Voucher: M&R 19933
(Fig. 49), Arroyo Largo, 6.7 mi E of the mouth, ca 175 m, B.Cfa.Sur
(near 27°36’N, 114°43’W), 8 Feb 1973.

Heckard and Chuang (1975) have reported Orobanche cooperi from
Baja California, Mexico, and California, with counts of 27 = 48, 72, and
96. Apparently, our count is the first diploid population reported for this
species, and the first species of New World Orobanche with a haploid

number of 12. Our plants differ from typical O. coo peri in having glabrous
anthers.

Fics. 30-64. 30. Eriogonum thurberi, n = 20, metaphase I; 31, telophase I; 32,
telophase II. 33. Streptanthus arizonicus, n = 14, anaphase I; 34, metaphase I. 35.
Wislizenia refracta, n = 20, metaphase I; 36, telophase I. 37. Astragalus magdalenae
var. magdalenae, n = 11, metaphase I; 38, one complement of anaphase I; 39, telo-
phase II. 40. Dalea bicolor var. ined, n = 7, metaphase I; 41, anaphase I. 42. Erraszu-
rizia benthamii, n = 14, telophase I. 43, 44. Lupinus arizonicus var. barbatulus,
n — 24, metaphase I; 45, telophase I; 46, telophase II. 47. Castilleja aff. lanata,
n — 48, metaphase I; 48, 2 = 60, metaphase I. 49. Orobanche cooperi, n = 12,
anaphase I. 50. Salvia similis, n = 12, telophase I. 51. Encelia palmeri, n = 17,
metaphase I. 52. Encelia stenophylla, n = 17, diakinesis; 53, metaphase; 54, late
anaphase I. 55. Greenella ramulosa, n = 4, diakinesis; 56, telophase I; 57, telophase
Il. 58. Machaeranthera crispa, n = 5, diakinesis; 59, metaphase I. 60, 61. Porophyl-
lum tridentatum, n= 15, metaphase I. 62. Viguiera lanata, n = 17, diakinesis:
63, anaphase I; 64, telophase IT.

234 MADRONO [Viokws

Salvia similis Brandegee. n = 12. Voucher: M&'R 20108 (Fig. 50),
in an arroyo 3 mi SW of Cadejé, ca 50 m, B.Cfa.Sur (near 26°20'N,
112°24’W), 14 Feb 1973.

Encelia palmeri Vasey & Rose. n = 17. Voucher: M&R 20003 (Fig.
51), Arroyo Calvario, 6 mi N of San Andrés, ca 130 m, B.Cfa.Sur (near
27°20’N, 114°26’W), 10 Feb 1973.

Encelia stenophylla Greene. n = 17. Voucher: M&R 19678 (Figs. 52-
54), north slope of SE peak, Picachos de Santa Clara, ca 350 m, B.Cfa.
Sur (near 27°O7’N, 113°37’W), 3 Feb 1973.

Shreve and Wiggins (1964) reported this species for Cedros Island but
not for peninsular Baja California. Our collections show it fairly common
not only at Picachos de Santa Clara but also from Turtle Bay to Arroyo
Malarrimo and to south of San José de Castro, from about 25 to 850
meters elevation.

Greenella ramulosa Greene. n = 4. Voucher: M&R 19926 (Figs. 55-
57), southeast side of Bahia Tortugas, ca 10 m, B.Cfa.Sur (near 27°39’N,
114°51’W), 8 Feb 1973.

This perennial species is restricted to the west coast of central Baja
California. The only other species of the genus is an annual, found in
Arizona, which likewise is 2 = 4 (Solbrig et al., 1964). The proposal by
Ruffin (1974) to reduce Greenella to Xanthocephalum is not adopted
here.

Machaeranthera crispa (Brandegee) Turner & Horne. 7 = 5. Voucher:
M&R 20154 (Figs. 58, 59), San Zacarias, ca 170 m, B.Cfa.Sur (near
27°08’N, 112°56’W), 15 Feb 1973.

This count is consistent with the somatic number of 2” = 10 given by
Turner and Horne (1964).

Porophyllum tridentatum Benth. nm = 15. Voucher: M&R 19965
(Figs. 60, 61), Arrovo de las Casitas, 1.5 mi above the mouth, ca 50 m,
B.Cfa.Sur (near 27°30’N, 114°36’W), 9 Feb 1973; confirmed by John
L. Strother.

Johnston (1965) published a count of 2m = 30 for what he called
Porophyllum tridentatum var. crassifolium; but that is a valid species in
the opinion of John L. Strother (pers. comm.). Shreve and Wiggins
(1964) reported P. tridentatum only from the Magdalena Plain and
adjacent islands, some 325 kilometers to the southeast.

Viguiera lanata (Kellogg) A. Gray. 1 = 17. Voucher:M&R 19970
(Figs. 62-64), Arroyo de las Casitas, 8 mi above the mouth, ca 300 m,
B.Cfa.Sur (near 27°33’N, 114°35’W), 9 Feb 1973.

Shreve and Wiggins (1964) reported this species as known only from
Cedros and Natividad Islands, but they expressed the opinion that
it should also occur on the mainland. Our collections confirm that
expectation.

SL

1977] REVEAL & MORAN: CHROMOSOME COUNTS 23%

ACKNOWLEDGMENTS
Field support was provided by National Science Foundation grant GB-22645,
and publication costs were partially provided by BMS 75-13063. We are grateful
to C. Rose Broome for her comments upon the manuscript.

LITERATURE CITED

BoLKHOVSKIKH, Z. V. et al. 1969. Chromosome numbers of flowering plants.
Leningrad.

Hecxarp, L. R. and T. I. CHuanc. 1975. Chromosome numbers and polyploidy in
Orobanche (Orobanchaceae). Brittonia 27:179-186.

Jounston, R. R. 1965. In: Documented chromosome numbers of plants. Madrono
18:122-126.

ReveaL, J. L. 1967. In: Documented chromosome numbers of plants. Madrofo
19:134-136.

. 1968. Notes on Eriogonum-IV. A revision of the Eviogonum deflexum
complex. Brittonia 20:13-33.

. 1976. Eriogonum (Polygonaceae) novelties from Baja California, Mexico.
Brittonia 28:337-340.

and C. A. Hanson. 1967. Two new eriogonums from Baja California,

Mexico. Madrono 19:55-57.

and P. A. Munz. 1968. “Eriogonum,.” In: P. A. Munz, Supplement to A
California Flora. Berkeley, pp. 33-72.

and R. SPELLENBERG. 1976. Miscellaneous chromosome counts of western
American plants—III. Rhodora 78:37-52.
and E. L. Sryer. 1973. Miscellaneous chromosome counts of western
American plants—II. Great Basin Naturalist 33:19-25.
. 1974. Miscellaneous chromosome counts of western American plants—I.
Southw. Naturalist 18:397—402.

Rurfrin, J. 1974. A taxonomic re-evaluation of the genera Amphiachyris, Amphi-
pappus, Greenella, Gutierrezia, Gymnosperma, Thurovia, and Xanthocephalum
(Compositae). Sida 5:301-333.

SHREVE, F. and I. L. Wiccrns. 1964. Vegetation and flora of the Sonoran desert.
Stanford Univ. Press.

Sorsric, O. T. et al. 1964. Chromosome numbers in Compositae V. Astereae II.
Amer. J. Bot. 51:513-520.

STEBBINS, G. L. 1942. Polyploid complexes in relation to ecology and the history of
floras. Amer. Naturalist 76:36—-45.

STOKES, S. G. and G. L. Stepprns. 1955. Chromosome numbers in the genus Eriogo-
num. Leafl. W. Bot. 7:228-233.

Turner, B. L. and D. Horne. 1964. Taxonomy of Machaeranthera sect. Psilactis
(Compositae, Astereae). Brittonia 16:316-331.

FOUR NEW SPECIES OF CENTAURIUM (GENTIANACEAE)
FROM MEXICO

C. Rose BRrRooME
Department of Botany, University of Maryland, College Park 20742

Work on a taxonomic revision of the New Work members of Cen-
taurium (Gentianaceae) has resulted in the recognition of four previously
undescribed species from Mexico. Three are known to me only from
herbarium specimens, but living material of a fourth has been collected
and observed.

236 MADRONO [Vol. 24

Centaurium gentryi Broome, sp. nov. Herba annua parva caule sim-
plici ramis paucis, rosula basali, foliis caulinis setaceis, et floribus paucis
pentameris lobis corollae tubo longioribus. A C. madrensi foliis basalis
multo majoribus persistentibus, nodis caulinis foliosis paucioribus, flori-
bus majoribus differt. Fig. 1a, b.

Type: Mexico, Chihuahua, Rancho Byerly, Sierra Charuco (approx.
27°30’N, 108°40’W), on rocky igneous slopes in pine-oak forest, 1525-
1770 m, 17-25 Apr 1948, Gentry 8035 (Holotype: UC; isotypes: DS,
MEXU, MICH, US).

Plants annual with erect simple stems 11—26 cm tall, branched mostly
above the middle with only ca 3 elongate internodes, 10-55 mm long,
beneath the first branch. Leaves strongly dimorphic, the basal 3—6 pairs
in a rosette, these lance—ovate to broadly ovate, 5-12 mm long, 3—6 mm
broad, the cauline leaves linear or subulate, 3-12 mm long and 0.5—1 mm
broad, appressed to the stem. Inflorescence a simple paniculate or race-
mose cyme, sparingly branched, with the branches ascending at angles
of 20-30° from the main axis, often uniflorous; pedicels of central flowers
in a dichasium 46—56 mm long, but as short as 6 mm on lateral flowers.
Flowers pentamerous, usually 1-6 per plant. Calyx 4.5—9.2 mm long, the
lobes 4.2—-8.7 mm long, subulate—attenuate and moderately keeled, with-
out prominent hyaline margins, about equally the corolla—tube. Corolla
vivid rose—pink without a prominent white eye, (11—) 18—-22.5 mm long;
corolla-tube (4—) 7—8.2 mm long, not constricted above ovary, the top
slightly flared into a throat; corolla—lobes (7—) 12-15 mm long, (3.5—)
5.5-6.5 mm broad, lance—ovate, obtuse at tips. Stamens 4.5—7 mm long,
inserted just below summit of ovary, shorter than or equalling style at
anthesis; anthers linear before anthesis, 1.2-2 mm long when coiled;
pollen grains 23.8-40.8 um in diameter. Ovary 4.2—7.5 mm long at
anthesis; style exserted, 4.6-7 mm long, the stigma bilobed with lobes
slightly broader than long. Capsule oblong, not resinous, ca 8 mm long
and 3 mm broad; seeds reddish-brown, 0.3 mm long.

The Rio Mayo region of western Mexico has yielded a number of plant
novelties, but the area has been little collected since Gentry’s expeditions
in the 1930s and 1940s. This very distinctive Rio Mayo Centaurium is
accorded the rank of species, in spite of the fact that it is known only
from a single collection. Its large, showy flowers with long exserted styles
are reminiscent of those of C. chironioides (Griseb.) Druce, but in its
habit, inflorescence, and leaf morphology it more resembles C. madrense
(Hemsley) B. L. Robinson, a species of the Sierran region of Sinaloa,
Durango, Nayarit, and Jalisco. It is probably more allied to those two
Mexican species than to the C. calycosum (Buckley) Fern. complex of
the United States and bordering states of Mexico. It differs from C.
madrense by having much larger basal leaves, fewer leaf—bearing nodes
beneath the inflorescence, broader and shorter calyx—lobes, and conspicu-
ously larger corollas.

1977 | BROOME: CENTAURIA 237

Fic. 1. Centaurium gentryi. a. habit (x0.5), b. flower (x1.2)—based on Gentry
8035; and C. pterocaule, c. habit (x0.3), d. flower (x1.2)—based on Smith M81.

Centaurium pterocaule Broome, sp. nov. Herba annua, caule erecto
alato, floribus paucis pentameris magnis, lobis corollae tubo longioribus.
Centaurio chironioide affinis, caule et videtur simplici, alis usque ad 1 mm
latis, undulatis, foliis inferioribus obovatis latioribus differt. Fig. Ic, d.

Type: Mexico, Hidalgo, Zimapan, Coulter 941 (Holotype: K; iso-
type: GH).

Plants annual with erect, simple stems 23-39 cm tall, relatively stout
and prominently 4—winged, the wings on the lower stem ca 1 mm wide
and ruffled. Leaves cauline, 9-33 mm long, 6-13 mm wide, obovate,
obscurely 3—nerved, obtuse—-tipped at lower nodes, and becoming nar-
rower and shorter upwards, narrowly obovate to oblanceolate with acute
or acuminate tips, reduced to lanceolate bracts in the inflorescence. Stem—

238 MADRONO [Vol. 24

wings and leaf margins blue-black upon drying. Inflorescence a simple
determinate panicle with both dichasial and monochasial branching, the
branches up to 20 cm long and leafy, only once or twice compound or
uniflorous; pedicel of central flower of a dichasium 6—17 mm long, but
as short as 2 mm on lateral flowers. Flowers pentamerous, showy, usually
4—10 per plant. Calyx 6-8 mm long, the lobes 4—-5.5 mm long, narrowly
triangular—attenuate to narrowly subulate—acuminate, about equally the
corolla—tube, the hyaline margins narrow. Corolla apparently deep rose
without white eye, 16-23 mm long; corolla—tube 6—9 mm long, slightly
exceeding and constricted above the ovary; corolla—lobes 10-14 mm long,
4—7.5 mm wide, ovate or lance-ovate with acute or slightly acuminate
tips. Stamens 4.5—5.5 mm long, inserted in corolla—tube just below the
summit of the ovary and nearly equalling the stvle at anthesis; anthers
linear—oblong before anthesis, 1.6—2.2 mm long when coiled; pollen grains
26.6—29.4 um (Carlson 3148) or 33.9-46.2 um (Smith M81) in diameter.
Ovary oblong, thick-walled, ca 7-7.5 mm long; style exserted from co-
rolla—tube, 5—7 mm long, the stigma bilobed with lobes slightly broader
than long. Mature capsule not seen; partially mature one 7.6 mm long
and 3.4 mm broad, oval with rounded apex; seeds not seen.

Mexico, in states of Morelos, Hidalgo, and San Luis Potosi, in mon-
tane pine forests at 1800-2500 m. Flowering specimens have been col-
lected in January and April.

Additional Collections: MEXICO. Moretos: in pine forest on Old Mexico City
Highway, 16 Apr 1960, Smith M81 (TEX); between Cuernavaca and Mexico City
on old road, km 62, 2070 m, 27 Jan 1955, Carlson 3148 (DUKE). San Luts Potosi:
Cerro de la Silleta, near Xilitla, Paray 369 (MEXU).

Considerable variation exists among the four collections that | have
included in this taxon. The Carlson specimen has strikingly large, broadly
obovate leaves at the lower nodes reminiscent of Schultesia lisianthoides
(Griseb.) Benth. & Hook. ex Hemsley or Centaurium strictum (Schiede)
Druce. The broad lower leaves are absent in Coulter 941 and Smith M81.
The inflorescence is narrowest and shortest in Carlson 3148, but more
open and divaricate with few branches in the other specimens. All speci-
mens share the distinctive characters of the broad stem—wings, sharply
keeled calyx—lobes, and large flowers. They also display a bluish color in
the vegetative organs similar to that found in C. chironioides and C.
pauciflorum (Mart. & Galeotti) B. L. Robinson, two other Mexican
species probably closely related to C. pterocaule.

The occurrence of these few specimens over so wide a geographic range
within central Mexico suggests a species which at an earlier time has been
more abundant. Undoubtedly the widespread destruction of forests in
this most populated area of Mexico has drastically reduced the popu-
lations of this species. My several searches from 1969 to 1975 were
unsuccessful.

1977] BROOME: CENTAURIA 239

Centaurium wigginsii Broome, sp. nov. Herba annua caule simplici
foliis caulinis obovatis oblongis decussatis, ramis divaricatis superne,
floribus parvis pentameris roseis. Centaurio stricto ubique affinis, sed inflo-
rescentia apertiore bracteis foliaceis, pedicellis longioribus, corollis majori-
bus roseis profundioribus, capsulis majoribus differt. Fig. 2.

TypE: Mexico, Sinaloa, Highway 40 (Mazatlan-Durango highway),
34.3 mi E of Concordia, growing with Centaurium nudicaule and C. qui-
tense on wet roadbank with oak and pine above, 1770 m, 17 Feb 1971,
C. R. Broome 763 (Holotype: DUKE; isotypes: K, MEXU, NY, UC,
Us).

Plants annual with erect, simple stems (8.5—) 12—25 (—43) cm tall,
branched from above middle or sometimes from base. Leaves cauline,
markedly decussate, green, not in a definite basal rosette but clustered at
lower nodes and longer than the contracted internodes, becoming more
remote upward and shorter than the elongated upper internodes; leaf-
blades obovate or oblong, the lower (5—) 12-22 mm long, (2—) 5-8 mm
wide, the margins upturned and becoming narrower, (ob—) lanceolate
above with flattened margins. Inflorescence open, branches rather divari-
cate, these branched again only once or twice, monochasially or dichasi-
ally; pedicels 0.5-10 (—19) mm long. Flowers pentamerous, (1—) 5-30
per plant. Calyx 3-5.3 mm long, the lobes lance—acuminate with well—
developed hyaline margins, 2.5-5 mm long, about equaily the corolla—
tube. Corolla salverform, rose—pink with white eye, (6—) 8-10 mm long;
corolla—tube (3-) 5-7 mm long, not constricted above ovary; corolla—
lobes 3.4—4 mm long, 1.7—2.3 mm broad with rounded and slightly erose
tips. Stamens 2.5—3.6 mm long, inserted just below summit of ovary and
equalling or exceeding the style; anthers oblong to sagittate, 1.2-1.8 mm
long before dehiscence; pollen grains 20.9-29.4 um in diameter. Ovary
4.2-5 mm long, bearing 2—3 rows of ovules on each carpel margin; style
exserted from corolla—tube, 1.6-3 mm long, the stigma bilobed with the
lobes flabellate. Capsule fusiform, 6.5-9 mm long, 1.8-3 mm broad;
seeds ca 0.3 mm long, reddish-brown, ca 250 per capsule. Chromosome
number: n = 22.

Known only from the Sierra Madre Occidental west of the summit of
the high ridge of El Espinazo del Diablo along Mexico Highway 40, at ca
1600-2200 m, in Sinaloa and Durango, and one station in northern
Nayarit. Moist, partly shaded steep clay banks of the oak—pine zone.
Flowering specimens have been collected in February, March, and April.

Additional Collections: MEXICO. Duranco: ca 2 mi SW of Revolcaderos enroute
to Mazatlan on Hwy 40 on foot trail descending to rocky stream with oak—pine
woods above, ca 2200 m, 26 Mar 1975, Almeda 2529 (DUKE) ; along the Mazatlan-
Durango highway 3-15 km toward El Salto from the Sinaloa boundary at El
Palmito, 1950-2200 m, 13 Apr 1965, McVaugh 23591 (MICH). Stnatoa: Hwy 40
ca 4.7 mi N of El Carrizo in oak-pine forest zone at 1950 m, 26 Mar 1975, Almeda
2526 (DUKE) ; dry hillside among pines and oak 49 mi E of Villa Union, 1630 m,
18 Mar 1955, Wiggins 13179 (DS). Nayarit: vicinity of Acaponeta, 9 Apr 1910,
Rose, Standley & Russell 14273 (F, GH, NY).

240 MADRONO [Vol. 24

we
EO

Fic. 2. Centaurium wigginsii. a. habit (x0.9), b. a lower leaf (x1.3), c. flower in
top view (x2.5), d. flower in side view (x2.5), e. capsule with marcescent corolla
(x2.5)—based on Broome 763.

1977] BROOME: CENTAURIA 241

The species is named for Professor Ira L. Wiggins, distinguished stu-
dent of the Mexican flora, who made one of the earliest collections.

Centaurium wigginsii is related to C. strictum, a Mexican endemic
found no farther north than Jalisco, in the Nueva Galicia region. The
two species are quite similar in leaf morphology and habit, but differ sig-
nificantly in the following characters:

C. wigginsi C. strictum
Branching angle divaricate strict
Inflorescence branches with few conspicuously
inconspicuous bracts __ leafy
Pedicel of terminal flower 3.5-19 mm 1-6 mm
Corolla color bright rose—pink whitish to bluish
or pale—pink
Corolla length (6—) 8-10 mm 5—7.7 mm
Chromosome number n= 22 n= 21

Experimental hybridization of several American species of Centaurium
with C. wigginsii has confirmed that C. strictum is genetically similar as
well (Broome, 1973). The F, hybrids in crosses between the two had up
to 70% pollen stainability and produced vigorous F,.’s. Significant (de-
fined as greater than 1% pollen stainability in the F, hybrid) affinities
exist between C. wigginsi and only two other Mexican species as indi-
cated by pollen stainability of the F,’s: C. quitense (18-38%) and C.
nudicaule (0.5-2.5%). It is my belief that C. wigginsii was derived
directly from C. strictum by aneuploid increase, in the northernmost part
of the range, followed by geographical isolation of the two species. Cen-
taurium wig ginsii is now sympatric with C. nudicaule in the same habitat,
and is geographically but apparently not populationally sympatric with
C. setaceum (Benth.) B. L. Robinson.

Centaurium capense Broome, sp. nov. Herba annua caule simplici,
ramosissimo e basi, foliis caulinis anguste ovatis, floribus numerosis par-
vis roseis, lobis corollae tubo brevioribus, stigmate bilobo. Centaurio flori-
_bundo, C. tenuifloro, C. pulchello floribus similis, a his speciebus pedi-
cellis longis, rosulis basalibus nullis, et ramificatione divaricata saepe
trichotoma differt. Fig. 3.

_ Type: Mexico, Baja California Sur, along stream below Santiago,
_ between Santiago and Rivera, Cape Region, 6 May 1931, Wiggins 5665
_ (Holotype: US; isotypes: DS, F, GH, MICH, UC).

_ Plants annual with erect, simple stems 7.5-47 cm tall, branched at
most nodes. Basal rosette of leaves lacking, but the lower internodes often
| contracted and the lower leaves clustered, the upper 3—6 cauline inter-
nodes to 63 mm long and often much longer than the subtending leaves.
_ Leaf-blades thin, (ob—) lance-ovate to lanceolate, 12-32 mm long, 1.5-
_ 12 mm wide, the largest ones borne at the middle nodes of the stem, much
_Teduced on lateral branches, the larger leaves with 5 main nerves. Inflo-

MADRONO LVol. 24

Ls)
Aes
bo

Fic. 3. Centaurium capense. a. habit (x0.5), b. flower in side view (x2), c. flower |
in top view (x2), d. style and stigma (x14.5), e. anther after dehiscence (x10.5),
f. mature capsule, marcescent corolla removed (x2)—based on Wiggins 5608.

rescence a broad, divaricately branched compound cyme with the main —
branches arising from all but the lowermost one or two cauline nodes
and diverging from the main axis at angles of 40—-60°, these branched |
again 3—4 times, the branching mainly monochasial but with a high fre-
quency of dichasial (trichotomous) branching at all levels; pedicels of |
central flowers in dichasia 5—20 mm long, usually shorter than the flowers |
but slightly longer on lateral flowers. Flowers pentamerous, 50 to several

1977 | BROOME: CENTAURIA 243

hundred per plant. Calyx 5.5-10 mm long, the lobes 5-10 mm long, fili-
form, shorter than the corolla—tube, with inconspicuous hyaline margins.
Corolla light to deep pink with small white eye, (9—) 10-16 mm long;
corolla-tube (6—) 7-12.5 mm long, exceeding ovary by ca 2 mm and
slightly constricted above it; corolla—lobes 3—4.7 mm long, 1-2 mm wide,
lanceolate with blunt and minutely erose tips. Stamens inserted just
above summit of ovary, exceeding style; anthers oblong, 0.7-1.5 mm
long after anthesis, frequently bent over on filaments and shedding pollen
directly onto stigma beneath; pollen grains 21.8—28.5 4m in diameter.
Ovary 5.2—8.5 mm long, ca 1 mm wide, bearing 2—3 rows of ovules on
each carpel margin; style included or slightly exserted from corolla—tube,
deeply divided with stigma—lobes ovoid, longer than broad. Capsule nar-
rowly cylindrical, 9-11 (—12.5) mm long, 1.5—2 mm broad; seeds 0.15-
0.2 mm long, deep reddish—brown.

Known from Baja California Sur only in the Cape Region in the Lagu-
nan Woodland biome (Axelrod, 1958) of the mountains just southeast
of La Paz and southward. In moist, sandy soil along streams and in
washes. Flowering specimens have been collected from March through
early May.

Additional Specimens: MEXICO. Baya CaLirorniA Sur: 22 mi S of La Paz, 8
May 1931, Wiggins 5608 (DS); La Huerta, 19 May 1889, Brandegee sn. (UC);
ca 6 mi SW from Santiago, Arroyo San Mateo, 30 Apr 1959, Thomas 7722 (DS);
along stream below Santiago between Rivera and Santiago, 6 May 1931, Wiggins
5660B (DS) ; Santiago, 31 Mar 1936, Bailey 176 (F); El Reparito, S fork of Cafion
San Pedro, ca 770 m, 23°20’N, 109°55’W, 8 May 1959, Moran 7361 (DS); Arroyo
Culebriado near junction of trails to Rancho la Fragua and Rancho Sauce (SE of
Cerro Giganta), ca 400 m, 26°3.5’N, 111°28.5’W, 31 Mar 1960, Carter & Ferris 4065
(UC) ; Arroyo del Salto, E of La Paz, in moist granitic sand under palms, 24°12’N,
110°7.5’W, 30 Mar 1949, Carter 2576 (UC); San José del Cabo, 2 Apr 1892, Bran-
degee sm. (UC); San Lazaro Canyon, ca 100 m, 23°08’N, 109°48’W, 2 May 1959
Moran 7329 (DS); Potrero de Almenta, Arroyo de Almenta, E slope of Sierra de la
Victoria inland from Caduano, 1036 m, 9-11 May 1959, Thomas 7825 (DS, GH, US).

This interesting species has been distributed as Centaurium exaltatum
(Griseb) W. F. Wight or as C. nudicaule, two other small—flowered
species which also occur on the Baja California peninsula. On closer

inspection, the new species was found to differ from those taxa in several
characters:

C.exaltatum —C. capense C. nudicaule
Stigma lobe shape _flabelliform ovoid flabelliform

or reniform
Style division shallow deep shallow
Corolla length 10-19 mm 10-16 mm 6.5-10 mm
Capsule length 10-15 mm 9-12.5 mm 6-9.5 mm
Capsule width 1.8-3.5 mm 1.5-2 mm 1.2-3 mm
Cauline leaf shape lanceolate jance—-ovate linear or

to lanceolate subulate

244 MADRONO [Vol. 24

Type of cyme narrow, rather broad narrow, branches
strict divaricate ascending

Central pedicel longer than shorter than _ longer than
flower flower flower

The origins of this species are obscure. My initial hypothesis that this
endemic was derived from migrant populations of Centaurium exaltatum
has been rejected on morphological grounds. The style is unusual as it is
distinctly branched beneath the papillose surface of the stigma lobes.
The lobes are devoid of papillae in the central portion of the inside sur-
face. These are features found, to my knowledge, only in those species
allied to C. minus Moench, including the California taxon, C. floribun-
dum (Benth.) B. L. Robinson. Besides C. floribundum and C. capense,
no other American species have been found to have this stylar mor-
phology. The very narrowly cylindrical ovary and capsule of both species
also ally them to C. minus. Centaurium capense, like C. floribundum, in
all probability has its closest relatives in Europe or Asia.

ACKNOWLEDGMENTS

This paper is based on a portion of a dissertation submitted to the faculty of Duke
University in partial fulfillment of the doctoral degree. I wish to acknowledge the
support of National Science Foundation Systematics Training grants GB-6393 and
GB-23200 during this research. The illustrations were prepared by Karen Teramura
(Fig. 1), Susan Carlton Smith (Fig. 2), and Lyn Loveless (Fig. 3). My thanks to
the curators of the following herbaria for allowing me to study their material: DS,
DUKE, F, GH, K, MEXU, MICH, NY, TEX, UC, US.

LITERATURE CITED
AXE Lrop, D. I. 1958. Evolution of the Madro-Tertiary geoflora. Bot. Rev. 24:433-
509. |

Broome, C. R. 1973. Systematics of Centaurium (Gentianaceae) of Mexico and |
Central America. Unpublished doctoral dissertation, Duke University, Durham, |

NC.

e
MIBO

INDEX TO VOLUME XXIV
Classified entries: botanical names (new names are in boldface) ; major subject
headings, including key words from titles; reviews. Incidental references to taxa and
taxa merely enumerated in lists or tables are not indexed. Names of authors followed
by titles of articles are listed alphabetically in Table of Contents, pp. iii-iv. (The

1977] INDEX 245

Editor thanks John L. Strother for his kindness in preparing this index.)

Allium: taxonomy, 24; chromosome
counts, 26; A. serra, 25

Angelica callii, 78

Announcements: cover 4 of number 3;
cover 4 of number 4

Anthocarp anatomy, Nyctaginaceae, 104

Asclepias solanoana, floral ecology, 159

Bebbia: taxonomy, 112

Biomass ratios, root:shoot, 215

Centaurium: new species, 235; C. Ca-
pense, 241; C. gentryi, 236; C. ptero-
caule, 237; C. wigginsii, 239

Chenopodium: relationships of C. fla-
bellifolium and C. inamoenum, 63

Chromosome counts, miscellaneous, 227

Chrysactinia: taxonomy, 129; C. sect.
Tagetifolia, 134

Claytonia: chromosome counts and tax-
onomy, 62; C. arenicola, 62; C. saxo-
sa, 62

Coastal prairie, northern California, 83

Coastal scrub, northern California, 18

Cymophora: taxonomy, 1; C. accedens,
5; C. hintonii, 2; C. venezuelensis,
190

Darmera, correct name for Peltiphyl-
lum, 68

Drypetes standleyi, 65

Epilobium: taxonomy, 6; E. foliosum,
6; E. minutum, 6

Flaveria mcdougallii, 13

Floral ecology, Asclepias, 159

Frost sensitivity, shrubs of California
and Chile, 74

Germination flap, Gramineae, 123

Harnackia: taxonomy, 129

Helianthus exilis-bolanderi complex, 177

Hemizonia: intersectional hybridization,
55

Hulsea: taxonomy, 48; H. vestita subsp.
gabrielensis, 53

Hybridella: taxonomy, 29; H. anthemi-
difolia, 32; H. globosa var. myrio-
phylla, 34

Hybridization:
193

Hemizonia, 55; Pinus,

I. effusa,
tenuifolia,

Ipomopsis:
143.1.
146

Ivesia rhypara, 224

Lescaillea: taxonomy, 129

Lilaeopsis masonii, 81

Linanthus: taxonomy, 36, 147; chromo-
some counts, 40; L. floribundus
subsp. glabrus, 45; L. jamauensis,
147; L. nuttallii subsp. pubescens,
43; L. nuttallii subsp. tenuilobus,
43; L. orcuttii, 150; L. pachyphyl-
lus, 44; L. uncialis, 151; L. viscai-
nensis, 152

Madronho: dates of publication for vol.
24, iv; information for contributors,
cover 3; reviewers of manuscripts for
vol. 24, p. 63 of vol. 25; table of con-
tents for vol. 24, iii

Mirabilinae, anthocarp anatomy, 104

Moran, R., introduced, 140

Navarretia: taxonomy, 155; N. fossalis,
155

Nyctaginaceae, anthocarp anatomy, 104

Peltiphyllum, name correction, 68

Peltophyllum caudatum, 71

Phacelia marshali-johnstonii, 212

Pinus: hybridization, 193; P. aristata,
193; P. balfouriana, 193

Point Reyes Peninsula, California, vege-
tation, 18

Resprouting, shrubs of California and
Chile, 74

Reviews: Abbott, I. A. and G. J. Hollen-
berg, Marine algae of California, 124;
Henrickson, J. and R. M. Straw, A
gazetteer of the Chihuahuan Desert
Region, 128; Hoare, M. E., The tact-
less philosopher. Johan Reinhold For-
ster (1729-1798), 127; Thiers, H. D.,
California mushrooms, a field guide to
the boletes, 191

Root:shoot biomass
and Chile, 74

Sea Ranch, California, vegetation, 83

Shrubs, California and Chile, 74

Vegetation analysis, coastal prairie, 83

taxonomy, 141;
guttata, 145; I.

ratios, California

Ee er ee eee

Membership in the California Botanical Society is open to individuals ($12.00 per
year, regular; $8.00 per year, student). Members of the Society receive MapRoNO
free. Institutional subscriptions to MaproNo are available ($14.00 per year).

Back issues of Madrofio are available at the following rates:

Vol. 1 (1916-1929) and Vol. 2 (1930-1934), each consisting of 17 numbers: $17.00
per volume, $1.00 per issue.

Vol. 3 (1935-1936) through Vol. 20 (1969-1970), each biennial, consisting of 8
numbers: $16.00 per volume, $2.00 per issue.

Vol. 21 (1971-1972) through Vol. 23 (1975-1976), each biennial, consisting of 8
numbers: $24.00 per volume, $3.00 per issue.

Vol. 24 (1977) et seq., one volume per year, each consisting of 4 numbers: $14.00
per volume, $3.50 per issue.

Applications for membership (including dues), orders for abet ene requests
for back issues, changes of address, and undelivered copies of MapRoNo should be
sent to the California Botanical Society, Inc., Department of Botany, University of
California, Berkeley 94720.

INFORMATION FOR CONTRIBUTORS

Manuscripts submitted for publication in MaproNo should be sent to the Editor.
Membership in the California Botanical Society is requisite for publication in
ManproNno.

Manuscripts and accompanying illustrative materials must be submitted in dupli-
cate and should follow the format used in recent issues of MaproNo. Original illus-
trations should not be submitted until paper is accepted for publication. All manu-
scripts MUST BE DOUBLE SPACED THROUGHOUT, including title, text, tables,
captions, lists, literature cited, etc. Footnotes (which should be avoided wherever
possible), captions, and tables must be typed on sheets separate from the text.
Presentation of nomenclatural matter (accepted names, synonyms, typification)
should follow the format used for Rhus integrifolia in MaproNo 22:288. 1974. All
measurements should be given in S. I. (metric) units. Where appropriate, scales
should be included on figures rather than in captions as estimates of ratio of repro-
duction such as X 1%, X 1620, etc. Institutional abbreviations in specimen citations
should follow Holmgren and Keuken’s list (Index herbariorum, Part 1, The herbaria
of the world. Sixth edition. 1974. Regnum Veg. vol. 92). Abbreviations of names of
journals should be those in Botanico-Periodicum-Huntianum (Lawrence, G. H. M.
et al. 1968. Hunt Botanical Library, Pittsburgh). If the correct abbreviation cannot
be determined, the full title of the journal should be used. Titles of books should be
given in full, together with the place and date of publication, names of publisher,
and an indication of the edition, if other than the first.

Short articles such as range extensions and other brief notes are published in con-
densed form under the heading “Notes and News”. Authors of such articles should
follow the format used in recent issues of MApDRONO.

Authors are allowed up to 10 pages per year without page charges; charge for
additional pages is $30.00 per page. Subject to approval by the Editors, articles may
be published ahead of schedule, as additional pages of an issue, provided the author
assumes complete costs of publication.

STATEMENT OF OWNERSHIP, MANAGEMENT, AND CIRCULATION
(Act of Oct. 23, 1962; Section 4369, Title 39, United States Code)

Madrono, A West American Journal of Botany, is published quarterly at Berke-
ley, California.

The Publisher is the California Botanical Society, Inc., Life Sciences Building,
University of California, Berkeley, California 94720.

The editor is Barbara D. Webster, Department of Agronomy and Range Science, 4
University of California, Davis, California 95616. i

The owner is the California Botanical Society, Inc., Life Sciences Building, Uni-
versity of California, Berkeley, California 94720. There are no bondholders, mort-
gagees, or other security holders.

The average number of copies distributed of each issue during the preceding 12 —
months is 946; the number of copies of the single issue closest to the filing date
is 917. ‘

I certify that the statements made by me above are correct and complete.

Barpara D. WEBSTER, Editor

September 1, 1977

PsILOTACEAE SYMPOSIUM REPRINTS AVAILABLE.—The recent symposium entitled —
“Taxonomic and Morphological Relationships of the Psilotaceae,” which was pre- _
sented during the 1976 AIBS meetings at Tulane University, will be published in —
the January-March 1977 issue of BRITTONIA. This symposium, under the chair- —
manship of Richard A. White (Duke University) and sponsored by the American —
Fern Society, the American Society of Plant Taxonomists, and the Pteridological, ©
Paleobotanical, Structural and Systematic Sections of the Botanical Society of —
America, summarizes the major points of the current controversy surrounding this —
enigmatic plant group.

Contributors are: David W. Bierhorst (University of Massachusetts), The sys-
tematic position of Psilotum and Tmesipteris; Patricia G. Gensel (University of —
North Carolina), Morphologic and taxonomic relationships of the Psilotaceae rela- —
tive to evolutionary lines in early land vascular plants; Donald R. Kaplan (Univer- —
sity of California—Berkeley), Morphopological status of the shoot systems of
Psilotaceae; and Warren H. Wagner, Jr. (University of Michigan), Systematic —
implications of the Psilotaceae. The set of papers, together with introduction and _
discussion, will be printed as a unit and become available in April as a separate
through the New York Botanical Garden. Copies can be obtained by sending your ~ |
order, along with $3.00 per copy ($2.50 each in quantities of 30 or more), to: |
BRITTONIA, New York Botanical Garden, Bronx, NY 10458. No one will be 4
filled unless accompanied payment.

SARS

gy we
HAM? Page)
need

tr

MADRONO

A WEST AMERICAN JOURNAL OF BOTANY

VOLUME XXV
1978

BOARD OF EDITORS

Class of:

1978—SHERWIN CarLQuIsT, Claremont Graduate School
Lesitige D. GotrLers, University of California, Davis
Dennis R. PARNELL, California State University, Hayward

1979—Puitie W. RunpDEL, University of California, Irvine
IsABELLE TAvarRES, University of California, Berkeley

1980—JAMEsS R. GrirFin, University of California, Hastings Reservation
Frank A. Lanc, Southern Oregon College, Ashland

1981—DanIeEL J. CRAWFORD, Ohio State University, Columbus
James Henrickson, California State University, Los Angeles

1982—Dean W. Tayvtor, University of California, Davis
RicHARD VocL, California State University, Los Angeles

Editor — James C. HICKMAN
Department of Botany, University of California, Berkeley 94720

Published quarterly by the
California Botanical Society, Inc.
Life Sciences Building, University of California, Berkeley 94720

Printed by Gillick Printing, Inc., Berkeley, California 94710

David Leech, 1972

Volume 25 of Madrono is dedicated to Professor G. Ledyard Stebbins,
Jr., in recognition of the exceptional scope and importance of his contri-
butions to evolutionary botany that have extended from his dissertation
work with Antennaria through the revolutionary 1938 study of Crepis
with E. B. Babcock and to the present time; in recognition of his exten-
sive contributions to the taxonomy, reproductive biology, ecology, and
floristics of California plants; and in recognition of his service to the
California Botanical Society, particularly as its President in 1964.

TABLE OF CONTENTS

Brriincs, W. D., Terrestrial vegetation of. California (review) ..........-.....2-1...2.2 174
Boum, Bruce A. and RoBert OrNDUFF, Chemotaxonomic studies in the Saxi-

fragaceae s1, 9. Flavonoids in Jepsonid. 4-8 ee ee 39
BroPHY, Wiliam. (see MacNeill, €{ Don)” 2 40 a a et ee 54
Brown, Roy Curtiss, Biosystematics of Psilostrophe (Compositae: Helenieae).

Lf. Artificial hybridization ‘and: systematic treatment)... ee 187 |
BULLOCK, STEPHEN H., Plant abundance and distribution in relation to types of

seed dispersal: in chaparral) 20220525 eee ee 104 |

CanneE, JupirH M., Circumscription and generic relationships of Galinsoga

M@emmpositacyseLelanUMne ae) s sc.cee. cesses sree et cus sac cs0 regs c eos tor ldee es sees eee
GARGER. SANNET@AS De Story Onpines, (TEVIEW) ..2%csisc.c!o2cccc fects ccsg ace seks <dzcesbcs eae cose
CiarK, Curtis, Pollen shed as tetrads by plants of Eschscholzia californica

(GL E21 yeh ich fe Wek 61 YS se ee ee me eno eens ee eee ere
(OEE HOWARD As See OLOKEI AROCLOR 2 cere. cise ut21.c oe Secs eigu dese eS Apiecereicacet dacs sess
CouLTER, MAtco_m C., Additicn to the flora of the Farallon Islands, California
Deucan, BryAN and Grapy L. WessTER, Three new species of Jatropha (Eu-

DOE Olaceacyminom Wesvert\VECKICOQ! 2.55: f oc 5 oases feces ees
Denton, Metrinpa F., Ranunculus californicus, a new record for the state of
AYERS) E9121 6) eae gue Nr ear reer eee coe eda ee ce P< he

Dorn, Ropert D., A new species of Draba (Cruciferae) from Wyoming and
LOT eB A a oe ee eo oe ee eran roe
Dorn, Rospert D., Great Basin vegetation in Carbon County, Wyoming ..............
Duncan, THomas, Ranunculus geranioides H. B. K. ex DC. in Costa Rica and
_ CEN IZ 00s Wa RH Re ee er ee ere eee pear
Ewan, JosEepH, Northwest botanical manuscripts (review) ..........-....:-2:::::ceeeseee0+=*
HIRO AN AN DARC . GSCCrIEO Mes OCA ica c28 sata Sebel ant chic ec tes ds ee Bata eee
FERLATTE, WILLIAM J., Notes on two rare, endemic species from the Klamath
Region of northern California, Phacelia dalesiana (Hydrophyllaceae) and
WROULOVOCH A Priggler (COMPOSILAC) sec eii nc. . doco test decadone aes Wee beset ater Se ae
ForcELLA, FRANK, Flora and chorology of the Pinus albicaulis — Vaccinium sco-

LICH ASS OGL ALL OMe collet ce RO aust ett Bn Br ested tes aces ache ncn ee me ee
FoRCELLA, FRANK, Irregularitiy of pinyon cone production and its relation to
plnyOMcone moth pPredaiOn’ .2.s.4 2 ocak saee see ee oe
FRYXELL, PAUL A., Gossypium turneri (Malvaceae), a new species from Sonora,
IVAN eG Ose te yee we Rie yl Ok 28 ee ss ten 2 ee ee teehes eat e a ak heed esa A Sees
GANDERS, FreD R. and HELEN KENNEDY, Gynodioecy in Mammillaria dioica
(CU AC Cae) eae terete ei rt ee eee DLS es coher iret acy rene ee
GotrLies, LESLIE D., Stephanomeria malheurensis (Compositae), a new species
PO MIMRC) ECS OM geese mony. emer aer eee ae 7ecet hue. Aeon Ree dad oc nae tee ese

Gray, JoHn T., The vegetation of two California mountain slopes ....................-...
GRIFFIN, JAMES R., Maritime chaparral and endemic shrubs of the Monterey
HS yy ae CCUG Iie alt inl ay tne ee tse tie Sdn ds ea al eee eR Ae
HeEnpErson, Douctass M., Notes on the flora of east-central Idaho ............2........
HERMANN, F. A Validation of the name Juncus bufonius var. occidentalis ..........
Howarp, ALICE Q., Endangered species in California: Federal procedures and
RS rcANLEUT Ss mtg JO) Malena eee sae sy eS ats csvset et bee cease etees ceswctee besetcet <2 sa ASsugy saz ceugeecsant ethers Sccees
OWARD, WwEIce:@. (see Powell, W. RODCEL) oie. occssssncucc.sScscecsctelecees escncsaccbes sscle ue estes
Jounson, ANN F., A new subspecies of Abronia maritima from Baja California,
UN Fess] (6 cere rene ae cory 2 SD ge Me A, 222s 2 Se a Reh ore? creck eee, SR ee
Jounson, DALE E. and Rosert Ornpourr, Lasihenia californica (Compositae) ,
another name for a common goldfteld .c....cccc coc ccc ccs ca ocecenccedncceecceeee
KEELEY, Jon E., STERLING C. KEELEY, and JANET LEE, Knobcone pine south-
wang: rance extensions imat ne: olerra.Nevaday <0... ce eee tacos ee
KEELEY, Jon E., A correction on the indigenous distribution of knobcone pine
KEELEY, STERLING C. (see Keeley, Jon E.)
ROONINEDY, ELEN (See. Ganders, Ered R. )ie2cicccccz. eh: ceceeseczide ccs dectev ten esssdivecduasacesgstcenccee
Kurjt, Jos, Germination of Comandra (Santalaceae)
i remANED (see Keeley, Jom: El.) ssc. sees odes ctecsentaceassuns ocleee ceccaecasececs hecavlucese-eescacece cad
Lewis, HARLAN and Jerzy RzEDOWSKI, The genus J7vichostema in Mexico ..........
Love, Ruopa and Marc Freican, Interspecific hybridization between native and
naturalized Crataegus (Rosaceae) in western Oregon ...........00.....222222000eeceeeeeeeee
MacrartanE, Rocer M., On the taxonomic status of Fritillaria phaeanthera
1S USUI po 4G Be TE Wk 6) ree em Oe nee ag eee
Macneitr, C. Don, WittiAm BropHy, and ALAN R. Smit, Ferns of the New
York Mountains, California, with biogeographic comments __...
Major, JAcK, Manual of the vascular plants of Wyoming (review) _................-.
Mixter, JoHN M., Diploid Claytonia perfoliata from southern Mexico

ill

81
110

59
175
234

30
132

101
105

228
185
211

139

234

236

NetsEess, Kurt R., Noteworthy collection of Agrostis humilis Vasey (Poaceae)
ORNDUEF, ROBERT (see Bohm, BrucexAc) oc: eke eae ae ee
OrnDUEF®, ROBERT (See Jonnson, Dale it:) 0 see ee
PALKovic, LAWRENCE A., Reduction of Gunnera kilipiana to synonymy with G.
WVENTCOMO 2a osccs is Sac sa cteg tesa a neo ep se acter
Pitz, GreorcE E., Systematics of Mirabilis subgenus Quamoclidion (Nyctagi-
WAACEAC) pecsscet ee eet A sa ck es ee ee Sete Se te ee ly ce ee ee ee
Porter, Duncan M., Nomenclature changes in Spilanthes, Lycopersicon, and
Opuntia tor the Galapagos slands: 222 ee oe oe eee
PowELL, A. MicHAEL and SHIRLEY A. POWELL, Chromosome numbers in As-
LCLACEAC.” x fa ceceeesteesc ee cette aaa dete he tute ee econ eee ee eee, ne
POWELL, SHIRLEY A, (See Powell, cA, Michael) ee es
PowELL, W. RoBert and ALIcE Q. Howarp, Rare taxa in the literature ................
QUIBELL, CHARLES F. and JoHNn L. StroTHER, Noteworthy collection of Tees-
dalia coronopifolia (Bergeret) Thellung (Cruciferae) ......2.....02..24..-2..4...-
RAE, STEPHEN P., Cordylanthus mollis ssp. mollis (Scrophulariaceae), rediscov-
ery of extinct record within Napa County, California —....000200 0.00002...
REDDER CHARLOT@EE Gr. (see Reeder. }Om init ey sa... ee cere se
REEDER, JOHN R. and CHARLOTTE G. REEDER, Tragus racemosus in Arizona ......
REVEAL, JAMES L., Combinations in the genus Eriogonum (Polygonaceae) not
properly proposed by Munz in “A Flora of Southern California” _...020.2000......
IRZEDOWSKI, JERZY ‘(see [bewis, lari), sore ee
SCHLESSMAN, Mark A., Systematics and reproductive biology of Lomatium fari-
nosum (Geyer ex Hooker). Coulter'& Rose (Umbelliferac) 22
SHAVER, Gaius R., Leaf angle and light absorptance of Arctostaphylos species
(Ericacéae) along environmental pradients, 222s.
SHowers, Davip W., Noteworthy collection of Asplenium septentrionale (L.)
lofi ov GASplemaceae)) cece et oe see cert 5 eee ee epee ee
SMITH, ALAN IR. (see MacNeil Ce Don) ets eae tae et eee
SPELLENBERG, RicHARD, New plant distribution records from the southwestern
United! States:and northern: Miexico iz 2 cee ee
STEBBINS, G. LEpvaRp: (see Taylor, Dean Wm.) 2.....cnne ee eee
STEKEL, PETER and Epwarp A. Cope, Terrestrial vegetation of California (re
VLQW.)). = sce e hon sd seeee Sel sba Ses See eagle oe cs tate ence et 2 pee ea nS Oe ence ee ere
STROTHER, JOHN J. (see Quibell » Charles. I6:) 22 c.cs22 gicestt Soe ee ee ee es
Tanowirz, Barry D., Hemizonia conjugens (Compositae): distribution, chro-
mosome number; and: relationsaips 223.2 2 ee
TAYLor, DEAN Wo., A survival handbook to Sierra flora (review) .................-----
TayLor, DEAN Wo., Vascular plants of the Nevada Test Site and central-south-
ern Nevada: ecological and geographical distributions (review) ....................
TayLor, DEAN Wm. and G. LEepyArD STeBBins, A new species of Eupatorium
(Asteraceae): irom ‘Calitornia. (2.2.20 sc eae ee eee
Topsen, THomas K., Scrophularia laevis (Scrophulariaceae), a legitimate spe-

CHES) ecacce sc agstee ees cee sean 2S a SO neta Rec tee 7 SRD i ae Be ee ee
Turner, B. L., Taxonomy of Axiniphyllum (Asteraceae — Heliantheae) ............
TurRNER, B. L., A new species of Vizguzera (Asteraceae — Heliantheae) from

Nayarit <-IMiexICO 2 s1s22ect.o sete ccdeasenadee sae ac ace Seek ee appre ee ee
VASEK, FRANK C., Jeffrey pine and vegetation of the southern Modoc National

POPGSt.: ccc ceastecetcosceiisecentee saeco nceep ac aGhne seca sa pees Sad see Be et ie eee en

Watson, THOMAS J., JR., Chromosome numbers in Xylorhiza Nuttall (Aster-
aceae —. ASterede) cs.c 6 ies. ctei stesso eee eas ee ee ee
WEesBsTER, GRADY L., The genus Epilobium (Onagraceae) in Australasia: a sys-
tematic.and evonitionary study (review yess ee ee ee
Wesstrer, Gravy L., Daleae Imagines, a illustrated revision of Errazurizia Phillip-
pi, Psorothamnus Rydberg, Marina Liebmann, and Dalea Lucanus emend.
Barneby, including all species of Leguminosae tribe Amorpheae ever re-
ferred. to Daléa (review) <c..0c2s 2S ce ee ee
Werester, Gravy L.: (Gee: Dehgan, Bijan) iid... e2. ea ee ee

iv

Zo

109

106

22

ADRONO

VOLUME 25, NUMBER 1 JANUARY 1978
Contents
SYSTEMATICS AND REPRODUCTIVE BioLocy oF Lomatium farinosum

(GEYER EX HOOKER) COULTER & ROSE (UMBELLIFERAE),

Mark A. Schlessman 1
JEFFREY PINE AND VEGETATION OF THE SOUTHERN

Mopoc National Forest, Frank C. Vasek 9
THREE NEW SPECIES OF JATROPHA (EUPHORBIACEAE)

FROM WESTERN MExico, Bijan Dehgan and Grady L. Webster 30
CHEMOTAXONOMIC STUDIES IN THE SAXIFRAGACEAE S.L, 9.

FLAVENOIDS IN JEPSONIA, Bruce A. Bohm and Robert Ornduff 39
STEPHANOMERIA MALHEURENSIS (COMPOSITAE),

A NEw SPECIES FROM OREGON, L. D. Gottlieb 44
TAXONOMY OF AXINIPHYLLUM (ASTERACEAE—HELIANTHEAE),

B.L, Turner 46

NOTES AND NEWS

REDUCTION OF GUNNERA KILIPIANA TO SYNONYMY

WITH G. MEXICANA, Lawrence A. Palkovic 53
FERNS OF THE NEw YorK MOovuNnNTAINS, CALIFORNIA,

WITH BIOGEOGRAPHIC COMMENTS, C. Don MacNeill,

William Brophy, and Alan R. Smith 54
DIPLomw CLAYTONIA PERFOLIATA FROM SOUTHERN MEXxIco,

John M. Miller oy
NOMENCLATURAL CHANGES IN SPILANTHES, LYCOPERSICON, AND

OPUNTIA FOR THE GALAPAGos IsLanps, Duncan M. Porter 58
RarE TAXA IN THE LITERATURE,

W. Robert Powell and Alice Q. Howard 59
POLLEN SHED AS TETRADS BY PLANTS OF ESCHSCHOLZIA

CALIFORNICA (PAPAVERACEAE), Curtis Clark 59

COMBINATIONS IN THE GENUS ERIOGONUM (POLYGONACEAE)
NOT PROPERLY PROPOSED BY MUNZ IN

A WEST AMERICAN JOURNAL OF BOTANY

“A FLORA OF SOUTHERN CALIFORNIA”, James L. Reveal 60

REVIEW
PETER H. RAVEN AND TAMRA ENGLEHORN RAVEN

The Genus Epilobium (Onagraceac) in Australia:

A Systematic and Evolutionary Study Gravy L. WEBSTER 61
BOOKS RECEIVED 63
REVIEWERS OF MANUSCRIPTS 63
SPECIAL ANNOUNCEMENTS 64

IJBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY

MaproNo is published quarterly by the California Botanical Society, Inc., and is
issued from the office of the Society, Herbarium, Life Sciences Building, University
of California, Berkeley. Established 1916. Second-class postage paid at Berkeley.
Return requested.

Editor — BarBARA D. WEBSTER
Department of Agronomy and Range Science
University of California, Davis 95616

Associate Editor — Grapy L. WEBSTER
Department of Botany, University of California, Davis 95616

Board of Editors
Class of:

1977—Wiu1AM Louis CuLBErsoN, Duke University, Durham, North Carolina

Date M. SmitH, University of California, Santa Barbara
1978—SHERWIN CarLQUIST, Claremont Graduate School

Lest D. GottLis, University of California, Davis

Dennis R. PARNELL, California State University, Hayward
1979—Puire W. RunNDEL, University of California, Irvine

ISABELLE TAVARES, University of California, Berkeley
1980—JaMEs R. GriFFIn, University of California, Hastings Reservation

FRANK A. Lane, Southern Oregon College, Ashland
1981—DaniEL J. CRAWFoRD, Ohio State University, Columbus

James Henrickson, California State University, Los Angeles
1982—Dean W. Taytor, University of California, Davis

RicHArRD VoGcL, California State University, Los Angeles

CALIFORNIA BOTANICAL SOCIETY, INC.

OFFICERS FOR 1977

President: Dr. WinsLow R. Briccs, Carnegie Institution of Washington,
Stanford, California 94305

First Vice President: Dr. Atva Day, Department of Botany, California Academy of
Sciences, Golden Gate Park, San Francisco, California 94118

Second Vice President: Dr. JoB Ku1jT, Department of Biological Sciences,
University of Lethbridge, Lethbridge, Alberta, Canada

Recording Secretary: Dr. CHARLES F. QUIBELL, Department of Biological Sciences,
Sonoma State College, Rohnert Park, California 94928

Corresponding Secretary: Dr. RupoLF Scomm, Department of Botany,
University of California, Berkeley, California 94720

Treasurer: Dr. G. Douctas Barse, California Department of Food and Agriculture,
1220 N Street, Sacramento, California 95814

The Council of the California Botanical Society consists of the officers listed above
plus the immediate Past-President, DENNIs R. PARNELL, Department of Biological
Sciences, California State University, Hayward, CA 94542; the Editors of Madrono;
and three elected Council members: L. R. HEcKARD, Jepson Herbarium, Department
of Botany, University of California, Berkeley, CA 94720 (1975-1977); James R.
GrirFin, Hastings Reservation, Star Route 80, Carmel Valley, CA 93924 (1976-
1978) ; Harry D. Turers, Department of Ecology and Systematic Biology, Califor-

nia State University, San Francisco, CA 94132 (1977-1979).

SYSTEMATICS AND REPRODUCTIVE BIOLOGY OF
LOMATIUM FARINOSUM (GEYER EX HOOKER)
COULTER & ROSE (UMBELLIFERAE)

Mark A. SCHLESSMAN
Department of Botany, University of Washington, Seattle 98195

The large western North American genus Lomatium is replete with
systematic problems. Because many of the approximately seventy species
currently recognized are poorly represented in herbaria, their taxonomic
limits are vague. Almost nothing is known about reproductive biology in
the genus.

Lomatium farinosum is a member of the diminutive, “tuberous” group
of species designated by Marcus E. Jones (1908) as section Cous,. It is
native to the sagebrush-bunchgrass steppe of the Columbia River Basin
and flowers from late March to early May. The species consists of two
geographically significant varieties which are morphologically distin-
guishable only by differences in floral color (Schlessman, 1976). Variety
farinosum, which occurs throughout most of the range of the species, has
white petals and reddish-purple anthers and stylopodia. Variety ham-
bleniae, found only in the extreme western portion of the range, has
yellow petals, anthers and stylopodia (Fig. 1). Previous investigators
have noted the morphological similarity of the two taxa but have main-
tained them as separate species, presumably on the basis of their appar-
ent geographical isolation (Mathias & Constance, 1942, 1945; Hitch-
cock, Cronquist, Ownbey & Thompson, 1961).

MATERIALS AND METHODS

Field studies in 1976 and 1977 included collection of living and dried
specimens, cytological materials and floral visitors, as well as observa-
tions of floral phenology and behavior of pollinators. Chromosome counts
were obtained using standard acetocarmine squash techniques. Pollen
viability was estimated by determining the stainability of pollen from
dried specimens in a solution of aniline blue and lactophenol. My collec-
tions, which include vouchers for chromosome number and pollen viabil-
ity determinations, are kept at WTU. Specimens from the following
herbaria were also utilized: ID, K, NY, OSC, ORE, Reed College, UC,
Walla Walla College, WS and WTU.

Experimental tests of the breeding system and artificial hybridizations
were carried out in an insect-free cage in the greenhouse. To test for
apomixis, flowers were emasculated in bud and plants left undisturbed
until the development of fruit could be determined. Seed set due to self-
ing in the absence of floral visitors was determined by placing three

Madrofo, Vol. 25, No. 1, pp. 1-64. January 30, 1978.

1

2 MADRONO [Vol. 25

var. farinesum
| Var. farinosum with yellow anthers
e var. hambleniae
»* war. hambleniae with white petals
© sympatric populations of the two varieties

Fic.1. Distribution of Lomatium farinosum.

individuals of var. farinosum and three of var. hambleniae in the cage
while the flowers were in bud and leaving them undisturbed for the
course of the experiment. For hybridization studies, flowers of the pistil-
late parent were emasculated in bud and the remaining flowers removed
from the inflorescence. Pollen was transferred once or twice a day for
two to four consecutive days.

RESULTS

The distributions of the two varieties of Lomatium farinosum overlap
in Douglas, Grant, Kittitas and Yakima counties, Washington. Data
from herbarium labels indicated that sympatric populations of the two
varieties would be found at Grand Coulee, Steamboat Rock, Dry Falls
and Sun Lakes in Grant County, Washington. Searches revealed such
populations only at Steamboat Rock and Sun Lakes (Fig. 1).

Although floral color provides the only clear distinction between the
two varieties (Table 1), variation is found in both. Individuals with
yellow, rather than purple, anthers have been collected in three popula-
tions of var. farinosum; and plants with white, rather than yellow, petals
occur in one population of var. hambleniae. Several intermediate forms |
are present at both sites where populations of the two varieties are
known to be sympatric (Fig. 1). These intermediates may have the
following combinations of characters: yellow petals, purple anthers and ©

1978] SCHLESSMAN: LOMATIUM 3

7? oy 4°° Wr
e °¢ &
a. a 4 .° @ @ & @

D E F

Fic. 2. Camera lucida drawings of the chromosomes of Lomatium farinosum.
n=11. A-D, var. farinosum, Schlessman 54, 56, 58, 60; E-F, var. hambleniae,
Schlessman 73, 94. All figures represent prophase I except A, which is metaphase IT.

purple stylopodia; or white petals, purple anthers and yellow stylopodia;
or white petals, yellow anthers and yellow stylopodia.

Chromosome counts were obtained from four populations of var. favi-
nosum and from two of var. hambleniae. All counts are n = 11 (Fig. 2).
No significant differences among the pollen viabilities of the two varieties
and the intermediates were found (Table 2).

Lomatium farinosum is andromonoecious, each plant bearing both
functionally staminate and hermaphroditic flowers. The staminate flowers
have well-developed: stylopodia but lack styles. They do not produce
functional ovaries or set seed (Hardin, 1929; Schlessman, 1976). The
hermaphroditic flowers are strongly protogynous, the styles withering
before pollen is shed. Each plant produces one to several compound
umbels, which are borne singly on scapose peduncles (see illustration in
vol. 3, p. 553, Hitchcock, Cronquist, Ownbey & Thompson, 1961). The
period of anthesis of each umbel overlaps with that of the preceding one.

The ratio of staminate to hermaphroditic flowers in an umbel varies
according to the sequence in which the umbels are produced. The first
umbel is composed completely or predominantly of staminate flowers.
In each sequential umbel the proportion of staminate flowers decreases
so that the flowers of the last umbels are predominantly hermaphroditic
(Fig. 3). These bisexual flowers tend to be clustered in the outermost
umbellets and are usually the outermost flowers within an umbellet.
Flowering occurs centripetally in each umbel and umbellet. Within um-

4 MADRONO [Vol. 25

TABLE 1. COMPARISON OF THE VARIETIES OF LOMATIUM FARINOSUM AND NATUR-
ALLY OCCURRING INTERMEDIATE Forms. For the first five characters, mean values
are followed by standard deviations in parentheses. Data were obtained from over
120 specimens.

Character var. farinosum var. hambleniae Intermediates
1. Height in cm 17 (6) 20 (4) if 52 1 COES))
2. Ray length
in fruit (cm) 3.5 (2) aa) 4 (1)
3. Pedicel length ;
in fruit (mm) 14 (6) 17 (6) 17 (6)
4. Maximum fruit 7
length (mm) 5.5. 1) 6.4 (1) 5.5 (1)
5. Maximum fruit
~ ~ 4 ~
Sie (ar 2:5 (0:5) 2.5 (0.5) 2.55 (0)
6. Longest leaf
1 cm or more 1 cm or more 1 cm or more
segments
7. Oil tubes on rier Ae 4-7
commisure
8. Fruit surface glabrous glabrous glabrous
9. Habit acaulescent acaulescent acaulescent
10.B ech linear to linear to linear to
pe accu uabG lanceolate lanceolate lanceolate
11. Petal color white yellow white or yellow
12. Anther color purple yellow purple or yellow
Se SUG lets eat purple yellow purple or yellow
color
TABLE 2. POLLEN VIABILITY OF LOMATIUM FARINOSUM.
Pia Number of Standard
determinations Mean (%) Range deviation
var. farinosum 25 92 58-100 10
var. hambleniae Ti 88 59-100 12
intermediates 13 86 51-100 13

bellets the flowers are tightly clustered so that geitonogamous pollina-
tion can occur without transfer of pollen by insects.

Floral visitors to Lomatium farinosum include leaf-cutting bees (Mega-
chilidae) , halictid bees (Halictidae) , cuckoo bees (Anthophoridae) , sphe-
cid wasps (Sphecidae), mason wasps (Vespinae), tachinid flies (Tachini-
dae), bee flies (Bombylidae) and midges (Chironomidae). Typically, an
insect will land on one of the outer umbellets and wander over that

wal

1978] SCHLESSMAN: LOMATIUM
100

% staminate flowers
RO Ol
ro) | (o>)

>

io)

umbel
100
75
S 50 B
= 25
9
1 4
umbel

Fic.3. Percentages of staminate flowers in the umbels of Lomatium farinosum.
Umbels are numbered in the sequence in which anthesis occurs. Lines connect the
mean percentages and vertical bars indicate standard deviations. A, var. farinosum,
data from 30 specimens; B, var. hambleniae, data from 30 specimens.

6 MADRONO [Vol. 25

TABLE 3. RESULTS OF ARTIFICIAL HysripizaTions. Collection numbers are my
own. The letters after collection numbers indicate colors of the petals, anthers and
stylopodia, respectively (e.g., WPY denotes an intermediate form with white petals,
purple anthers and yellow stylopodia).

Flowers
Cross Number of producing

axe Q 3 flowers seed %
var. farinosum x

var. hambleniae SURES aS ie a

var. hambleniae x 474 x 158 e) 2 67

var. farinosum 179 x 174 10 10 100

474x174 10 10 100

480 x 158 15 ES) 100

total 38 total 37 x 96

intermediate x 200 (WYY) x 158 11 3 27

var. farinosum 200 (YPP) x174 10 4 40

200 (WYY) x 174 10 10 100

total 31 total 17 x 55

intermediate x 200 (YPP) x 474 10 0) 0

var. hambleniae 200 (WPY) x 474 8 8 100

total 18 total 8 x 44

umbellet from several seconds to three or four minutes before flying to
another umbellet. If any anthers have dehisced, the entire underside of
the insect may be exposed to pollen. Floral visitors may spread pollen
from flower to flower within inflorescences as well as between inflores-
cences. No differences between var. farinosum and var. hambleniae in the
kinds of floral visitors or pollinator behavior were observed. In popula-
tions consisting of both varieties and intermediate forms, floral visitors
did not exhibit constancy for any floral type.

None of the 30 flowers emasculated to test for apomixis (10 for var.
farinosum, 20 for var. hambleniae) set seed. The six plants used to test
for selfing in the absence of insects produced a total of 261 flowers, only
10 of which (4%) yielded fruit. Seed production from crosses between
the two varieties ranged from 67—100%, and that from crosses between

intermediates and typical forms ranged from 27—100% (Table 3.) All |

flowers which bore fruit produced two morphologically normal seeds.

DISCUSSION

- It is clear from their almost identical morphologies, chromosomal num- |
bers, sympatry and naturally occurring intermediates that the white- _

EO

1978] SCHLESSMAN: LOMATIUM 7

and yellow-petaled plants examined are conspecific. My crossing studies
strongly suggest that the naturally occurring intermediates arose through
hybridization between the two typical forms, and that some gene flow
still exists between them. In my opinion the significant geographical
trend in distribution of the two color forms warrants taxonomic recog-
nition of two varieties. This treatment is consistent with that of other
investigators dealing with infraspecific variation in floral color in Lo-
matium (Mathias & Constance, 1945; Hitchcock, Cronquist, Ownbey &
Thompson, 1961). Although the data are insufficient to establish a pat-
tern, intervarietal crosses appear to be more successful than those be-
tween either variety and the intermediate forms. Cytological abnormali-
ties undetectable by determinations of pollen viability may reduce the
fertility of intermediates. The two crosses between intermediates and
typical forms that resulted in full seed set may have involved progeny
of backcrosses to one of the typical forms.

Dichogamy is a common adaptation for outcrossing in self-compatible
plants (Baker, 1960). Although protogyny apparently prevents autog-
amy in Lomatium farinosum, the compact umbellets and the centripetal
sequence of anthesis increase the likelihood of geitonogamy with or
without the aid of insect vectors. The seed set in my selfing experiments,
which were conducted in the absence of any floral visitors, was probably
due to geitonogamous selfing. Studies to determine the extent to which
insect-mediated geitonogamy may occur in L. farinosum and other “‘tu-
berous” species of Lomatium have been initiated. The position of the
hermaphroditic flowers is also an adaptation for outcrossing in L. farino-
sum, Since hermaphroditic flowers are often the first of an umbellet or
umbel to reach anthesis, pollinations occurring soon after the flowers
open are likely to be xenogamous.

Regular patterns of change in proportions of staminate and hermaph-
roditic flowers have been reported for many andromonoecious Umbelli-
ferae (Miller, 1883; Hardin, 1929; Bell, 1971; Schlessman, 1976).
The widespread occurrence and general constancy of these patterns indi-
cate that they are genetically controlled. Experiments utilizing altered
environmental conditions or growth hormones have produced minor vari-
ations in the ratio of staminate to hermaphroditic flowers, but no change
in overall pattern (Braak & Kho, 1958; Quagliotti, 1967). These patterns
may have evolved under selection brought about by dichogamy (Bell,
1971; Schlessman, 1976). In the case of a protogynous species such as
Lomatium farinosum, little or no pollen would be available to hermaph-
roditic flowers in the first umbels to reach anthesis and few of these
flowers would set seed. Selection for the conservation of reproductive
effort would result in elimination of hermaphroditic flowers from the first
umbels and a preponderance of them in later-flowering umbels. Cruden
(1976) has reported preliminary evidence of ecotypic variation in pro-
portions of staminate and hermaphroditic flowers in Heracleum lanatum

8 MADRONO [Vol. 25

Michx. Since my data represent several populations, such variation may
contribute to the large standard deviations in percentages of staminate
flowers in L. farinosum.

Bell (1971) has proposed that uniformity of floral structure in Um-
belliferae represents an ancient adaptive peak for utilization of unspe-
cialized pollinators. He has suggested that studies of the comparatively
minor changes in breeding systems superposed since this adaptive peak
was reached may have wide applications to evolutionary studies within
the family. Investigations of reproductive biology have clarified the
infraspecific relationships of Lomatium farinosum. Comparative studies
of floral biology and breeding systems should elucidate phylogenetic
relationships among the species of this taxonomically “difficult” genus.

TAXONOMY
Key to the varieties of Lomatium farinosum
Petals white; anthers purple; stylopodia purple . . . var. farinosum
Petals, anthers and stylopodia yellow . . . . . .. var. hambleniae

Recognition of var. hambleniae as an infraspecific taxon requires the
following new combination. The correct citation and synonymy of these
taxa are as follows.

Lomatium farinosum (Geyer ex Hooker) Coulter & Rose var. farinosum,
Contr. U.S. Natl. Herb. 7:210. 1900.

Basionym: Peucedanum farinosum Geyer ex Hooker, Lond. Jour.

Bot. 6:235. 1847. Type: USA, Idaho, “On an isolated rock in the

Coer [sic] d’Alene Mountains, on wet clay, with Sedum steno peta-

lon [sic] and Platyspermum,” April, 1844, Geyer 325 (Holotype:

K!).—Cogswellia farinosa (Geyer ex Hooker) M. E. Jones, Contr.

West. Bot. 12:33. 1908.

Lomatium farinosum (Geyer ex Hooker) Coulter & Rose var. hambleniae
(Mathias & Constance) Schlessman, comb. nov.

Basionum: Lomatium hambleniae Mathias & Constance, Bull. Tor-

rey Bot. Club 69(3):153. 1952. Type: USA, Washington, Grant

County, ‘“‘on a level scabrock bench at Dry Falls, Grand Coulee,”

22 April 1941, Frances G. Hamblen s.n. (Holotype: UC! Isotype:

WS!).

ACKNOWLEDGMENTS
I thank Dr. Melinda F. Denton for advice throughout this study and the prepa-
ration of this paper. This work was supported by a grant-in-aid of research from
the Society of Sigma Xi, a Graduate School Special Fellowship from the University
of Washington, and by the Department of Botany, University of Washington. The
Ecosystems Department of Battelle Pacific Northwest Laboratories provided access
to the Arid Lands Ecology Reserve at Hanford, Washington.

LITERATURE CITED
Baker, H. G. 1960. Reproductive methods as factors in speciation in flowering
plants. Cold Spring Harbor Symp. Quant. Biol. 24:177-191.

1978 ] VASEK: PINUS 9

BELL, C. R. 1971. Breeding systems and floral biology of the Umbelliferae or evi-
dence for specialization in unspecialized flowers. In V. H. Heywood (ed). The
biology and chemistry of the Umbelliferae. Academic Press, New York, 93-107.

Braak, J. P. and Y. O. Kuo. 1958. Some observations on the floral biology of the
carrot (Daucus carota L.). Euphytica 7:131-139.

CruDEN, R. W. 1976. Intraspecific variation in pollen-ovule ratios and nectar secre-
tion—preliminary evidence of ecotypic adaptation. Ann. Missouri Bot. Gard.
63:277-289.

Harpin, E. 1929. The flowering and fruiting habits of Lomatium. Res. Stud. State
Coll. Wash. 1(1) :15-—27.

Hitcucock, C. L., A. Cronquist, M. OwnBeEy and J. W. THompson. 1961. Vascular
plants of the Pacific Northwest, vol. 3. University of Washington Press.

Jones, M. E. 1908. New species and notes. Contr. West. Bot. 12:1-81.

Matuias, M. E. and L. Constance. 1942. New North American Umbelliferae. Bull.
Torrey Bot. Club 68(2) :121-124.

. 1945. Umbelliferae. Jn North American Flora. New York Botanical Gar-
den. 28:161-295.

Mutter, H. 1883. The fertilization of flowers. Macmillan, London.

QuactioTTi, L. 1967. Effects of different temperatures on stalk development, flower-
ing habit, and sex expression in the carrot Daucus carota L. Euphytica 16:83-103.

SCHLESSMAN, M. A. 1976. A systematic study of Lomatium farinosum (Geyer ex
Hooker) Coulter & Rose (Umbelliferae) and its relatives. M.S. Thesis, Univer-
sity of Washington.

JEFFREY PINE AND VEGETATION OF THE
SOUTHERN MODOC NATIONAL FOREST

FRANK C. VASEK
Department of Biology, University of California, Riverside 92521

Jeffrey pine (Pinus jeffreyi Grev. & Balf.) ranges the length of Cali-
fornia in forested areas and extends into southern Oregon and northern
Baja California (Griffin and Critchfield, 1972). It occurs in several forest
and woodland communities but also forms a fairly distinctive Jeffrey pine
forest type east of the Sierra Nevada—Cascade ranges (Society of Ameri-
can Foresters, 1954). Jeffrey pine forests reach widespread development
near the Owens River headwaters and Mono Lake and in eastern Plumas
and Lassen counties. Elsewhere, Jeffrey pine occurs as an element of the
mixed conifer forest (Griffin and Critchfield, 1972), a vegetation type
roughly equivalent to the yellow pine forest of Munz and Keck (1949)
and to several forest types recognized by the Society of American For-
esters (1954). In some of the latter, Jeffrey pine and ponderosa pine
(Pinus ponderosa Laws.) mingle freely. Such forests are classed as pon-
derosa pine forest types without particular analysis, even though Jeffrey
pine sometimes predominates.

10 MADRONO [Vol. 25

As an element of mixed conifer forests, Jeffrey pine reaches its northern
limit in the Siskiyou Mountains of southern Oregon (Whittaker, 1960;
Franklin and Dyrness, 1973). To the south, it has generally been in-
cluded in forest communities dominated by ponderosa pine (Munz and
Keck, 1949; Griffin and Critchfield, 1972; Horton, 1960; Minnich, 1976;
Thorne, 1976). However, Jeffrey pine increases in importance toward
higher elevations and toward the interior (Haller, 1962; Griffin and
Critchfield, 1972), and forested areas dominated by Jeffrey pine to the
virtual exclusion of ponderosa pine have been noted in the Sierra Nevada
of Kern County (Twisselman, 1967) and Fresno County (Klyver, 1931),
on Mt. Pinos (Vogl and Miller, 1968), in the San Gabriel Mountains
(Thorne, pers. comm.), and in the San Bernardino Mountains (Vasek,
1966). Furthermore, P. jeffreyz dominates forests in the peninsular ranges
(Santa Rosa Mountains, Laguna Mountains) of southern California.
Whereas, P. ponderosa reaches a southern limit near Cuyamaca Lake in
San Diego County, California (Haller, 1959), P. jeffreyi continues south
to the Sierra Juarez and the Sierra San Pedro Martir of Baja California
where it is a dominant in forested areas (Haller, pers. comm.).

Several questions arise concerning the community definition, composi-
tion, and ecology and the phytogeographic history of Jeffrey pine forests
in the south relative to those in the north and of such forests relative to
mixed conifer forests. Accordingly, a long-range program of vegetation
sampling was initiated in the northern Jeffrey pine forests to provide a
data base from which north-south comparisons and interpretations might
be made. That project is a long way from fruition. However, shortly after
sampling several plots in the Modoc National Forest, several vegetation—
soil maps (U.S. Forest Service, 1953a) for the sampled area came to my
attention. These maps are part of a series prepared by the State Coopera-
tive Soil-vegetation Survey (U.S.D.A. Forest Service, 1958). Most of the
available vegetation—soil map coverage applies to the coast range forests |
and coverage is sparse for the region east of the Sierra Nevada—Cascade |
ranges. Consequently, the coincidence in sampling areas is fortuitous.

My few ground samples could be correlated with broad scale coverage
of more than 200 sections (520 km?) and an opportunity was presented |
to address several general questions regarding species composition and |
species associations, and the environmental and distributional relations of |
northern Jeffrey pine forests. Consideration of these parameters may lead
to a better understanding of the northern Jeffrey pine forests and provide
a focal reference for investigation of those to the south. |

This paper characterizes a Jeffrey pine forest near the northern limit of |
that forest type and draws relationships to comparable forests farther
south.

1978] VASEK: PINUS 11

Fic. 1. Map showing the distribution of Pinus jeffreyz and the locations of
sampling areas in California: A, southern Modoc National Forest; B, Plumas N.F.;
C, Tahoe N.F.; D, Toiyabe N.F. (Nevada); E, Devils Gate Pass; F, Deadman
Summit. Adapted from Griffin and Critchfield, 1972.

METHOpsS

This study employed two approaches: a simulated aerial point sam-
pling based on vegetation—soil maps and direct ground sampling.

Vegetation—soil maps were compiled by the U.S. Forest Service (1953a)
from aerial photographs plus direct ground observations and supple-
mental soil samples. Vegetation—soil maps resemble a jig—saw puzzle in
which units of irregular shape and variable size are each coded for char-
acteristics of the constituent vegetation and soil types (Colwell, 1974).
Each cover element comprising more than 20% of the total cover, and

12 MADRONO [ Vol. 25

the included species, each comprising at least 20% of the cover in that
element, is listed in order of decreasing cover value (U.S. Forest Service,
1953b). Soil series, depth, texture and exposure are coded for each unit.
Thus a large body of semi—quantitative descriptive information is avail—
able for a significant segment of forest.

Vegetation—soil map coverage for the Modoc National Forest and for
all northeast California includes only that portion of the Big Valley Dis-
trict, Modoc National Forest (Fig. 1), located in Lassen County, south
of the Modoc County town of Adin. The National Forest Boundary is
irregular and encloses a mapped area of 208 full sections and parts of 27
other sections. Vegetation of this forest segment was sampled as follows:

1) A series of 10 transects was made from west to east across alternate
tiers of sections. The first transect started southeast of Adin, and south
of the county line, with section 36, ROE T39N, and included 10 consecu-
tive sections. The last transect, 29 km to the south, started with sections
31 to 34, R8E T36N, skipped 11 private, unmapped sections, and ended
at sections 34 and 35, RIOE T36N, for an interrupted transect of 6
sections.

2) Each transect consisted of 2 tracks parallel to the north section line
and located respectively '4 and 34 of the way to the south section line.

3) Five sample points were equally spaced along each track, starting at
the west edge of each section, providing 10 sample points per section.

4) The vegetation—soil unit at each sample point was recorded.

5) The accumulated data for 105 full sections (1050 sample points)
were tabulated to yield frequency data for each major species and several
non-specific cover elements (e.g. bare ground).

Direct ground samples were taken from the Modoc, Plumas, Tahoe
and Toiyabe National Forests (Fig. 1) during Sept. and Oct. 1974. Sam-
pling areas potentially dominated by Jeffrey pine were selected using a
published range map (Griffin and Critchfield, 1972) before going into the
field. In each area sample points were located at 100—pace intervals
(about 160 m) along a line-of-sight transect established on a diagonal
to the general slope. Sample points were used to center temporary circu-
lar sample plots of 100 m*. In each plot dbh was recorded by species for
each tree taller than 2 m; height was recorded by species for each tree
shorter than 2 m; and 2 diameters were recorded by species for each
shrub. Basal area for trees and crown cover for shrubs were calculated
from half—diameters by assuming circular shapes. For non-circular shrubs,
the longest diameter (a) and a perpendicular short diameter (b) were
measured and the crown cover area calculated from A = 7 (a/2*b/2).
For prostrate or sprawling shrubs such as Ceanothus prostratus and Sym-
phoricar pos acutus, density was arbitrarily assumed to be one plant per |
plot and ground cover was estimated visually for each plot.

Crown cover for Jeffrey pine was estimated from basal area. The crown
cover/basal area for 10 trees on which both parameters were measured |

1978] VASEK: PINUS 13

ranged from 30 to 165; 100 was selected as representative. The same con-
version factor was applied to other tree species but possible error was not
estimated.

Five to 8 plots were scored in each sampling area. Data include density,
cover, and frequency for each species. Relative density, relative cover,
and relative frequency for each species were calculated and summed to
obtain an Importance Value (Mueller-Dombois and Ellenberg 1974), a
convenient statistic for comparing forest stands in different areas. In
addition, exploratory data, gathered during preliminary method evalua-
tions in 1972, were included to characterize forests in Mono County
(Fig. 1).

Individual pine trees sometimes posed problems in identification by
manifesting several traits from two species. These were listed as hybrid
derivatives under the species most closely resembled.

In the Modoc Forest, P. ponderosa includes 7 putative hybrid deriva-
tives, and P. jeffrey includes 6 putative hybrid derivatives. In the Plumas
Forest, 6 hybrid derivatives are included in P. ponderosa and 2 in P.
jeffreyi. In the Tahoe region, all trees observed were typical Jeffrey pine.
However, Pinus washoensis occurs nearby (Haller, 1961), and it is pos-
sible that some Washoe pine data is included under Jeffrey pine in that
area.

Identity of the juniper subspecies posed another problem. The south-
ern Modoc Forest is at the transition between the usually dioecious,
forest—dwelling Juniperus occidentalis ssp. australis and the slightly
smaller, commonly monecious, woodland forming J. occidentalis ssp.
occidentalis (Vasek 1966, Vasek and Thorne 1977). This paper deals
primarily with J. 0. australis but some influence of J. 0. occidentalis in
the open woodland areas cannot be ruled out.

RESULTS AND DISCUSSION

Vegetation structure based on vegetation map data. Vegetation of the
southern Modoc National Forest consists of a mixture of tree, shrub, and
herbaceous species arranged on and around a series of small volcanic
mountains. Topographic relief of the region ranges from elevations of
about 1280 m at Adin to about 1950 m at the top of Hayden Hill. The 3
vegetational phases (forest, woodland and scrub) are identified on the
basis of whether trees or shrubby elements dominate, or whether only
shrubby species occur. The 3 phases are arbitrarily defined for descrip-
tive convenience since forests grade into woodlands and woodlands into
scrub.

Forests occupy about 39% of the area (Table 1) based on frequency
of sample points in forest areas. Woodland occupies about 43%, and
scrub or brush vegetation about 17%. In general, then, the mapped vege-
tation within the National Forest boundary is relatively open, more than
60% having no or rather sparse tree cover.

14 MADRONO [Vol. 25

TABLE 1. FREQUENCY OF SPECIES AND COVER ELEMENTS IN THREE SUB-GROUPS
BASED ON VEGETATION StRuCcTURE. I, shrubs only; IJ, shrubs dominant; III, trees
dominant. % = percent occurrence in total sample; N= number of sample points.

A
For Abies concolor, the symbol W is used for convenience, rather than W.

Species or element Symbol Observed Frequency
I II III Total %
Trees
Abies concolor W -—- 8 82 90 8.6
Calocedrus decurrens I — 11 78 89 8.5
Pinus ponderosa Y —= 140 625 465 44.3
Pinus jeffreyi J pa 269 342 611 58.2
Quercus kelloggii B -—— 17 10 27 2.6
Juniperus occidentalis Jo a 292 71 363 34.6
Woody shrubs
Ceanothus velutinus Cv 5 3 5 13 122
Ceanothus prostratus Cpo 1 43 () 116 1.0
Arctostaphylos patula Ap 50 105 43 198 18.9
Cercocarpus ledifolius Cl 62 215 115 392 37.3
Amelanchier pallida Aa 26 9 1 36 3.4
Prunus emarginata,
P. demissa Pe,Pd 3) 1 a 4 0.4
Sym phoricar pos
vaccinoides Sal: —- 3 2 5 0.5
Brush
Artemisia tridentata Atr 82 254 61 397 3S
Purshia tridentata Pt 59 183 oy! 293 27.9
Chrysothamnus
NAUSEOSUS Chn 1S 57 13 85 8.1
Chrysothamnus
viscidiflorus Chv 8 12 — 20 1.9
Artemisia arbuscula Aar 49 82 11 142 13:5
Artemisia cana Arc 3 — — 3 0.3
Herbs and other
W yvethia mollis Wm 4 32 ons 87 8.3
Grasses and herbs Gr 130 303 164 597 56.9
Bare ground Ba 115 222 232 569 54.2
N 183 454 413 1050/4602
% 17.4 43.2 39.3

Areas of dense forest grade into less dense forest patches, some with a
shrubby understory or intermittent open or shrubby areas. Open forests
grade into woodlands. Vegetation structure appears to be influenced by,
or correlated with, soil characteristics. Forests preferentially occur on
north slopes with deep soil and low surface rock cover (Table 2). Wood-
land tends to occur on south slopes with greater surface rock cover and
soil of intermediate depth. Scrub vegetation occupies shallow soils (‘Table
Dre

1978] VASEK: PINUS 15

TABLE 2. FREQUENCY OF SEVERAL SOIL CHARACTERISTICS IN THREE VEGETATION
STRUCTURAL SUBGROUPS (as in Table 1). N= number of sample points; NC =
number of sample points with unclassified soil. Surface rock categories RO, R4, R3,
R2 and R1 respectively indicate 0-5, 5-20, 20-50, 50-80 and 80-100% imbedded
large rocks or bed rock; average surface rock cover is calculated from class mid-
point values. Soil depth is indicated in feet, the letter s indicating stones equal to
about 1 ft of soil; soil series with arbitrary numbers.

I II Ill Total %
N 183 454 413 1050
NC 18 4 1 29
Aspect
North 67 130 256 453 44.1
South 98 320 156 574 55.9
Surface rock (Ave. = 14.9% surface rock cover)
RO (0-5%) 64 129 249 442 43.0
R4 = (5-20%) 48 176 109 335 32.4
R3 (20-50%) 45 114 42 201 19.6
R2 (50-80%) 8 oy 10 49 4.8
Ri (80-100% ) — — 2 2 0.2
Soil depth (average = 2.43)
5 ~~ -—— 17 17 7
5s — — St ot 3.0
4 | 22 26 55 5.4
4s 10 25 188 Z28 Dies
3 8 20 10 38 S34
35 7 148 93 248 24.1
2 8! 19 1 Si 320
2s 86 154 35 275 26.8
1 2 4 -— 6 0.6
1s 34 58 11 103 10.0
Soil series
714 28 129 107 264 25.7
720 45 147 197 389 37.9
d22 11 54 80 145 14.1
761 3 10 — 13 1.3
763 6 5 -— 11 ila
765 28 44 9 81 7.9
776 39 38 10 87 se)
het| 1 16 1 18 1.8
Misc. 4 fi 8 19 1.9

The several soil series are arbitrarily numbered (Colwell, 1974) and
several of the major soils and their frequency distribution are listed in
Table 2. Soil 720, a loam with good permeability, good drainage and
slight acidity is most frequent. Other soils differ as follows: loam 722 is
neutral with excessive drainage; gravelly-sandy loam 714 drains exces-
sively as does loam 765; stony loam 766 and stony—sandy loam 777 drain
poorly and have slow permeability; sandy loam 763 has imperfect drain-
age; and clay loam 761 is alkaline with slow permeability.

16 MADRONO [Vol. 25

TABLE 3. AVERAGE FREQUENCY OF SOUTH-FACING ASPECT, SURFACE Rock Cover,
AND DEPTH OF SOILS ON WHICH THE INDICATED SPECIES OR COVER ELEMENTS OCCUR.
Notations as in Tables 1 and 2.

South Surface Soil
aspect rock depth
Av. % Av. % Av. ft.
Trees
W 222 4.7 3.6
I 24.7 8.9 Bee
Y 38.4 7.6 2.9
i] 47.7 9.7 2.9
B 40.7 14.1 hea |
Jo Vid 25.9 17
Woody shrubs
Cv 7.7 (es 3.5
Cpo 47.8 10.5 3.0
Ap 55.3 11.8 2.4
Cc} 52.5 13.5 2.4
Aa 44.4 10.8 17
Brush
Atr 51.0 127 22
Pt 51:5 14.1 ou
Chn 76.2 15.7 22
Chv 62.1 9.3 1.8
Aar 83.2 41.7 1.0
Herbs and other
Wm 57.9 9.7 2.6
Gr 62.5 17.3 Z-2
Ba 59.7 20.8 2.2

Vegetational composition based on vegetation maps. The 21 species
observed, plus grasses (including associated herbs and bare ground), are
grouped according to growth form in Table 1.

Pinus jeffreyvi has the highest frequency followed by P. ponderosa,
Artemisia tridentata, Cercocarpus ledifolius, Juniperus occidentalis, and
Purshia tridentata. The non-specific cover elements, grasses and bare
ground, also have high frequencies which total about 25% of all the fre-
quency occurrences (= relative frequency), and emphasize the open
nature of the vegetation. The high frequency of bare ground in forested
areas doubtless reflects the nature of the geology with its exposed volcanic
ridges, peaks, and abrupt slopes.

Tree species array along a gradient initially identified on the basis of
soil depth (Table 3). Abies concolor occurs on soils averaging 110 cm in
depth whereas Juniperus occidentalis occurs on soils averaging 52 cm.
Other tree species fill in the gradient in the sequence listed in Table 3.

The same species follow a similar gradient in surface rock cover from
about 5% average surface rock for soils of A. concolor to 26% for J.
occidentalis (Table 3). However, the internal gradient sequence differs

|
}

1978] VASEK: PINUS L]

TaBLE 4. RELATIVE FREQUENCY OF PLANT SPECIES AND COVER ELEMENTS ON
SEVERAL COMMON SOILS.

Soil series

720 722 714 765 776 Ci 763 761 Other

Trees
W 67 30 1 2
I 85 8 3 1 2
Y 47 20 30 1 2
J 46 20 31 1 1 1
B 63 15 4 11 7
Jo 28 11 21 15 13 5 4 3
Woody shrubs
Cv 69 8 Z3
Cpo 90 1 8 1
Ap 79 14 6 1
Cl 47 20 27 4 2
Aa 67 8 8 6 11
Brush
Atr 38 13 29 18 1 1
Pt 43 9 23 20 3 2
Chn ale 28 26 27 1 ii
Chv 50 3 43 3
Aar 3 2 3 59 13 8 | 5
Herbs and others
Wm 86 6) 1 9
Gr 19 13 35 13 13 3 Z 2 1

Ba 42 10 16 8 15 3 Z Z 1

slightly from the soil depth sequence in that soils of Calocedrus have a
little more surface rock, on average, than do soils on which P. ponderosa
occurs.

A pattern of nearly parallel soil gradients is extended to the parameter
of slope aspect. This gradient ranges from a 22% ocurrence of A. con-
color on south facing soil formations ( = 78% preference for north
aspects) to 78% preference for south aspects for J. occidentalis. On slope
aspect, however, Quercus kelloggii assumes a gradient sequence position
close to P. ponderosa rather than between P. jeffreyi and J. occidentalis
as observed in the 2 other gradients.

All tree species grow on soil 720 and usually one or more other soils.
Calocedrus is most frequent on soil 720, followed by Abies and then Q.
kelloggii. Most notably, J. occidentalis occurs on a wide range of soil
types, with significant occurrence on heavy-—textured soils such as 776
and 777 (Table 4).

Woody shrub species assort along a soil depth gradient (Table 3)
Closely parallel to that for tree species. Along a gradient from deep to
shallow soils, Ceanothus velutinus is followed by C. prostratus, Arcto-
staphylos patula, Cercocarpus ledifolius, and Amelanchier pallida.

18 MADRONO [Vol. 25

TABLE 5. Sor CHARACTERISTICS ALONG AN ASSOCIATION GRADIENT IDENTIFIED ON
THE BAsIs OF RELATIVE DOMINANCE OF JEFFREY PINE VS. PONDEROSA PINE AT ONE
EXTREME AND MOUNTAIN JUNIPER AT THE OTHER. Relative dominance is indicated
by the sequence of species symbols.

South Surface Soil Soiliceries

Associa- Rel. aspect rock depth

tion N Freq. % Av.% Av. fit. 120" 414.9722" 165°" 746.” Gea
ny. 39 AD 3533 6.7 32 92 5) oe MS eS
YJ 308 33:3 33.6 5.6 3.4 41 33 24 —- —- —
JY 118 12.3 io228 10.7 Pad | 50 28 17 — —~ —
J 97 10.5 52.6 8.1 2.6 47 38 14 — —~— —
Joly 59 64 483 126 2.4 43 «33: 92° (= ee
JoJ 88 95 85.2 21.7 19 S728 SSeS
Jo 216 23:3 82.6 31.5 1.4 H2 19 — 23 Van 8

The surface rock cover gradient for woody shrubs is less regular and
less extensive than for trees. Ceanothus velutinus grows on soils with
about 7% surface rock. Other shrub species form a group with soil sur-
face rock cover averages ranging from 10.5—13.5%

A stepped gradient in slope aspect is apparent for soils on which
shrubs occur (Table 3). A strong preference (92%) for north aspects
places C. velutinus in a class by itself. The other shrub species array near
neutral values (i.e. 50%) with A. pallida and C. prostratus showing
slight preference for north aspects and Arctostaphylos and Cercocar pus
showing slight preference for south.

Shrubs closely parallel the pattern described for trees with regard to
soil type. Generally a decreasing occurrence on soil 720 follows a slightly
different species sequence than was observed in the soil depth gradient
(Table 4). Ceanothus prostratus is most frequent on soil 720 whereas
Cercocar pus ledifolius is least frequent.

Soft shrub or “brush” species occupy similar gradients, but in contrast
to trees and woody shrubs, the soil depth gradient range is lower, the
range of surface rock cover higher, and the range of preference for soils
with south aspect greater (Table 3). Generally, Artemisia tridentata and
Purshia tridentata occur on the most favorable of the ‘“‘brush soils”. Arte-
misia arbuscula characteristically occupies heavy soils and is essentially
excluded from soil 720 (Table 4).

The cover elements conferring an open character to the vegetation, i.e.,
Wvyethia mollis, grasses and bare ground, usually occupy soils that gen-
erally face south (Table 3). Those are somewhat shallow but deeper than
most of the “brush” soils. Surface rock cover is moderate for soils of
W. mollis and rather extensive for grasses and bare ground.

19

: PINUS

VASEK

1978]

OT 6t 167
80°19 8'T9
ee )
£°00¢ T9°STOZ col 99 OSOLE
Ve £70 T ST
ce 188 I el
is O8"2 s 8
6:92 COT? el LOT cane
OT £00 i T
aus v8'0 c z
OT cO'0 I T
VOT OS'29 8 8
9°9P 99°892 v2 OOT OL el
6°62 08°92 L ssl
[Oa eb 0 I C
68 09 T 6 eT
L'S¢ eO:L Ce LI vy vS‘9LT
Cc 6L 9¢°698 IG vit O7’ SIT
c OV oy COL st c9
07 CC br I ¢
Le ST él it eT
“AT @) A d A

£ 29 VLC
Svs S8e
L 8
CC LCs SOOTE
L£c0
T8°8
T6'T 68°0
19°0 Chsy
£00
80
0720
os'0 OO'rT
6718 LS VE
cost
€T0
O¢'0 c9'0
Ov'cOl ch Ce
VE S8Z OS’ SsT
C9 8Y v9'0
AI 1H

eae afdures

6'S8 C972
Tr6 OTP
9 9
979s CL OPC
LEC
OO' rr 00'V
ene v8'9OL
L8 Ol
89°0
oc St
vl 702 SOLE
Sr cle 6S'0¢
canal
ST 6l
TT I

IIAOD 991} BBRIIAY
I9A0) dSBIIAY
sjo[d Jo Jaquinny

[210],

Saplour99va SOgADI1AOYGuks
DILUAO {IDI DSOY

x UNUIINIAL SaQuyy
DIDJUIP1A] DIYSANG
SnIsnap UoOWMasuad
SNSOASNDU SNUMYIOS KAY D
SNU1INIaa SNYjoUDIaD
SNIDAISOAG SNYJOUDID
sniofipa] Sn¢s4pI0949)
DIDIJUIPIA] DISIMAIAP
pinqog SsojkydvjsojIsp
Dpiyjog AalyIUDaM PY
Syp]Uap1gIIO SNAIGIUN
tKasffal snutg

DsSoAapuog Snutg
SUadANIAP SNAPIIOID
40109U09 Sa1qy

sa19adS

TI] bore ye wnmissisoIsIA Saq~y = xx ‘(APaed 10) pyeujsa 19A09 = y {9g pure y ‘CG 2a27DJa4 Jo UOTYBUIUINS UWOI} PaAlap an[ea
oue}yiodwu = ‘A’] +.W Ul JaAOD UMOID = Q ‘ADUONbaIy = YF ‘AysuUap = q ‘8 BGR], Ul pd}sl] SUOI}LIOT ‘LNAWATY INVNIWOQ V SV ANIg AaaIaal
ONIAVAH] LSAYOT IVNOILVN OOGOJT NI SVANY AIAWVS ¢$ NI SAIOadS IVINNAYAG AO SOLLSIMALOVYVHD NOILVIAOTIA GNV YTAOD NMOAD ‘9 ATAVGE

20 MADRONO [Vol. 25

Jeffrey pine gradient segment. Since Jeffrey pine comprises a major
focus in this study, its position in a gradient relative to other species was
examined in detail. Seven association categories were identified (Table
5) based on relative dominance of P. jeffreyvt. Pinus ponderosa and Juni-
perus occidentalis occur at either end of this gradient segment. Pinus
jeffreyz occurs in the 5 intervening categories according to its increasing
dominance relative to P. ponderosa (steps 1-4) and its decreasing domi-
nance relative to J. occidentalis (steps 4-7).

In this analysis, mixed forests dominated by P. ponderosa are most
frequent (Table 5), followed by Juniperus woodlands. Unmixed stands
of Jeffrey and ponderosa pines are much less frequent than those of
western juniper, reinforcing other indications concerning the general dry,
open nature of the southern Modoc forest. Even though selection of study
area was based on Jeffrey pine dominance, unmixed Jeffrey pine forests
occupy only a narrow band along the available gradient (Table 5).

On this gradient, P. ponderosa occupies deep soils with little surface
rock and shows definite north aspect preference. Juniperus occidentalis
occupies shallow soils with much surface rock and shows strong south
aspect preference. Unmixed stands of ponderosa pine strongly correlate
with soil 720. Jeffrey pine and pondeosa—Jeffrey stands occur on soil 720
but more often on soils with excessive drainage. Jeffrey pine dominance
relative to ponderosa pine correlates with south slope preference, greater
surface rock cover, and shallow soils. Juniperus occidentalis occupies a
variety of soil types especially, where it occurs alone, the heavy soils
(Table 5). Its dominance relative to Jeffrey pine correlates with greater
south slope preference, greater surface rock cover, and shallower soils.
Soil preferences are clear but broad overlapping ranges of tolerance per-
mit cohabitation over considerable area.

Vegetation of the southern Modoc Forest based on ground samples.
This study also involved comparison of the southern Modoc Forest with
other northern California forests having Jeffrey pine as a dominant ele-
ment. Ground-level vegetation samples from 5 forest areas of the south-
ern Modoc Forest indicate an average cover of 61% and an average tree
cover of 49% (Table 6). The 5 tree species and 12 shrub, brush and
suffrutescent species observed suggest low species diversity. The most
important species, on the basis of Importance Values, include 3 trees and
4 shrubs in the sequence Pinus jeffreyi, Cercocarpus ledifolius, Pinus |
ponderosa, Juniperus occidentalis, Artemisia tridentata, Purshia triden-
tata and Ceanothus prostratus (Table 6). Considerable variation in im- _
portance values suggests that few species are dominant in the region.

The occurrence of several vegetational phases is apparent from the
variation among sample areas (Table 6). Few species were observed in
samples areas II and V; each was dominated by 2 tree and one shrub |
species. However, a mesic forest of ponderosa and Jeffrey pine occurs in

VASEK: PINUS

1978]

of 1D £
tof

ID of
*(AL) 1D
of

Af of 49 1D
fof wD 1D
Ni¢

c (0)

AI

o3PIy MOTIIM

5

eq eg

AID Of f
eg eq

Ol Tie

Af

(ie)

ee

of

1D

ii

nrg fa

{ of eg 19

Nie

AOI

ZT

IIA wosuyo [

v

‘TOAOD UT MOT 3Nq “JUasaid sayeoTpuT (

is a
Id HV Of ID f
roars |
AV 1D £

odd 1D [
(Gd) 1D
(fOf) Id
{ of
(VV) £
{

Vv f

Hi

{ 1D eg AN
fA DV eg ID
Ne

AOI

o¢ ‘TE

ITH vepAry
2

era f A

ea fA

SULIdS 193807
@

eg eg
WIMATI SL

In eq OV
wd) MATL

AM

(A) £10
(CP) i 10
ROI
QV 1D
IOV Ark F

HY uy ig f A eg

eq f M
Af eat

P21 Ysy

I

) ‘aATVVALOP priqdy sqeqoid =

NOILVLADAA dV, HLIM NOILVTdYAOD) UNV LISadO4 IVNOILVN JOdOJ. NYAHLNOS AHL AO INAWOAS V NI

apod punosy

apoo dey
uoT}eUIUENS

8
A
2

S Sapor
bp U0T}e}039\
¢ UOI}e}939 A
Z a[dures
I punoiy
sapoy
U01}84a59 A
dep
diysumo J,

asuPry
U01}99S

Paly
ajdures

« T IQ®L Ul payst] are sapod satsadg ‘siinp
S€TdWVS GNNOU) JO NOILVIOYT “/ ATAV]

22 MADRONO [Vol. 25

area IT whereas a drier more open forest—woodland of J. occidentalis, P..
jeffrey and C. ledifolius occurs in area V.

Greater diversity of species in areas I, IV, and especially III appar-
ently correlates with considerable local topographic diversity and vegeta-
tional mosaics. A mesic Jeffrey pine forest in area I includes Abies con-
color and Calocedrus decurrens but also more xeric open patches of Arie-
misia tridentata, Cercocar pus ledifolius and Juniperus occidentalis. Areas
III and IV are drier, including open slopes, ridges and spurs, with a
variety of shrub—woodland elements and slopes forested with Jeffrey pine.

Comparison of methods. Comparison of ground observations (Table 6)
and the vegetation map discussed above suggests that different forests
may have been under consideration. However, the 2 sets of information
are based on methodologies differing in scale of observation, qualifica-
tions for inclusion, and extent of coverage.

To reconcile these differences, ground samples were correlated with
vegetation map units by determining their location on topographic maps
and soil vegetation maps, listing the map vegetation codes at those loca-
tions, and applying similar codes to ground vegetation samples (Table 7).
The 2 sets of vegetation codes still look somewhat different. However, if
the apparent greater heterogeneity of the ground samples, due to the
scale of observation, is ameliorated by summing the codes for several
samples in each sample area, and if bare ground symbols are added to
ground vegetation codes to allow for the different method of recording
ground cover, the 2 sets of summation codes (Table 7) are quite similar.

A few differences remain. At Ash Creek, Calocedrus is recorded in
ground but not in map samples, with a low cover value in one plot (Table
7). This difference is doubtless due to low density sampling variation and
perhaps to a sampling location at the edge of mapped vegetation. At
Foster Spring, ground samples indicate P. ponderosa on benches and gen-
tle slopes and P. jeffreyi in cold air drainage ways. However, ground
samples code exactly the same as vegetation map samples. Generalizing
to the scale of the map, therefore, does result in a slight loss of resolution.

At Hayden Hill, the vegetation map omitted J. occidentalis (probably
by error) and ground samples included only minimum P. ponderosa
cover. Observational notes between ground sample plots record occa-
sional cones of P. ponderosa on the ground and old stumps of lareg pon-
derosa trees. Therefore, there is some P. ponderosa in the area even
though significant occurrence was not recorded in sample plots. Depend-
ing on the time of lumbering operations, considerable P. ponderosa timber
could have been standing at the time of original mapping.

At Johnson Mill, practically identical codes prevail for both ground
and map samples.

At Willow Ridge, map codes list P. ponderosa and grass cover but
ground samples record P. jeffreyi and low total cover but not P. ponder-
osa (Table 7). Perhaps P. ponderosa was prevalent at the time of map-

1978] VASEK: PINUS 23

TABLE 8. CROWN COVER AND VEGETATION CHARACTERISTICS OF 3 SAMPLE AREAS
IN PLuMAsS NATIONAL Forest. I, Forest road $27, 10 mi. N of Beckwourth; II, 0.2
mi. E of Lake Davis, N of Portola; III, Frenchman Lake, 2 mi. W of dam. Head-
ings and notations as in Table 7.

Species Sample area Total
I II III D F Cc Te

Abies concolor 166° © 125.35 3.50 26 8 158.58 20.4
Calocedrus decurrens 9.50 189.86 Oe 7 199.36 22.4
Pinus ponderosa 102.49 154.95 55 6 257.44 28.4
Pinus jeffreyi 90.23 53.68 132.64 85 13 276.55 39.5
Juniperus occidentalis 166.18 0.16 5 4 166.34 1570
Amelanchier pallida 0.08 0.06 5.89 8 5 6.03 510
Arctostaphylos patula 32.00 5.66 18.01 62 12 5010Y 21.3
Artemisia tridentata 0.39 0.80 35.28 222 7 36.47 Bons
Cercocar pus ledifolius 4.53 16 2 4.53 3.9
Ceanothus velutinus 6.83 5.34 104.96 36 8. 14743 18.9
Ceanothus prostratus* 80.00 63.00 53.00 14 14 196.00 26.5
Eriogonum marifolium 0.42 35 2 0.42 6.0
Eriogonum ovalifolium 0.12 21 1 0.12 3.4
Haplopappus acaulis 0.07 3 if 0.07 12
Haplopappus bloomeri 4.42 80 5 4.42 14.2
Penstemon deustus 0.06 0.01 11 2 0.07 3.0
Phlox sp. 1.89 64 2 1.89 9.6
Purshia tridentata 2.43 0.12 3.21 44 12 5.76 15.9
Symphoricarpos acutus* 18.00 3.00 0.90 8 8 21.90 9.2

Total 516.88 601.98 389.89 822 119 1508.75 300.1
Number of plots 5 5 5 a5
Average cover 103.38 120.40 77.98 100.58

Average tree cover 74.01 104.80 32.84 63.88

ping and was subsequently lumbered out. The presence of P. ponderosa
on favorable soils about 14 mi distant might account for the discrepancy
between map and ground vegetation codes.

Grass cover values were not recorded at Willow Ridge. Grass patches
were noted but bare ground was more prevalent. For practical purposes,
in this comparison grass cover and bare ground are equivalent. Corre-
spondence between ground samples and map samples is rather close and
lends confidence to their complementary use in regional comparisons.

The Jeffrey Pine Forests of Plumas and Tahoe National Forests.
Ground level vegetation samples from 3 forest areas in Plumas National
Forest (Table 8) indicate an average cover of 100% and an average tree
cover of 64%. Both values are higher than comparable values in the Mo-
doc Forest suggesting a denser vegetation and more mesic conditions. The
most important species, on the basis of importance values, include 5 trees
and 6 shrubs in the sequence Pinus jeffreyi, Artemisia tridentata, Pinus
ponderosa, Ceanothus prostratus, Calocedrus decurrens, Arctostaphylos
patula, Abies concolor, Ceanothus velutinus, Purshia tridentata, Junip-
erus occidentalis and Haplopappus bloomeri.

24 MADRONO [Vol. 25

Fairly uniform importance values suggest that more than a few species
assume dominant roles in the region. The 5 tree species are the same as
those in the Modoc Forest. The 14 species of shrubs and suffrutescent
plants represent a net increase of 2 species in 15 sample plots as com-
pared with 12 species observed in 33 sample plots in the Modoc Forest.

Greater variety of species and increased dominance of more species,
especially shrubs, probably derives from greater gross environmental and
topographic variation and greater disturbance in the Plumas Forest.

Topographic variation ranges from the west slope position (high pre-
cipitation) of Davis Lake, and the high elevation (2000 m) and near
sub-alpine ridges of the mountains north of Beckwourth, to the low broad
ridges and volcanic tablelands at Frenchman Lake in a local rain shadow.
The general topographic diversity provides several different habitat sys-
tems in which different species assume roles of major importance. The
compositional differences among the 3 sample areas (Table 8) produce an
overall effect of fairly uniform average importance values for the region.

The relative amount of disturbance follows the same sequence: slight
disturbance at Davis Lake derives from the management practice of
cutting understory white fir; greater disturbance north of Beckwourth
derives from current logging, especially construction of logging roads;
much greater general disturbance from logging and burning over many
years is evident at Frenchman Lake. The variety of successional stages
apparent at Frenchman Lake contribute to marked diversity and general
openness of the vegetation despite a relatively mesic appearance (com-
pared to the Modoc Forest). Consequently, successional species like
Ceanothus velutinus, Arctostaphylos patula, and Ceanothus prostratus,
as well as generalists like Artesmisia tridentata, assume greater impor-
tance values than they do in the Modoc Forest. The more mesic condi-
tions in the Plumas Forest probably account for the lower importance
values of Cercocarpus ledifolius, Amelanchier pallida, and perhaps Pur-
shia tridentata.

The longevity of disturbance at Frenchman Lake is indicated by the
large size of Ceanothus velutinus clones (Zavitkovski and Newton 1968),
some of which reach 20 m in diameter. An instructive appraisal of the
rate of secondary succession might be derived from collation of clone size
and age distributions.

Vegetation samples from 5 forest areas in the Tahoe National Forest
(including one in Toiyabe N.F.) indicate an average cover of 76% and
an average tree cover of 66% (Table 9). This forest is fairly open but
considerable heterogeneity among sample areas is evident in cover and
composition (Table 9). Relatively mesic conditions are indicated south
of Truckee at high elevations by the dominance of A bies-concolor and the
presence of Pinus lambertiana. Relatively xeric conditions are indicated
south of Carson City at low elevations by the absence of all trees except

29

VASEK: PINUS

1978]

Os 59 80°09 coer Toes 98°0S OVS T9A9) 99.1] ABVIDAY
6S°SL eS L9 9¢ OTT 8v'6S 69°7S 60°¢8 JIAO) ABLIDAY
O¢ g 9 l 9 9 syoyd yo Joaquin
vo00¢ LO L977 Lvt 8eOl cs Obe L869 ve OLY EV ONS CS 86P [P10
8°6 $26 a a SO'L 06'T oro SNINID SOG ADIAOY FU S
oil OO'T é G O0'T DIMMAO{1]DI DSOY
Ve 99°0 v v £00 £9°0 212904 SaQLyy
OT T0°0 C c 10°0 D0 f{2U219IDA SNILINE)
e cv L6O9L SI LLC 06°¢Z 67°62 O8'¢ 86°61 DIDUsp1Ay DIYS AN
8b Te'T € 82 10'0 Oe ds x074d
TT OT'O T v OL'0 SNISnNap UOUWAasuag
SY LL0 ¢ 92 Z0'0 $20 249010019 Sngdndojdv
Gas c0'0 ¢ V4 Z0'0 SNSOISNDU SNUULDY JOSKAY J
S21 92°69 ta ra 10'y 00'9 ccc 00'LS 4SNIDAISOAG SNYJOUDID
6€ S6'V v 6 700 6'L SNIDINPAOD SNYJOUDAD
v'62 CSCC el £07 SI'vl 60°9 nae DIDIUIPIA] DISIMAIAL
O°9T 8S SOT L 99 10°0 LOZ 06°SOT pinjiog SojkygnjsoqIsy
CV £O°Z S ¢ 0 SISUAPDAAU SOJKYGDISOJIAP
Sv TT'0 9 L $0'0 90°0 Dpiyjog AatyIUDjaM P
a) 87112 8 ST vZ'00Z vO'TT SypjUuap1gI0 snsagiuns
6001 IL LL11 o¢ L672 7b'00¢ £0°69T IS Z9T S176 09° €SZ wasfal snurg
eC 9¢6T I jt 9f°61 SUadANIAP SNAPIIOIDY
T8¢ SOLUS CL 6S T8°TOS O¢'8 vol O9 Se 40109U09 SaIQYV
CT £16 T T €1'6 DUDIIAIQUD] SNULT
‘AT 9) A d A AI III II I
[®10.L eaie s[duies satoads

‘L IRL Ul se ssurpeayy “S6e “SQ JO M ‘Tu g'Z pue Aj UosIeD JO S ‘OS ‘SQ APMYSIET PIO—A ‘eyNIL Jo Ss ‘Tu g
‘PeOY YD WOg—A]I '68 AEMYSIP 99"1S JO | Tu J's pue | Tw TT ‘A ‘Tu ST ApaAtjoadsei pue vayonay, Jo N ‘Tu CT ynoqe ‘peoy sseg souudzyT
—III ‘II ‘I ‘Sisauoj IvNOIMVN @#0AIO], GNV OHV], NI SVTUY AIUWVS ¢ JO SOLLSIMALOVUVHD NOILVLANAA UNV VAYY YIAOD NMOUD “6 ATAVT

26 MADRONO [ Vol. 25

TABLE 10. VEGETATION AT DeviL’s GATE Pass, Mono County, CALirorniA, AUG
1, 1972. Importance values are based on 2 belt transects (A, B), each 50x2 m;
crown diameters of all shrubs, and trunk diameters of all trees over 2 m high, rooted
within the transect area, were measured; small trees were included in determining
frequency; otherwise, methods are as in Table 7 to 9; Ceanothus velutinus is also
common in the area but did not occur on the observed transects.

TY. TENG

Species A B
Pinus jeffreyi 131.8 70.3
Cercocar pus ledifolius 30.4 1235
Artemisia tridentata 104.2 70.6
Ribes viscosissimum 24.7 9.3
Leptodactylon pungens 8.8 37.6
Symphoricar pos vaccinoides 16.1
Purshia tridentata 15-5
Juniperus occidentalis 8.0

Total plants 47 150

TABLE 11. IMPORTANCE VALUES FOR 3 AREAS NEAR DEADMAN SUMMIT, MONO
County, Cair. AuG. 1 AND 2 1972. Based on 7 belt transects, 50x 2 m. I, Lower
mountain slopes; II, flat at base of mountain slopes; III, pumic flat, 4 mi. E of
summit; Pinus murrayana also was common in areas I and III but did not occur
on sample plots.

Species I 1G III
Abies magnifica 55.8 197.6

Pinus monticola 39.5 —

Pinus jeffreyi 204.7 102.4 86.6 150.8 142.9 112.2 97.8
Purshia tridentata 21348 - 122.32 67103:0 132.4 50.8
Leptodactylon pungens 26.9 54.3 — 7.0
Artemisia tridentata 42.8 88.7
Chrysothamnus parryi 12.7 414
Lupinus breweri 14.2

P. jeffreyi. Cold conditions are indicated for the general east Sierra region
south of Truckee by the absence of P. ponderosa (Haller, 1959). How-
ever, north of Truckee the east side ponderosa pines are morphologically
and physiologically like those of the interior Pacific Northwest (Haller,
pers. comm.). Primarily they are highly cold tolerant and compete well
with Jeffery pine. South of Truckee, where ponderosa pine is absent or
sparse (for reasons not apparent), P. jeffreyi assumes a role of strong
dominance, even more exaggerated than in the Modoc Forest. The cold
conditions and strong Sierra Nevada rain shadow probably combine to
eliminate other species selectively.

The most important species in the Tahoe Forest, based on importance
values are, in sequence Pinus jeffreyi, Purshia tridentata, Abies concolor,
Artemisia tridentata, Juniperus occidentalis, Arctostaphylo patula and
Ceanothus prostratus.

1978] VASEK: PINUS 27

An increase in the number of shrub and suffrutescent species, as com-
pared to the Modoc and Plumas forests, may be accounted for by the
wide elevational range and pronounced rain shadow. Disturbance by fire
was evident only in sample area I (Table 9) where Arctostaphylos
patula, Ceanothus cordulatus and Ceanothus prostratus are most promi-
nent (Skau et al., 1970).

Arctostaphylos nevadensis and Quercus vaccinifolia in sample plots
south of Truckee but not in sampled areas to the north indicate a high
elevation forest transitional to the Sierra Nevada red fir forest (Oosting
and Billings, 1943). Furthermore, I observed a few red fir (Abies mag-
nifica) in the area but not on the sample plots.

Jeffrey pine forests in Mono County. An open stand of Jeffrey pine at
Devil’s Gate Pass (Table 10) consists of a few large trees with a multiple
understory of Cercocarpus ledifolius, Artemisia, and Ribes. Still farther
south in Mono County, an extensive open forest of Jeffrey pine occurs on
the high elevation pumice plateau near Deadman Summit where open
ground and a shrub understory of Purshia and Artemisia are significant
(Table 11). Toward the base of the Sierra Nevada, Jeffrey pine and
Purshia increase (Table 11) to positions of strong dominance similar to
the sample from Toiyabe National Forest (Table 9, area V). On the
steep slopes, Jeffrey pine forests grade into dense forests of red fir with
an admixture of western white pine but with virtually no shrub under-
story (Table 12). Lodgepole pine (Pinus murrayana) is common in the
region but not in the few transects observed.

SUMMARY AND CONCLUSIONS

The southern Modoc National Forest was analyzed from vegetation—
soil map data that expanded ground observational coverage to over 500
km’. Information from the vegetation map agreed satisfactorily with
ground observations when scale of observation was accounted for. Thus,
vegetation map data provide a viable base for quantitative interpretation
of vegetation over large areas. The vegetation pattern is one in which
forests occur on favorable, deep soils, brush vegetation occurs on poor
soils in the basins, and intermediate soils are occupied by open forests
and woodlands. The observed gradient places Abies concolor and Calo-
cedrus decurrens in deep soils at the most mesic end of the vegetation
gradient. Pinus ponderosa, P. jeffreyi and Juniperus occidentalis occur
respectively along the gradient of decreasing soil depth and quality until
only brush species, Artemisia tridentata, and finally A. arbuscula, occur
on the shallowest, poorest soils.

Pinus jeffreyi occurs in pure stands, i.e., without other tree species,
with a frequency of only 16%, and P. ponderosa occurs in pure stands
with a frequency of only 8%. More often they occur in mixed forests and
Jeffrey pine also occurs with Juniperus occidentalis in open forests or
woodlands.

28 MADRONO [ Vol. 25

The pattern of vegetational distribution is mediated by soil character-
istics but complicated by other factors such as successional status and
management practices. Woody shrubs may have roles in secondary suc-
cession and their fairly low frequencies except on volcanic ridges suggests
that the Modoc Forest is generally stable and relatively undisturbed.
Logging over the years may have influenced some compositional changes
in the forest. Primarily, Pinus ponderosa seems less prominent now than
indicated on the vegetation map compiled in 1953, at least in 2 specific
localities. Such compositional changes, mediated by lumbering, render
difficult the interpretation of the forest in terms of natural plant com-
munities. Nevertheless, those forests are the de facto communities and
bear considerable resemblance to pristine vegetation.

The southern Modoc Forest was compared to forests with Jeffrey pine
in the Plumas and Tahoe Forests to the south. Each forest has its range
of characteristics with respect to cover, important species, diversity of
species and successional status.

In broad terms, Jeffrey pine forests range along at least 2 general gra-
dients, both modified by altitudinal variation. One gradient proceeds
essentially west—east and is influenced by decreasing moisture associated
with the Sierran rain shadow. Depending on elevation, the more mesic
forest to the west is either a red fir or Sierra Nevada ponderosa pine
forest. The more xeric vegetation to the east is either a western juniper
woodland or Great Basin sagebrush.

The second gradient runs north-south and is probably primarily influ-
enced by temperature. The major range for Jeffrey pine occurs to the
south where it occupies a broader elevational range (Haller, 1959).
Jeffrey pine as a forest type stops north of the southern Modoc Forest
study region. Still further north Jeffrey pine occurs mainly as an element
of ponderosa pine forests. However, patches of Jeffrey pine occur in the
Warner Mountains (Milligan, 1969) in forests considered to be depau-
perate Sierran forests by Critchfield and Allenbaugh (1969). Vegetation
of the Warner Mountains is considered to be transitional to the Great
Basin (Milligan, 1969) and cold dry conditions appear limiting to the
northeast. To the northwest, however, excessive moisture may be lim-
iting. Haller (1959) indicates more rapid growth by ponderosa than
by Jeffrey pine in areas of ample moisture. The southern Oregon limit
for Jeffrey pine on serpentine (Whittaker, 1960) may be instructive in
this regard. Thus, the northern range of Jeffrey pine seems limited where
conflicting moisture-temperature gradients meet and are superceded by
other factors.

The narrow occurrence of unmixed Jeffrey pine on a broad gradient in
the Modoc Forest may be a consequence of its position near the northern
limits of its range. Therefore, relative breadth of occurrence on a gradient
toward the center of the range of distribution should be determined in
addressing the question of whether Jeffrey pine forests merit widespread

1978] VASEK: PINUS 29

recognition as a separate community or whether they represent merely
a phase of the mixed conifer forest.

ACKNOWLEDGMENTS
Financial support from the Academic Senate Committee on Research, University
of California, Riverside is gratefully acknowledged. I appreciate the comments of
J. R. Haller, P. W. Rundel and W. R. Powell on early versions of the manuscript.

LITERATURE CITED

CoLtweELL, W. L. 1974. Soil-vegetation maps of California. U.S.D.A., Forest Service
Res. Bull. PSW-13.

CRITCHFIELD, W. B. and G. A. ALLENBAUGH. 1969. The distribution of Pinaceae in
and near northern Nevada. Madrofto 20:12-26.

FRANKLIN, J. F. and C. T. Dyrness. 1973. Natural vegetation of Oregon and Wash-
ington. U.S.D.A. Forest Service Gen. Tech. Report PNW-S8. U.S. Govt. Printing
Office, Washington, D.C.

GRIFFIN, J. W. and W. B. CrircHFIELp. 1972. The distribution of forest eres in
California. U.S. Forest Service Res. Paper PSW-82. Pacific S.W. Forest and
Range Exp. Sta., Berkeley.

Hatter, J. R. 1959. Factors affecting the distribution of ponderosa and Jeffrey
pines in California. Madrofo 15:65-96.

————. 1961. Scme recent observations on ponderosa, Jeffrey and Washoe pines
in northeastern California. Madrono 16:126-132.

. 1962. Variation and hybridization in ponderosa and Jeffrey pines. Univ.
Calit. Publ. Bot. 34:123-166.

Horton, J. S. 1960. Vegetation types of the San Bernardino Mountains. U.S. Forest
Service, Pacific S.W. Forest and Range Exp. Sta. Tech. Paper 44.

Kryver, F. D. 1931. Major plant communities in a transect of the Sierra Nevada
Mountains of California. Ecology 12:1-17

Miriican, M. T. 1969. Transect flora of Eagle Peak, Warner Mountains, Modoc
County. M.S. Thesis. Humboldt State College, Arcata.

MinnicuH, R. 1976. Vegetation of the San Bernardino Mountains. Jn Latting, J.
(ed.), Proc. symposium on the plant communities of southern California. Calif.
Native Plant Soc. Spec. Publ. 2.

MUveELLER-Dompotrs, D. and H. ELLeNBERG. 1974. Aims and methods of vegetation
ecology. John Wiley & Sons, New York. 547 pp.

Mownz, P. A. and D. D. Keck. 1949. California plant communities. Aliso 2:87-105.

SkAu, C. M., R. O. Meeuwic and T. W. Townsenp. 1970. Ecology of eastside
Sierra chaparral. Nevada Agr. Exp. Sta. Bull. R71. 14 pp.

SOCIETY OF AMERICAN Foresters. 1954. Forest cover types of North America.
Washington, D.C. 67 pp.

Tuorne, R. F. 1976. Plant communities of California. In Latting, J. (ed.), Proceed-
ings of a symposium on the plant communities of southern California. Calif.
Native Plant Soc. Spec. Publ. 2.

TwIssELMAN, E. C. 1967. A flora of Kern County. Wasmann J. Biol. 25:1-395.

U.S.D.A., Forest Service. California [now Pacific Southwest] Forest and Range
Experiment Station. 1953a. Vegetation-soil maps of California. Lassen County:
lobed; 17C-1 to -4- 17D-2,3; 20B=1,2-21A-1,2.

. 1953b. Legends and supplemental information to accompany vegetation-

soil maps of California. Lassen County: 16D-4; 17C-1 to 4; 17D-2,3; 20B-1,2;

21A-1,2.
. 1958. Soil vegetation surveys in California. (Revised 1969). State cooper-
ative soil-vegetation survey. Berkeley, California.

30 MADRONO [Vol. 25

VASEK, F. C. 1966. The distribution and taxonomy of three western junipers. Brit-

tonia 18:350-372.
and R. F. THorNeE. 1977. Transmontane coniferous vegetation. Jn Major,

J. and M. G. Barbour (eds.). Terrestrial vegetation of California. Wiley Inter-
science, New York.

VocL, R. J. and B. C. MILER. 1968. The vegetational composition of the south
slope of Mt. Pinos, California. Madrofio 19:225-234.

WHITTAKER, R. H. 1960. Vegetation of the Siskiyou Mountains, Oregon and Cali-
fornia. Ecol. Monogr. 30:279-338.

ZAVITKOVSKI, J. and M. Newron. 1968. Ecological importance of snowbrush,
Ceanothus velutinus, in the Oregon Cascades. Ecology 49: 1135-1145.

THREE NEW SPECIES OF JATROPHA (EUPHORBIACEAE)
FROM WESTERN MEXICO

BIJAN DEHGAN AND GRADY L. WEBSTER
Department of Botany, University of California, Davis 95616

In the course of field studies for an infrageneric revision of the genus

Jatropha (Dehgan, 1976), three Mexican collections—two from Baja
California and one from Jalisco—were found to differ from all species
previously described from these areas (Mueller, 1866; Pax, 1910; Stand-
ley, 1923; McVaugh, 1945a, 1945b; Wilbur, 1954; Shreve and Wiggins,
1964). Following are descriptions of these taxa and comparisons with
related species. Since all three species were either dormant or at least
not flowering at the time of collection, cuttings and/or seedlings were
grown to maturity in the greenhouse. The descriptions that follow are
therefore based primarily on these greenhouse plants. _ |

1. Jatropha giffordiana Dehgan and Webster, sp. nov. sect. Loureira
subsect. Canescentes ; a J. canescenti et J. cinerea differt foliis glabrius-
culis exstipulatis, inflorescentia ¢ florem @ solitarium efferenti, petalis
rubris, pistillo 3-loculato, stylis 2 stigmatibus multifidis (Figs. 1-7).

Shrub ca 1-1.5 m high; caudex thickened, stem and branches succu-
lent, branches spreading and decumbent; bark fissured and peeling on
older branches; short shoots numerous and distinct on older branches,
2—5 cm long, pubescent, with small leaves crowded near the apex. Leaf-
with—petiole 3.5—-8.5 cm long, 3-7 cm wide; stipules not evident; blade
broadly ovate, unlobed, entire (completely devoid of glands), pubescent
on the veins abaxially but otherwise glabrous, palmatinerved with five
prominent and two weaker lateral veins, cordate at base, acute at apex.
Inflorescences gynodioecious, terminal on branches or more often on
short shoots, the plants pistillate and staminate; in the staminate race-
mose-paniculate, with many flowers and one central @ flower; pistillate
inflorescence racemose, of 2—5 flowers; axis villose, ca 5—12 cm long in the
male and 1.5—2.5 cm in the female; paracladia mostly 2-4 cm long in

DEHGAN & WEBSTER: JATROPHA 3

1978]

2: 3%
Detail of the stami-

locular fruit from the single

)

2. Pistillate flower
x4.

Growth habit.

t.

Jatropha giffordiana.

Fics. 1-7.
Detail of the pistillate flower, X3. 4. Stamina

5,

)

te flower

oe

nate flower, X4. 6. Staminate inflorescence with a

Cleared leaf, X1.

ie

X1.

ob}

tillate flower at the center

°

pis

32 MADRONO [Vol. 25

the male; lower bracts entire (not glandular), narrowly lanceolate, vil-
lose, ca 1-2 mm long in the male and 3—7 mm long in the female. Stamin-
ate flower urceolate, ca 8-10 mm long and 5—7 mm wide, with densely
villose pedicel ca 3-4 mm long; calyx lobes 5, elliptic, somewhat pointed,
entire (not glandular), ca 2-3 mm long and 1—1.5 mm wide, not imbri-
cate; petals connate to more than 2% their length, bright red, pubescent
adaxially; disc segments 5, spherical, massive, ca 1 mm high and wide;
stamens 10, monadelphous-biseriate, filaments connate to more than 24
their length, ca 5-8 mm long, anthers elliptic, ca 1.3—1.5 mm long. Pistil-
late flower campanulate, ca 10-12 mm wide (somewhat wider in the
staminate inflorescence), with densely villose, large and more or less
foliaceous calyx lobes ca 15-20 mm long, 10-13 mm wide, not imbricate;
petals connate to about the middle, bright red, villose adaxially; disc
dissected, of (8) 10 glands, somewhat taller than broad, ca 0.7-1.0 mm
wide and 0.8—1.1 mm high; ovary glabrous, of (2—) 3 carpels (always
three in the terminal pistillate flower of an otherwise staminate inflor-
escence) ; stylar column thick, connate to above the middle, dilated to 2
multifid stigmata. Capsules distinctly (2-) 3-lobed, ca 2—2.5 cm long
and wide, somewhat dry, tardily dehiscent; seeds grayish-brown, more
or less spherical, 1—1.5 cm long and wide, caruncle vestigial.

TyPer: Baja California, Cabo San Lucas, on dunes near the beach and
facing the rocky hill, 17 Mar 1974, Dehkgan B74.019. The population
consisted of 20-25 plants. Type specimens from greenhouse-grown plants
are deposited at DAV.

This species is named in honor of Professor Ernest M. Gifford for his
contributions to the morphology of vascular plants in general and of
shoot apices in particular.

Jatropha giffordiana is closely related to J. canescens and J. cinerea
(sensu lato) as shown by several common characters in their vegetative
morphology. McVaugh (1945b), with some justification, relegated J.
canescens to synonymy under J. cinerea and in fact referred to it as a
“race” of the latter. Taxonomic evidence presented elsewhere (Dehgan,
1976), however, shows the two species to be closely related but distinct.
The intermediacy of certain populations (‘“‘races” fide McVaugh) in Baja
California and western Mexico probably results from introgressive hy-
bridization between the two taxa. Jatropha canescens (sensu stricto) is
a disjunct relictual species otherwise occurring as isolated populations
in northern parts of the Sonora Desert and in Magdalena Island (see
Dehgan, 1976 for discussion). The extremely variable Jatropha cinerea
(sensu lato), on the other hand, is widespread in western Mexico, par-
ticularly in coastal areas and from San Felipe southward in Baja Cali-
fornia, but not on Magdalena Island.

Jatropha giffordiana differs from both J. canescens and J. cinerea in
its decumbent growth habit, gynodioecious inflorescence, absence of
stipules, bright red petals, and mainly 3-carpellate gynoecia with multifid

1978] DEHGAN & WEBSTER: JATROPHA 25)

stigmata. In fact, the multifid stigma is unique for the entire genus. The
occurrence of J. giffordiana as a small population apparently endemic
to the tip of Baja California, a more tropical and moister region below
the Tropic of Cancer, is also of some significance, since it may be an
ancestral relict.

The following key describes extremes of J. canescens-cinerea complex
and disregards intermediate individuals or populations as discussed by
McVaugh (1945b). Further clarification of this species complex, includ-
ing J. giffordiana, requires detailed biosystematic studies of various
populations. No populations intermediate between J. giffordiana and the
other two species have been observed. The descriptions of the two species
as given by Pax (1910), particularly that of J. canescens, are quite inac-
curate; and those of McVaugh (1945b) and Shreve and Wiggins (1964),
for the reasons given above, are somewhat confusing. And finally, it
should be emphasized that similarity of the vegetative characters, espe-
cially of the herbarium material, can be quite deceiving. The following
somewhat deliberately detailed diagnostic key should, we hope, clear up
the long-standing confusion with regard to the distinctiveness of the taxa
under consideration.

a. Gynodioecious shrubs with decumbent branches and distinct
short shoots; stipules absent; leaves sparsely pubescent on the
veins, otherwise glabrous; staminate inflorescence racemose-
paniculate with a single pistillate flower at the center; pistillate
flowers campanulate, the staminate urceolate; 2 calyx lobes
15-20 mm long, 10-13 mm wide; petals bright red; stylar col-
umn thick, dilated to 2 multifid stigmata . . . J. giffordiana.

a. Dioecious erect shrubs with or without distinct short shoots;
stipules early deciduous or persistent; leaves pubescent on one
or both surfaces; staminate inflorescence paniculate and lax or
nearly subsessile and compact without the central pistillate
flower; flowers subglobose or tubular in one or both sexes; @
calyx lobes 8-12 mm long, 6-10 mm wide; petals greenish-
yellow or pink-rose; stylar column narrow, dilated to 2 (rarely
oy indiStiomatas: “2 lo. ss © 2 wos im oe ec aoe = Dd,

b. Nodes swollen and with very short arrested shoots; stipules
linear-lanceolate and persistent; leaves orbicular, unlobed, en-
tire and densely pubescent on both surfaces; male inflorescence
subsessile, compact and many flowered; flowers of both sexes
+ tubular; petals greenish-yellow . . . . . . J. canescens.

b. Nodes not swollen and without arrested shoots; stipules lanceo-
late, early deciduous; leaves ovate-orbicular, often 3—5 lobed,
with glandular margin when young, pubescent only on the
adaxial surface; male inflorescence pedicellate, paniculate and
lax; staminate flowers tubular but pistillate flowers subglobose;
petals pink-rose . ......... . . . J. cinerea.

34 MADRONO [Vol. 25

Specimens of Jatropha canescens examined: SONORA: thorn scrub on
lava, flat at base of slope, 15 mi SE of Guaymas, 21 Jun 1972, Webster
é Lynch 17002 (DAV). SINALOA: cactus thorn scrub on silty soil,
plains ca 8 mi NW of Guamuchil, alt 20 ft, Webster & Lynch 17038
(DAV). Baja Cat. Sur: Magdalena Island, cactus thorn scrub on silty
soil, foothills, alt ca 100 ft, 20 Mar 1974, Dehgan & Webster 874.038
(DAV).

2. Jatropha moranii Dehgan and Webster, sp. nov. sect. Platyphyllae;
a J. purpurea differt foliis exstipulatis 5-lobatis, bracteis eglandulosis,
sepalis integris foliaceis, petalis chloroleucis recurvatis (Figs. 8-13).

Small shrub less than 1 m high with succulent stem and branches and
a distinct woody caudex; bark fissured but not peeling, brown with white
epidermal markings; short shoots not evident. Leaf-with—petiole ca 2.5-
5.5 cm long and 1.5—3 cm wide; stipules not evident; blades ovate, dis-
tinctly 5-lobed, the margin ciliate with knob-shaped stipitate glands 2—4
mm long, papillose abaxially but papillose and hirsute adaxially particu-
larly near the margins; palmately netted with 5 prominent veins, cordate
at the base, cuspidate at the apex. /nflorescence monoecious, subterminal
(occasionally appearing terminal or lateral); dichasia compound, para-
cladia of 1-2 dichasia each terminating in a pistillate flower; coflores-
cence present and often distinct; inflorescence axis downy, ca 2—4.5 cm
long, axes of paracladia ca 1-1.5 cm long; lower bracts entire (rarely
with 1 or 2 glands), hirsute, lanceolate, ca 3-7 mm long. Stamuinate
flowers subglobose, ca 8-12 mm long and 6—9 mm wide, with downy short
pedicel 4-6 mm long; calyx lobes 5, elliptic, pointed, entire (not glandu-
lar), downy, ca 5—9 mm long and 2—5 mm wide, imbricate at base; petals
recurved, connate to about % their length, white, glabrous on both sur-
faces; disc segments 5, spherical, massive, ca 1—-1.5 mm high and wide;
stamens 10, monadelphous-biseriate, connate for most of their length;
filaments ca 5-8 mm long; anthers elliptic, 1-1.3 mm long. Pistillate
flowers campanulate, ca 10-14 mm long and 12-18 mm wide, with downy
pedicel 8-11 mm long; calyx lobes 5, broadly elliptic, pointed, entire
(not glandular), papillose on both surfaces, large and more or less foli-
aceous, ca 10-25 mm long and 10-15 mm wide, not imbricate; petals
recurved, connate below the middle, white, glabrous on both surfaces;
disc segments 5, broader than tall, ca 1.5-2.5 mm wide and 0.8-1 mm
high; stylar column thick, connate to above the middle, dilated into 3
bifid stigmata. Capsules trilocular and distinctly trilobed, ca 1.5-2 mm
long and wide, somewhat dry, tardily dehiscent; seeds grayish brown,
more or less spherical, ca 1-1.5 cm long and slightly less wide, the ca-
runcle lacerate.

Types: Baja California Sur, Cabo San Lucas, 6 Aug 1932, John
Thomas Howell 10606 (CAS, holotype).

1978] DEHGAN & WEBSTER: JATROPHA O08

Fics. 8-13. Jatropha moranii. 8. Growth habit (greenhouse grown plant). 9. In-
florescence and leaves, ca. X1. 10. Pistillate flower, X3. 11. Detail of pistillate flower,
X6. 12. Staminate flower, X5. 13. Detail of staminate flower. XS.

Additional collection: Cabo San Lucas, above the air strip, alt. ca 10
m, on alluvial rocky areas, Jul 1968, John E. Bleck & Charles Glass
680—Dehgan B74.052 from greenhouse grown plants (DAV).

This species is named after Dr. Reid Moran of the San Diego Natural
History Museum for his contributions toward the understanding of the
flora of Baja California.

Jatropha moranii resembles J. purpurea in growth habit and in struc-
ture of the inflorescence. However, several characters suggest that the

36 MADRONO [Vol. 25

two may not even be closely related. Jatropha moranizi is quite distinct in
its lack of stipitate glands on bracts and calyx lobes and its lack of stip-
ules and its recurved white petals. Since we have seen no specimens of
J. purpurea in Baja California and no herbarium specimens from Baja
California, we suggest that references to J. purpurea in Baja California
(e.g., by Standley, 1923; Shreve and Wiggins, 1964) may be based on J.
morant. These two species can be distinguished by the following synoptic
key:
Stipules dissected into gland-tipped segments; petiole slender,
20-40 mm long; leaf blade 3—4 cm wide, shallowly 3-lobed, the
median lobe narrowly triangular and much longer than the
lower lobes, upper margins of lobes dentate but less conspicu-
ously glandular than the basal margin. Bracts glandular-ciliate;
sepals glandular-ciliate, linear-lanceolate in the pistillate flower ;
petals red, not recurved . . . . J. purpurea.
Stipules ingens petiole stout, 5 18 mm lone leat blade 1.5-3
cm wide, shallowly 5-lobed, the median lobe widely triangular
and nearly like the other lobes, margins evenly glandular
throughout. Bracts mostly without glands; sepals entire (not
glandular), broadly elliptic and more or less foliaceous in the
pistillate flower; petals greenish-white, recurved . . J. moranit.
Specimen of J. purpurea examined: SINALOA: Cerro Llano Redondo,
west of Caymanero, 8 Oct 1944, Howard Scott Gentry 7088 (DS).

3. Jatropha mcvaughii Dehgan and Webster, sp. nov. sect. Curcas; a J.
curcas differt ramis cortice fissurato, foliis pubescentibus, inflores-
centiis dioeciis, petalis in dimidio inferiore connatis, stylis crassis,
seminibus 10-12 mm longis non incrustatis (Figs. 14-18).

Syn. Jatropha curcas var. rufus McVaugh, Bull. Torrey Bot Club 72:

284. 1945.

Shrub or small tree, 1.5-3.5 m high; branches and foliage pubescent;
bark fissured or cracked, but not peeling. Leaf-with-petiole 25-35 cm
long when mature; stipules narrowly lanceolate, early deciduous; blades
ovate, 5-7 (9) lobed with the upper lobes extending to near the mid-
rib and lower lobes shallow, mostly 15-25 cm long and nearly as wide,
palmatinerved, with 7(9) primary nerves, broadly cordate at the base,
cuspidate at the apex, pubescent on both surfaces; margins entire (com-
pletely devoid of glands). /uflorescence dioecious, terminal on branches,
with typical jatrophoid compound dischasia in both sexes but with much
smaller number of flowers in the pistillate inflorescence; axis tomentose,
ca 7-12 cm long in the staminate and 3-6 cm in the pistillate; paracladia
terminating in a single flower; those of the pistillate inflorescence ca
1—2.5 cm long, but those of the staminate inflorescence 1—1.5 cm long;
lower bracts entire, lanceolate, pubescent, 4-9 mm long in the male,

1978] DEHGAN & WEBSTER: JATROPHA ci)

Fics. 14-18. Jatropha mcvaughii. 14. Pistillate inflorescence, X3. 15. Detail of
pistillate flower, X4. 16. Fruit, ca X1. 17. Staminate inflorescence, X4. 18. Detail of
staminate flower, X8.

somewhat longer in the female. Staminate flowers + tubular with corolla
tube longer than the lobes, 8-12 mm long and 5-8 mm wide; pedicel
tomentose, ca 4-8 mm long; calyx lobes 5, elliptic, pointed, entire (not
glandular), smaller than in the female, ca 4-7 mm long and 2-3 mm
wide, imbricate; petals obovate, connate to about /% or more of their
length, greenish-yellow, hirsute adaxially, villose abaxially; disc seg-
ments 5, massive, ellipsoid, ca 2—3.5 mm long and 1-1.5 mm wide; sta-
mens 10, monadelphous, scarcely b/seriate, the filaments connate to
about ™% their length, ca 3-5 mm long; anthers oblong elliptic, some-
what flattened at the apex, ca 1.5—2 mm long. Pistillate flowers ca 5-11
mm long, with tomentose pedicel, =_ campanulate, corolla tube shorter
than or nearly equalling lobes, ca 8-11 mm long and 10-14 mm wide;
calyx lobes 5, elliptic, pointed, entire (not glandular), ca 7-10 mm long
and 3.5-5 mm wide, imbricate; petals obovate, connate to about '4 their
length, greenish-yellow, hirsute adaxially, villose abaxially; disc segments

38 MADRONO [Vol. 25

5, massive, broader than long, ca 1-1.3 mm long and 2.5-4 mm wide;
ovary glabrous, of 3 carpels; stylar column thickened, connate to about
middle, not dilated but with 3 bifurcate, massive, dark green stigmata.
Capsules ellipsoidal, ca 2 cm long and 1.5 cm broad, + fleshy, at length
drying and tardily dehiscent; seeds light brown, 10-12 mm long and
8-10 mm wide, the caruncle appressed to the beak and nearly vestigal,
ca 1 mm or less long and 1.5—2 mm wide.

Type: Mexico, Jalisco, Playa Scandida, Dec 1974, Dehgan B74206
(DAV). The species has been observed from Mazatlan to an elevation of
ca 350 m on the road to Durango.

Additional specimens examined: SINALOA: Ymala, Aug 1891, Palmer
1413 (US, holotype); Mazatlan, Ortega 7299 (CAS); Culiacan and
vicinity, Howard Gentry 7046 (CAS). Nayarit: thorn woodland 11-12
mi NE of Singayta, alt ca 200 ft, 25 Jun 1972, Webster & Lynch 17070
(2), 17073 (3!) (DAV).

This species is named in honor of Professor Rogers McVaugh, in recog-
nition of his contributions toward understanding of the genus Jatropha
in particular (1944, 1945a, 1945b) and to the systematics of neotropical
flowering plants in general.

Jatropha mcvaughii was earlier described as J. curcas var. rufus by
McVaugh (1945b). Although originally distinguished by McVaugh solely
on the basis of pubescence, it actually differs from J. curcas in a number
of characters: fissured bark, dioecious inflorescences, longer corolla tube,
thick stylar column with undilated stigmata, and smaller smoother seeds.
Although the color and quantity of pubescence does furnish a convenient
recognition feature for J. mcvaughi, the dioecious flower production
seems systematically more important. We thus conclude that while J.
mcvaughii clearly belongs to sect. Curcas and is closely related to J.
curcas, it differs sufficiently from the latter to be considered a distinct
species, as is evident in the following synoptic key:

Bark smooth, branches and mature foliage glabrous; leaf blades
unlobed or with (3)5—7 very shallow lobes; monoecious, bi-
sexual or often unisexual; petals greenish or yellowish-white,
connate at the base; styles slender, dilated into massive stig-
mata; capsule ca 3 cm long and 1.5 cm broad; seeds 15-32 mm
long, blackish-encrustate-striate . . . sy Oe Cureass
Bark fissured or cracked, branches and TELL foliage pubes-
cent; leaf blade with 5-— 7( —~9) deeper lobes; dioecious; petals
piecnich= elton, connate to about half their iene Beil thick,
undilated, with fleshy stigmata; capsule ca 2 cm long and 1 cm
broad; seeds 10-12 mm long, light brown and without striations

J. mcvaughnu.

1978] BOHM & ORNDUFF: JEPSONIA 39

LITERATURE CITED

Deucan, B. 1976. Experimental and evolutionary studies of relationships in the
genus Jatropha L. (Euphorbiaceae). Ph.D. Dissertation, Department of Botany,
University of California, Davis, California. 436 pp.

McVaucu, R. 1944. The genus Cnidoscolus: generic limits and intrageneric groups.
Bull. Torrey Bot. Club. 71:457-474.

. 1945a. The Jatrophas of Cervantes and of the Sessé and Mocinho Her-
barium. Bull. Torrey Bot. Club 72:31-41.

. 1945b. The genus Jatropha in America: principal intrageneric groups.

Bull. Torrey Bot. Club 72:271-294.

MUELLER ARGOVIENSIS, J. 1866. Jatropha. In De Candolle (ed.), Prodromus Syste-
matis Naturalis Regni Vegetabilis 15(2) :1076-—1105.

Pax, F. 1910. Euphorbiaceae-Jatropheae. Jn A. Engler (ed.), Das Pflanzenreich
IV.147(Heft 42) :1-148. Verlag Wilhelm Englemann, Leipzig.

SureveE, F. and I. L. Wiccrns. 1964. Vegetation and flora of the Sonoran Desert.
Stanford University Press, Stanford, California. Vol. I: x + 840 pp.

STANDLEY, P. C. 1923. Jatropha. In Trees and shrubs of Mexico. Contr. U.S. Nat.
Herb. 23:634-642.

Wixeur, R. L. 1954. A synopsis of Jatropha, subsection Eucurcas, with the descrip-
tion of two new species from Mexico. J. Elisha Mitchell Sci. Soc. 70:92-101.

CHEMOTAXONOMIC STUDIES
IN THE SAXIFRAGACEAE S.L.
9, FLAVONOIDS OF JEPSONIA

Bruce A. BoHM
Department of Botany
University of British Columbia, Vancouver V6T 1W5 Canada
ROBERT ORNDUFF
Department of Botany, University of California, Berkeley 94720

Jepsonia is a small genus of the Saxifragaceae restricted to California
and northern Baja California. Ornduff (1961) described the distylous
nature of the flowers and, more recently, presented a detailed account of
the ecology, morphology and systematics of the genus (Ornduff, 1969).
No chemical study of the genus appears to have been done. An investi-
gation of the polyphenolic constituents of Jepsonia was thus undertaken
as part of a general chemotaxonomic survey of the family. It was hoped
that flavonoid data might yield additional characters useful for charac-
terizing the species and offer insights into the enigmatic relationships
between Jepsonia and other genera in the family.

MATERIAL AND METHODS
The plant collections used in this study are: J. heterandra FEastw.,
Bagby, Mariposa Co., Cal., 20 Mar 1970, G. D. Cromwell 101, RSA;
J. malvifolia (Greene) Small, Santa Catalina Island, Los Angeles Co.,
Cal., 8 Mar 1970, R. F. Thorne 39392, RSA; J. parryi (Torr.) Small,

40 MADRONO [Vol. 25

Old Mission Dam, San Diego Co., Cal., 12 Apr 1970, Cromwell 107,
RSA; Old Mission Dam, San Diego Co., Cal., 2 Nov 1970, Cromwell
109, RSA; Los Alamos Canyon, Riverside Co., Cal., 15 Feb 1972, Crom-
well 709, RSA; Jatay, Baja Cal., 21 Mar 1970, Thorne 39421 RSA;
Baja Cal., Spring 1973, Tom Mulroy (Pomona College) (s.n.) RSA.

Flavonoid constitutents were isolated and purified by the procedures
described by Wilkins and Bohm (1976a). The compounds were identified
by chromatography against standards, partial and total hydrolyses, and
ultraviolet spectral methods (Mabry et al., 1970). These procedures
were used on the pooled extracts of J. parryi and the extract from J.
heterandra. Compounds present in J. malvifolia were determined solely
on the basis of chromatography against standards because of the small
amount of plant material which was available.

RESULTS

The flavonoids of Jepsonia are based upon the common flavonols
kaempferol, quercetin, and myricetin (Table 1). All compounds are
3—0-glycosylated derivatives. The compounds indicated as ‘““K—acyl”’ and
“Q-acyl” are gallic acid esters of the corresponding kaempferol—3—0-
glucoside and quercetin—3—O—glucoside. The quantities available were
too small to allow detailed study but the derivatives have chromato-
graphic characteristics and color test behavior identical to those of
the flavonol—3—O—glucoside—6’’—gallyl derivatives identified in Tellima
(Collins et al., 1975) and Heuchera (Wilkins and Bohm, 1976a, and
unpubl.). Gallotannins were shown to be present by chromatography and
characteristic color reaction using ferricyanide reagent but very limited
material precluded further study.

One population of Jepsonia parryi was sampled in the autumn and in
the spring and both collections were chemically identical. The flavonoid
profiles of these samples were identical although there were differences
in the relative concentrations of the compounds. No significance can be
attributed to this quantitative variation without extensive additional
sampling of this species.

DISCUSSION

Taxonomic opinion has been divided on the number of species of
Jepsonia. The genus was considered to be monotypic by Jepson (1925,
1936) and Munz (1959) although its “polymorphous” nature was rec-
ognized by Munz (1959). At the other extreme Small and Rydberg
(1905), Bacigalupi (1944) and Ornduff (1969) recognized three species.
Ornduff (1969) based his taxonomic conclusions primarily on an array
of morphological traits that separate the three species and on hybridiza-
tions among the species. He pointed out the importance of studying liv-
ing material in the identification of species of Jepsonia.

We undertook study of the flavonoids of Jepsonia with the hope that

1978] BOHM & ORNDUFF: JEPSONIA 41

TABLE 1. FLAVONOIDS AND A GALLIC AciD DERIVATIVE OF JEPSONIA.

Je Je I.
parryi malvifolia* heterandra

Kaemferol—3—O0-rhamnoside + + 24
Kaempferol—3—0—glucoside + at ae
Kaempferol—3—0-galactoside trace ND? —
Quercetin—3—0—glucoside + = at:
Myricetin—3—0—glucoside + = ae
Kaempferol—3—0-rutinoside + + Js
Kaempferol—3—0-xylosylxyloside + + a
Quercetin—3—0-rutinoside + + ales
Quercetin—3—0—xlyosylxyloside + + +
Kaempferol—acyl° + + S:
Quercetin—acyl* + ay _
Gallotannin test* + ae a

“) determined by comparative chromatography only.

») ND = not determined.

“” kaempferol- and quercetin—3—0-glucoside—?”—gallate.
” blue coloration with ferricyanide reagent.

additional characters of systematic value would be found that might
shed light on the relationships among the species and of the genus (cf.
Wilkins and Bohm, 1976a; Bohm and Wilkins, 1976). This study showed
that there are very few differences among the species of Jepsonia in their
flavonoid biochemistry. Jepsonia parryi and J. malvifolia, both occurring
in southern California and Baja California, have an identical array of
flavonoids. Jepsonia heterandra from the foothills of the central Sierra
Nevada differs from these two species in that it lacks kaempferol—3—0-
rhamnoside, kaempferol—3—0-—galactoside and the 6’”—0-gallyl derivative
of the flavonol glucosides. Kaempferol—3—0—galactoside occurs only as a
trace constituent of J. parryi; it was not sought in J. malvifolia due to
lack of plant material.

Two types of gallic acid derivatives occur: the 6’—0-gallylated flavo-
nol glucosides and an unidentified compound indicated only as positive
“gallotannin test”. The capacity to make gallic acid derivatives charac-
terizes the genus; the nature of the derivatives appears to be useful in
assessing relationships.

Despite the limited taxonomic value of the flavonoid differences with-
in Jepsonia it is of interest that J. parryi and J. malvifolia, which have
adjacent geographical ranges and show the largest degree of crossability
(Ornduff, 1969), should exhibit identical pigment profiles.

Ornduff (1969) stated that, while Jepsonia has no close relatives in

42 MADRONO [Vol. 25

the family, it may be allied with such genera as Bolandra, Boykinia,
Heuchera, Darmera (Peltiphyllum), Suksdorfia and Tellima. Since de-
tailed flavonoid data are available on all of these except Bolandra, some
intergeneric comparisons are possible.

Heuchera micrantha Dougl. var. diversifolia (Ryd.) R. B. & L. and
H. cylindrica Dougl. var. glabella (T. & G.) Wheelock possess exceed-
ingly complex flavonoid mixtures; at least 60 compounds occur in the
former and about 40 are known in the latter (Wilkins and Bohm, 1976a;
and unpubl.). The major compounds are flavonols which exist in a wide
variety of mono-, di-, and triglycosylated forms. The flavonols are kaemp-
ferol, quercetin, and myricetin but small quantities of the O—methylated
flavonols isorhamnetin, larycitrin, and syringetin also occur. Both Heu-
chera species possess 6’’—O—gallyl derivatives of kaempferol and querce-
tin glucosides, accumulate a small amount of the flavone luteolin, and
have a variety of tannins.

Tellima grandiflora (Pursh) Dougl. has a simpler array of compounds
but shares with Heuchera the ability to make gallylated flavonol deriva-
tives. It does not have flavone derivatives. Tellima is the only genus of
Saxifragaceae so far studied that produces 4’—0-glucosides (Collins and
Bohm, 1974). It also has a complex array of tannins (Wilkins and Bohm,
1976b, c).

Damera peltata (Torr.) Voss also has fewer compounds than Heu-
chera although they share some flavonol mono- and diglycosides. Tannins
are also present in Darmera but they appear to be simpler than those in
Heuchera or Tellima (Bohm and Wilkins, 1976).

Preliminary study of Suksdorfia ranunculifolia (Hook.) Engl. (Bohm,
unpubl.) showed a very simple array of flavonol mono-, di-, and triglyco-
sides. Gallylated flavonol glycosides were not observed in Darmera or
the one Suksdorfia species examined.

Finally, studies of Boykinia (Gornall and Bohm, unpubl.) show the
presence of flavonols and flavones in roughly equal amounts. A moder-
ately simple pattern of monoglycosides is present but the complex array
of diglycosides is reminiscent of Heuchera. Gallylated flavonol glycosides
do not appear to be present, but 6—hydroxylation and 3—O—methylation
occur, which characters have been seen in the family so far only in
Chrysos plenium (Bohm et al, 1977, and ref. cited therein).

As in all of the genera of Saxifragaceae whose flavonoid profiles have
been studied to date, Jepsonia has a unique combination of compounds.
However, the flavonols present and their glycosylated derivatives are
clearly related to those found in other members of the family. Jepsonia
most closely resembles Darmera in its flavonol glycosides per se. The
presence of gallylated flavonol glycosides and other simple gallic acid
derivatives in Jepsonia suggests possible relationships with the two tan-
nin-producing genera studied to date: Heuchera and Tellima.

1978] BOHM & ORNDUFF: JEPSONIA 43

ACKNOWLEDGMENTS
This work was supported by grants from the National Research Council of
Canada (to B. A. B.) and the National Science Foundation (to R. O.). Dr. Elijah
Tannen’s continued interest in our work is also greatly appreciated.

LITERATURE CITED

BACIGALUPI, R. 1944. Jepsonia. In Abrams, L. Illustrated Flora of the Pacific States
2:54. Stanford University Press, Stanford, California.

Boum, B. A., F. W. Cortins, and R. Bose. 1977. Chemotaxonomic studies in the
Saxifragaceae s.l. 7. The flavonoids of Chrysosplenium tetrandrum. Phytochem-
istry 16: in press.

, and C. K. WiLxkins. 1976. Chemotaxonomic studies in the Saxifragaceae
s.l. 6. Flavonoids and gallic acid derivatives of Peltzphyllum peltatum. Phyto-
chemistry 15:2012-2013.

Cotiins, F. W. and B. A. Boum. 1974. Chemotaxonomic studies in the Saxifragaceae
s.l. 1. The flavonoids of Tellima grandiflora. Canad. J. Bot. 52:307-312.

, B. A. Boum, and C. K. Wivkins. 1975. Chemotaxonomic studies in the
Saxifragaceae s.l. 2. Flavonol glycoside gallates from Tellima grandiflora. Phyto-
chemistry 14:1099-1102.

Jepson, W. L. 1923-1925. A manual of the flowering plants of California. Associ-
ated Students Store, University of California. Saxifragaceae: pp. 454-475.

————. 1936. Saxifragaceae: In A flora of California 2:116-159. California Book
Depository.

Marry, T. J., K. R. MarkuHam, and M. B. THomas. 1970. The systematic identifi-
cation of flavonoids. Springer-Verlag, New York.

Muwz, P. A. 1959. A California flora. University of California Press, Berkeley and
Los Angeles.

OrnpDuFfr, R. 1961. Heterostyly in Jepsonia (Saxifragaceae). Recent advances in
Botany (IX International Bot. Congress, Montreal, 1959) 1:885-887.

. 1969. Ecology, morphology, and systematics of Jepsonia (Saxifragaceae).
Brittonia 21:286-298.

SMALL, J. K., and P. A. RypBerc. 1905. Saxifragaceae. North American Flora 22:
18-20.

Witkins, C. K., and B. A. Boum. 1976a. Chemotaxonomic studies in the Saxifra-
gaceae s.l. 4. Flavonoids of Heuchera micrantha var. diversifolia. Canad. J. Bot.
54:2133-2140.

—. 1976b. Chemotaxonomic studies in the Saxifragaceae s.]. 3. Ellagitannins

of Tellima grandiflora Phytochemistry 15:211-214.

. 1976c. Chemotaxonomic studies in the Saxifragaceae s.l. 5. Complex

ellagitannins of Tellima grandiflora. Planta Medica 30:72-74.

STEPHANOMERIA MALHEURENSIS (COMPOSITAE),
A NEW SPECIES FROM OREGON

L. D. Gorrites
Department of Genetics, University of California, Davis 95616

Evidence has been amassed that supports the hypothesis that the
diploid annual plant informally referred to as ‘“‘“Malheurensis”’ in a series
of publications (Gottlieb, 1973, 1974, 1977, 1978) has evolved from the
population of Stephanomeria exigua ssp. coronaria (Greene) Gottlieb
with which it is still biotically sympatric. The two taxa are morphologi-
cally extremely similar; however, they can be reliably distinguished by
differences in achene length and weight. Reproductive isolation between
them in nature appears complete and is maintained by three factors: (1)
pollen movement is restricted by differences in breeding system; (2)
there is a crossability factor(s) that reduces seed set from interspecific
cross-pollinations compared to conspecific ones by about 50%; and (3)
several differences in chromosomal structural arrangement exist, includ-
ing a reciprocal translocation, which reduce fertility of F, hybrids to
25% (Gottlieb, 1973). Consequently, it is appropriate to validate the
new taxon as a species. It appears to be one of the very few examples of
the recent natural origin of a diploid plant species.

Stephanomeria malheurensis Gottlieb, sp. nov. Differt a S. exigua
ssp. coronaria acheniis longioribus (plerumque 3.3—3.8 mm longis) et
gravioribus (medie 87.3 + 0.44 mg per 100 achenia) pappi setis longio-
ribus (plerumque 5—6 mm longis) et numerosioribus (9-12 vel —-15).

Type: Oregon, Harney County, 27 mi S of Burns, between Mile Posts
25 and 26 on Highway 205, in parts of sections 11 and 12, T 27S, R 30
E, Willamette Meridian, elevation 1524 m, 2 Jul 1975, Gottlieb 7350,
(Holotype, OSC; isotype, NY).

Distribution: Known only from the type locality, the top of a broad
hill with soil derived from volcanic tuff in an area surrounded by basaltic
soils. The locality has been designated a Scientific Study Site by the
Bureau of Land Management, which has jurisdiction over the land, in
order to preserve the species and to permit additional scientific research
on it. The site, including approximately 160 acres, has been enclosed
within a barbed-wire fence.

Plants annual; taproot with lateral branches often > 30 cm long; the
basal leaf rosette generally < 15 cm in diameter at bolting; herbage
glabrous; rosette leaves generally entire to pinnatifid, oblanceolate to
spatulate; stem single, generally < 60 cm long; branches averaging 23
in number; length of branch between adjacent heads averaging 1.9 cm;
heads on short peduncles 5-15 mm long, often having shorter secondary
peduncles also bearing heads; involucres cylindrical or oblong with a

44

1978] GOTTLEIB: STEPHANOMERIA +

GL

Fic. 1. Representative achenes of Stephanomeria malheurensis (large achenes)
and S. exigua ssp. coronaria (small achenes).

series of equal-sized phyllaries averaging 8.0—9.5 mm long, equivalent in
number to the number of florets, subtended by fewer appressed calycu-
late bractlets; florets 5-6 per head; ligules averaging 8.2—9.4 mm long
and 3.2-3.6 mm wide, dark pink, pink, very light pink, white, or rarely
orange-yellow; styles white or pink; anther apex most often dark pink,
occasionally white; achenes tan or light brown, averaging 3.3—-3.8 mm
long, five-sided with a narrow longitudinal groove on each side, the sur-
face generally rugose-tuberculate; pappus bristles generally 9—-12(—15)
in number, thickened and often connate in groups of 2—4 at their bases,
averaging 5-6 mm long, plumose on their distal 50-60‘ . Chromosome
number, 2 = 8.

In addition to the differences in their achene sizes, S. malheurensis can
be distinguished from S. exigua ssp. coronaria in the uniform garden in
a number of quantitative characters including cotyledon length, number
of branches per stem, length of internodes along the branches between
adjacent heads, number of florets per head, and ligule length/width ratio
(Gottlieb, 1973, 1977). They also differ in breeding system: S. exigua
ssp. coronaria has a sporophytic self-incompatibility system that prevents
self-pollen from germinating on stigmas of the same plant, making it an
obligate outcrosser. Stephanomeria malheurensis is predominantly self-
pollinating. Comparative information regarding variation in the electro-
phoretic mobilities of a sample of their enzymes, cytogenetic behavior

46 MADRONO [Vol. 25

and fertility of interspecific hybrids, growth rates under different experi-
mental conditions, phenotypic variability, phenotypic plasticity, and re-
quirements for seed germination have been described in the previously
cited publications.

ACKNOWLEDGMENTS
I thank Prof. Kenton L. Chambers for translating the species description into
Latin and for reviewing the manuscript.

LITERATURE CITED

Gottiies, L. D. 1973. Genetic differentiation, sympatric speciation and the origin
of a diploid species of Stephanomeria. Amer. J. Bot. 60:545—553.

. 1974. Genetic stability in a peripheral isolate of Stephanomeria exigua

ssp. coronaria that fluctuates in population size. Genetics 76:551—556.

—. 1977, Phenotypic variation in Stephanomeria exigua ssp. coronaria (Com-
positae) and its recent derivative species “Malheurensis.” Amer. J. Bot. 64:
873-880.

————. 1978. The origin of phenotype in a recently evolved species. In Otto T.
Solbrig, (ed.). Plant Population Biology, Columbia Univ. Press.

TAXONOMY OF AXINIPHYLLUM
(ASTERACEAE-HELIANTHEAE)

B. L. TURNER
The University of Texas, Austin 78712

Axiniphyllum is a small herbaceous genus of four poorly known species
of south-central Mexico (states of Guerrero and Oaxaca). It was erected
by Bentham in 1872 to accommodate two rayless species, A. corymbosum
and A. tomentosum, the former serving as the type for the genus. The
latter species, however, had been described earlier under the name Polym-
nia scabrum; hence Axiniphyllum scabrum is the correct binomial for
this taxon, as noted by Blake (1930).

I became interested in the genus through my efforts to position what
appeared to be an unidentified Rumfordia. While it seemed close to
Axiniphyllum, it possessed well-developed rays. Rumfordia, with well-
defined ray florets, appeared more remote, except for the relatively re-
cently proposed R. pinnatisecta (Wilson, 1958). The latter, however,
seemed exceptional in Rumfordia, and careful comparisons of the 10 or
more species of the latter genus and the two original species proposed by
Bentham for Axiniphyllum have convinced me that Wilson erred in
placing R. pinnatisecta in Rumfordia. Rumfordia pinnatisecta is much
closer to Axiniphyllum and I have therefore transferred it to what I
believe is its correct phyletic position, alongside the newly proposed A.
sagittalobum.

1978] TURNER: AXINIPHYLLUM 47

As to the generic relationship of Axiniphyllum, Bentham (1872) noted
that it was “in many respects allied to Zaluzania and Sabazia.” The latter
genus is readily distinguished by its white rays, more delicate habit,
non-clasping leaves, style branches, etc.; in short, the similarity is pre-
sumably superficial. The same may be said for the alternate-leaved
Zaluzania (Olsen, 1977); Bentham, presumably, was unduly influenced
by the epappose achenes of both genera.

I would relate Axiniphyllum to Rumfordia as would Sanders (1977),
primarily through the two radiate species proposed here and the herba-
ceous R. alcortae Rzedowski. The latter, in most of its major characters
(involucre, achenes, and style branches), is like Rumfordia but ap-
proaches Axiniphyllum in its habit, leaf shape and inflorescence. The
following key serves to distinguish the two genera:

1. Style branches narrowly linear, prominently pubescent on the abaxial
surfaces; apical appendage elongate, conical, as long as or ionger than
the style branch width; ray achenes 4-sided, not conspicuously radially
flattened; outermost (5-6) involucral bracts densely stipitate-glandu-
lar, similar in shape and texture to those which they subtend

Se Axiniphyllum.

. Style branches linear, Babrodeo or merely pamiilece on the abaxial sur-
face; apical pp dase short, half or less as long as the style branch
width; ray achenes radially flattened, usually prominently so; outer-
most bracts 5, sharply relexed, not densely stipitate-glandular, marked-
ly different in texture and shape from those within . . Rumfordia.

—

AXINIPHYLLUM Benth.

Perennial herbs up to 1 m tall. Leaves opposite, connate, the blades
coarsely pubescent and variously lobed. Heads 1- several in loose corym-
bose panicles. Involucral bracts imbricate in 2—3 series, the outer coarsely
pubescent or glandular. Receptable nearly flat to short-conical, with well-
developed, acute, scarious, 2-4 nerved paleae. Ray florets present or
absent; when present, pistillate and fertile. Disk florets fertile with
elongate, cylindric, 5-lobed limbs. Anther sacs obtuse at the base, the
apical appendages narrowly ovate to ovate-cuspidate. Style branches
linear-subulate, markedly short, pubescent on the exterior surfaces, the
appendages elongate-conical, pubescent. Achenes black, glabrous, epap-
pose. Chromosome number unknown.

Type species: Axiniphvllum corymbosum Benth.

Key to Species
1. Heads without ray florets (2)
2. Heads small, 8-10 mm ey outermost involucral bracts 5-7 mm

long... - . . L.A. corymbosum
2. Heads large, 52 20n mm eh fonteemnoet involucral bracts 15-25 mm

long 2.A. scabrum

48 MADRONO [Vol. 25

1. Heads with ray florets (3)
3. Terminal lobe of leaf blade closely serrate, sagittate in outline;
flowering heads 2—5, on ultimate peduncles 4—8 cm long
ei at Det as, Wiser Sin hag oe 3.A. sagittalobum
3. Terminal lobe of leaf blade crenulate, variously shaped; flowering
heads 6-30, on ultimate peduncles 2—3 cm long
4. A. pinnatisectum

—

. AXINIPHYLLUM CORYMBOSUM Benth., Hook Icon. Pl. 12:17. 1872.
Type: MEXICO. Oaxaca: “woods in the province of Oaxaca, at an
elevation of 7500 feet”, Sep 1840, H. Galeotti 2089 (Holotype: K!).

Sparsely branched, perennial herbs up to 70 cm tall. Leaves sagittate,
irregularly dentate or lobed to nearly entire, the petioles winged, auricu-
late, connate. Heads 8-10 mm high, arranged in a very loose corymbose
panicle, the ultimate peduncles 2—5 cm long. Involucres 6—8 mm long,
the bracts imbricate in 2—3 series, scabrous-pubescent to variously stipi-
tate-glandular, often intermixed. Ray florets absent. Disk florets ca 40,
fertile; corolla sparsely pubescent, ca 5 mm long; achenes epappose,
glabrous, black, 4-sided, 2-3 mm long, ca 1 mm wide, the upper, broadest,
portion quadroid in cross section.

DISTRIBUTION. Montane forests of south-central Mexico (Guerrero
and Oaxaca) from 1800-2500 m, reportedly growing in the shade of

Fic. 1. Distribution of Axiniphyllum species: A. corymbosum (area surrounding
C; A. scabrum (area surrounding S); A. sagittalobum (asterisk) ; A. pinnatisectum
(dots).

1978] TURNER: AXINIPHYLLUM 49

oaks and pines as a ‘“‘subherbacea anual geofita” (Kruse 2666), although
the specimens concerned appear to have arisen from lignescent, perennial
rootstocks. Sept.-Nov. (Fig. 1).

Additional specimens examined: MEXICO. GUERRERO: Mazatlan,
cerro Alquitran, 7 Nov 1969, Kruse 2666 (FM, US); without locality,
Sesse & Mocino “1490” (fragment) also numbered “2825” (FM).

Bentham’s description of the species was accompanied by a good
drawing showing a somewhat more overly-pubescent corolla and a more
congested inflorescence (because of the immature nature of the speci-
men) than is typical of the taxon. Mature achenes, as he noted, were
lacking. However, in most other details, the few collections available to
me match the holotype and drawings quite nicely, except for its reported
annual habit, which is presumably an observational error to be attributed
to the collector since this is so noted by appropriate symbol on his col-
lection label. Galeotti also describes the corollas as “yellow & rosy” but
recent collections by Kruse (2666, FM, US) describe the flowers as
white, although in the dried state they appear to be somewhat on the
yellow side.

2. AXINIPHYLLUM SCABRUM (Zucc.) Blake, Contrib. U.S. Nat. Herb.
26:248. 1930. Polymnia scabra Zucc., Abh. Akad. Wiss. Munchen
1:313. 1832. Type: Grown in the Botanical Garden at Munich from
seeds collected in Mexico by Karwinsky s.n., 1829 (Holotype, M;
isotypes, BM!, P; phototype US!).

Polymnia aspera Mart. ex DC., Prodr. 5:515. 1836 (Holotype, P).

According to Blake, this name is based in part on portions of the above

type, the specific epithet itself being a “slip of memory” on Martius’

part, Zuccarini’s earlier name having been intended.

Axiniphyllum tomentosum Benth., Hook. Icon. Pl. 12:17. 1872. Type:

MEXICO: Without locality, 1846, Bates s.n. (Holotype, K; photo-

type US!).

Robust perennial herb up to 1 m tall. Leaves broadly sagittate, those
at mid-stem as broad as or broader than long, scaberrimous above,
densely canescent or tomentose beneath, irregularly dentate or lobed
to nearly entire, the petioles winged, auriculate, connate. Heads 15—20
mm high, arranged in broad open corymbose panicles, the ultimate pe-
duncles 1-7 cm long. Involucral bracts 2-3 seriate, densely stipitate-
glandular, the 5 or 6 outermost bracts (15-25 mm long) equalling or
exceeding the head proper, spreading or reflexed at maturity in the man-
ner of Rumfordia. Ray florets 2bsent. Disk florets numerous (50-80),
fertile; corolla pubescent, 6-7 mm long, the limb abrupt, about twice
as long as the tube; achenes epappose, 3-4 mm long, 1-2 mm wide,
black, glabrous, 4-sided in cross section near the apex.

DISTRIBUTION. Known only from montane regions in the state of
Oaxaca, Mexico, where it reportedly occurs in “weedy fields” and “shrub-
by slopes” in oak forest zones at about 2000 m. Aug-Nov. (Fig. 1).

50 MADRONO [Vol. 25

Additional selected specimens examined: Oaxaca: 11.5 mi N of
Telixtlahuaca, 6 Nov 1966, Anderson & Laskowski 4138 (FM, GH);
5 mi NE of Mexican Highway 190, near Oaxaca, 26 Aug 1965, Breed-
love 12202 (US); hills near Oaxaca, Aug 1894, Pringle 4826 (FM);
Sierra de San Felipe, 8 Sep 1894, Smith 284 (FM, US).

This is a very distinct taxon, easily recognized by its large heads with
prominent, often reflexed, outer involucral bracts. It is apparently rela-
tively common in spite of its restricted distribution, being known by at
least 10 separate collections from among the herbaria from which I
borrowed.

3. Axiniphyllum sagittalobum Turner, sp. nov.
Type: MEXICO. Guerrero: Districta Mina, Toro Muerte, 2800 m,
30 Oct 1939, G. B. Hinton et al. 14761 (Holotype, LL; isotypes,
MICH, NY, US).

Herbae perennae (?) ad 1 m altae caulibus saepe rigide erectibus un-
usuquisque 2—5 capitata. Folia irregulariter lyrata vel repanda parte
terminali sagittata irregulariter serrata utrinque pubescentia pilis brevi-
bus crispatis. Pedunculi 5—8 cm longi stipitato-glandulares. Flores radi-
ati 8, pistillati fructiferi ligulis “luteolis” ca 8 cm longis, 5-6 mm latis
apice trilobatis. Flores disci ca 40 hermaphroditi fructiferi. Styli rami
complanati lineario-lanceolati subtus pubescentes lineis stigmaticis infine
bene evolutis sed apicem versus gradatim cum appendice terminali con-
fluentibus. Achaenia uniformia nigra glabra pappis nullis.

Perennial (?) herb up to 1 m high, the stems mostly stiffly erect,
glabrate and unbranched. Leaves opposite, connate, irregularly lyrate to
repand, the terminal portion sagittate, irregularly serrate, pubescent on
both sides with short, crisp, hairs, especially along the veins. Heads 2—5
to a stem, borne upon elongate stipitate-glandular peduncles, 5-8 cm
long; involucre 2—3 seriate, the outermost whorls linear-lanceolate, dense-
ly stipitate-glandular, variously reflexed with age. Receptacle chaffy, pu-
berulent, short-conical, knobby (with age). Ray florets 8, pistillate,
fertile, “pale yellow”, the ligules ca 8 mm long, 5-6 mm wide, 3-lobed
at the apex; tube ca 2 mm long, densely pubescent with both glandular
and nonglandular, uniseriate trichomes. Disk florets ca 40, perfect, fer-
tile; corolla 5-6 mm long; tube ca 1.5 mm long, the limb abruptly am-
pliate, 5-lobed, sparsely pubescent. Style branches flat, linear lanceolate,
pubescent beneath, the stigmatic lines well developed below but gradu-
ally merging into the acuminate appendage. Ray and disk achenes simi-
lar, falcate (the outermost) to clavate, black, glabrous, epappose.

DISTRIBUTION: Known only from the type collection.

The species is undoubtedly closely related to Axiniphyllum pinnati- —
sectum but can be readily distinguished by its sagittate, markedly ser- |

1978]

TURNER: AXINIPHYLLUM

PS)

ieay,

Pic. 2.

Habit sketch of Axiniphyllum pinnatisectum (X 1%)

. Hinton 9756 (MICH).

52 MADRONO [Vol. 25

rate, terminal leaf lobes, somewhat larger heads and more conspicuous
outer involucral bracts. However, both of these taxa are too poorly repre-
sented in herbaria to speculate upon the constancy of these characters
and additional collections showing populational intergradation might
mark these as no more than varietally distinct. A wide range of pinnati-
fied leaf forms is found among the isotypes of this taxon but none ap-
proaches that found in A. sagittalobum.

4. Axiniphyllum pinnatisectum (P. G. Wilson) Turner, comb. nov.
(Fig. 2). Rumfordia pinnatisecta P. G. Wilson, Kew Bull. 1958:164.
1958. Type: MEXICO. Guerrero: Mina Dist., Aguazarca-File, pine
forest, 30 Nov 1937, Hinton et al. 11289 (Holotype, K!; isotopes
NY! US!)

Perennial herb up to 1.5 min height. Similar to Axiniphyllum sagitta-
lobum but readily distinguished by the characters listed in the key to
species.

DISTRIBUTION: Known only from the state of Guerrero where it re-
portedly occurs in pine forests at ca 2300 m elevation. Oct-Nov. (Fig. 1).

Additional specimens examined: GUERRERO: Dist. Mina, Armenia,
pine forests, 2340 m, 23 Oct 1936, Hinton et al. 9756 (GH, NY, TEX,
US).

Wilson presumably relegated this species to the genus Rumfordia
largely because of its radiate heads. In most other characters, however,
it is much closer to Axiniphyllum corvmbosum and I find no hesitation
in making the necessary transfer.

ACKNOWLEDGMENTS
This study is based upon material from the following herbaria: British Museum
(BM), Field Museum (FM), Gray (GH), Kew Gardens (K), Lundell (LL), Univ.
of Michigan (MICH), New York Botanical Garden (NY), Univ. of Texas (TEX)
and the United States National Museum (US). My thanks to the directors for the
loan of material. Dr. M. C. Johnston provided the Latin description, for which I
am grateful. Supported in part by N.S.F. Grant 1013950.

LITERATURE CITED
OLSEN, 7. 1977. Systematic study of Zaluzania (Asteraceae). Ph.D. Thesis. The
University of Texas, Austin, Texas.
SANDERS, R. 1977. Taxonomic study of Rumfordia (Asteraceae). Systematic Botany
2:(In Press).

NOTES AND NEWS

REDUCTION OF GUNNERA KILIPIANA TO SYNONYMY WITH G. MEXICANA—The tax-
onomy of many plants has been obscured by poorly collected and poorly preserved
material. The literature is replete with synonyms reflecting the tendency to multiply
taxa needlessly when treating fragmentary or incomplete data. Those taxa inhabit-
ing remote areas or possessing large, cumbersome organs are especially prone to
such treatment. Gunnera L., a gigantic herb of the tropical montane cloud forest,
usually placed in the Haloragaceae, has suffered much in this manner.

In the summer of 1912, J. A. Purpus collected Gunnera (Purpus 8568) in the
Sierra Chiconquiaco above Misantla in Veracruz, Mexico. T. S. Brandegee named
it a new species, G. mexicana Bdg., in 1922 (Brandegee, Univ. Cal. Publ. Bot.
12:181-188. 1922). C. L. Lundell studied material of Gunnera which Eizi Matuda
(no. 2763) had collected from Volcan Tacana in Chiapas in 1939 and found notice-
able differences between it and Brandegee’s published description. In 1940 he named
the Matuda specimen G. kilipiana Lundell (Lundell, Phytologia 1:449-453. 1940).

After examining the collections of D. S. Barrington from the type locality of G.
mexicana above Misantla where he collected in December, 1971, I was struck by
the great similarity between his material and Lundell’s type of G. kilipiana from
Chiapas. I have since examined the 2 sheets of G. mexicana from the Gray Her-
barium (both labeled “isotype’”’) and the holotype from the University of California,
Berkeley, and have examined living material of both in the field. I am convinced
that Brandegee erred in his description and misled Lundell, who subsequently
named a new species needlessly. Gunnera kilipiana is synonymous with G. mexicana.

The holotype of G. mexicana (UCB 206237) consists of a habit photograph, an
inflorescence and a piece of a leaf folded like a fan into a rough triangle with a
short stub of petiole at the narrowest end. The Gray Herbarium isotypes consist of
similarly folded, but smaller pieces. Careful examination reveals that, by carefully
matching cut edges, impressions of veins and decayed spots, all three pieces were
once part of the same leaf (minus its petiole). Brandegee based his description on
only the middle section of this single leaf.

Brandegee’s error was never discovered. Inasmuch as the material Lundell had
from Chiapas differed from Brandegee’s description, he distinguished it from G.
mexicana, noting that this, the only other Mexican species, was known to him from
the brief original description only. He explained that his new species “apparently
differs from G. kilipiana amply in its leaf form being attenuate at the base rather
than deeply cordate” (my emphasis). The photograph was not mentioned.

Brandegee’s published description states: “foliis . . . latitudine valde variabilibus,
prope apicem usque ad 14 cm latis, ca. 32 cm longis in petiolum brevissimum gra-
datim angustatis ... ,” and “This extralimital species differs from the generic
description in the shape of the leaves. The leaves in circumscription are rounded
at the top and attenuate into a very short petiole’ (my emphasis).

This description could not apply to the Gunnera which Purpus collected if Bran-
degee had described whole leaves. Most Gunnera have cordate leaves; a few have
peltate ones. Purpus easily recognized Gumnnera in the field; this would have been
unlikely if it had differed greatly from known species. From his journal of that trip
he states: “Otra vez llegé el aguacero, tremendo; subiendo, al pasar encontramos
totonacas que se protegian del agua con las hojas de la Gunnera y de un Caladium.
Con tal motivo, tuve que acostarme para que mi ropa se secara . . .” (in Sousa
Sanchez, Univ. Cal. Publ. Bot. 51:1-36. 1969).

If he had used it as an umbrella, as he says, it must like Caladium have had a
long petiole. I have used it myself in this fashion, and without the petiole, it is
useless. Purpus’s photograph, attached to both the holotype and an isotype and
published by Brandegee, shows perfectly round, deeply cordate leaves, raised and
presumably supported by what must be long petioles.

The material from the type locality bears obovate or reniform leaves, moder-

53

54 MADRONO [Vol. 25

ately lobed, with deeply cordate sinuses and long (exceeding 1 m) petioles. It re-
sembles G. insignis (Oerst.) A. DC. from Costa Rica and Panama somewhat, and
G. kilipiana very closely.

Further comparison of both types and material from the type locality of G.
mexicana as well as additional material from central Chiapas and southwestern
Guatemala shows great similarity in leaf surface features and inflorescence charac-
ters. Pubescence on the leaf surface is very similar to the above and much denser
than that of the two other Central American species, G. insignis and G. talamancana
Weber & Mora. Inflorescence characters, particularly in the thickness of the
branches and the position, size and shape of their subtending bracts are also quite
similar and consistent with synonymy.

I can find no real differences between material from the type locality for G.
mexicana and Lundell’s type. Lundell’s description is a very good description of the
Veracruz material. Since there can no longer be any conflict between Brandegee’s
and Lundell’s short (58 and 28 word) descriptions, I can see no other alternative
than to call them one species.

Even though Brandegee’s description is based on fragmentary material and thus
misleads, it is the earlier, and has priority, so G. kilipiana must be reduced to
synonymy.

Material Studied: MEXICO: Veracruz: Sierra Chiconquiaco above Misantla,
26 Dec 1971, Barrington 416a, 416b 417, 439 (GH); 18 Sep 1973, Palkovic 767,
708, 109, 770, 771, 172, 773, 174, 115, 770, 177, (78, 179; 780 €GH) > Jul 1912, Pure
pus 8568 (UC holotype of G. mexicana, GH isotopes). CuHtapas: Volcan Tacana, 23
Mar 1939, Matuda 2763 (GH, type of G. kilipiana).

This study represents a by-product of a dissertation completed at Harvard Uni-
versity under the supervision of Rolla M. Tryon with the support of NSF grants
GB 27911 and GB 39866. I would also like to thank Drs. Richard M. Straw and
Kenneth Wilson for their helpful reviews and comments.—LAWRENCE A. PALKOVIC,
Department of Biology, California State University, Los Angeles, 90032.

FERNS OF THE NEw YorkK MOUNTAINS, CALIFORNIA, WITH BIOGEOGRAPHIC COM-
MENTS.—Several chains of mountain ranges in the eastern Mojavo Desert are of
particular biogeographic interest. The southernmost chain extends from the Granite
Mountains through the Providence, Mid Hills, and New York Mountains to the
northeast. North of the eastern part of this chain is another chain, comprising the
Ivanpah, Mescal, and Clark Mountains. Still farther north is the Kingston Range,
and finally, in adjacent Nevada, the very high Spring Mountains form the north
end of this assemblage.

These mountains share several important features. They are high enough at their
crests nearly or quite to emerge from the desert zone of the region. They regularly
receive winter snow, and importantly, they generally intercept significant precipita-
tion from the frequent late summer invasions of moist air from the Gulf of Mexico.
This invests them with two rainy seasons, much like climatic regimens to the east.
These ranges exhibit great tectonic complexity, and the array of exposed geological
formations provides both chemical and physical edaphic diversity. Collectively,
these ranges lie equidistant from ranges with comparable elevations to the east in
Arizcna (Cerbat and Hualapai Mountains) and to the southwest in California
(Transverse Ranges).

These features favor a token intrusion of numerous biotic elements not otherwise
a part of the fauna and flora of California. Particularly, plants and animals with
ranges principally in the Arizona Uplands (Shreve, Publ. Carnegie Inst. Wash.
591:42-43, map 1. 1951) of the Sonoran Desert may reach California here. Also,
some organisms with southern Rocky Mountain affinities and ranges across boreal
Arizona form a part of this southeast California assemblage. This has been noted

1978 | NOTES AND NEWS 55

briefly, without emphasis, for vertebrates by Johnson et al. (Univ. Cal. Publ. Zool.
48:248. 1948), for higher plants by Parish (Ecology 11:498. 1930) and most
recently by Henrickson and Prigge (Madrono 23:164-168. 1975). There are a
number of examples of similar distribution patterns known for invertebrates,
chiefly insects, especially in the better known orders such as Lepidoptera (Emmel
and Emmel, Los Angeles County Mus. Sci. Ser. 26:22-23, 46, 58, 60, 78, 84, 89. 1973;
Ferguson, Moths of America North of Mexico 20.2A:133. 1971; MacNeill, unpubl.).

Recent collections of ferns from a section of the New York Mountains (Califor-
nia, San Bernardino Co.) reflect well the biogeographic patterns suggested by other
segments of the flora and fauna. These collections have provided new records for
the state and several additional records for this mountain range. Our observations
further reaffirm that substrate strongly influences fern micro-distributions. Follow-
ing is an annotated list of the species of ferns now known from the New York
Mountains. Eight of the nine species of ferns recorded for the New York Moun-
tains are rare or very uncommon in California as a whole and can be regarded as
intrusive elements into California from the east (even though one, Notholaena
jonesii, does extend westward to coastal counties). All eight occur in Arizona and
generally southward or eastward, often much more commonly than in California.

Unless otherwise stated, collections are from the New York Mountains. Geo-
logical determinations have been provided through the kindess of B. C. Burchfiel,
Rice University, and Pierina Nicholson, The Oakland Museum (here abbreviated
OM).

Polypodium hesperitum Maxon. 550 m NNE of New York Mt. Peak, 2130 m,
MacNeill and Brophy 097510D11 (OM, UC), in shaded granite fissures and crevices;
2n = 74 II (voucher UC). Previously known in California from only a few collec-
tions in the Transverse Ranges. Records for P. hesperium from Placer Co. in the
Sierra Nevada (Howell and Long, Four Seasons 3:8. 1970) and also northern
California (see map in Lang, Madrofio 20:58. 1969) require re-examination since
Lang (Madronfio 20:53-60. 1969; Madronfio 21:235—254. 1971) and Lloyd (Fremontia
3:18-21. 1975) seem to imply that the diploid cytotype (P. amorphum Suksdorf)
and the tetraploid cytotype (P. hesperium) are allopatric or nearly so south of
British Columbia. Polypodium amorphum ranges south along the Cascade-Sierran
axis to the central Sierra Nevada, where it is rare (Howell and Long, loc. cit.;
reported as P. montense Lang). Polypodium hesperium evidently occurs along the
intermountain face of the northern Cascades and extends south to Mexico along
the Rocky Mountain system. It ranges west through Arizona and, unaccountably
until now, is known from two stations in the Transverse Ranges of California. Our
collection from the New York Mountains is tetraploid and lacks paraphyses and is
thus not P. amorphum. The New York Mountain locality bridges the distributional
gap between Arizona collections and those from western San Bernardino and River-
side Counties.

Notholaena jonesii Maxon. Keystone Basin, 1750 m, MacNeill and Smith s.n.
(OM, UC), crevices of blue-gray limestone cliffs and boulders; 2” = 54 II (voucher
UC). First collection from the New York Mountains. Previously known from scat-
tered localities in southern California (including Providence and Clark Mts.), Utah,
and Arizona. In Keystone Basin, this species was found associated only with the
bluish-gray limestones of the Pennsylvanian Bird Spring Formation, most abun-
dantly on a south-facing slope. On Clark Mountain we found this fern only upon
a similar-appearing blue-gray limestone.

Notholaena limitanea Maxon var. limitanea. Keystone Basin, 1800 m, MacNeill
097512A2 (OM), s.n. (UC), 20 plants found in crevices of north-facing bluish-gray
limestone of the Bird Spring Formation. First collections from California; previously
recorded from southern Utah, Arizona, southern New Mexico, west Texas, Chihua-
hua, and Sonora, with an additional variety from southeastern Arizona and adja-
cent New Mexico to Hidalgo (Tryon, Contr. Gray Herb. 179:86. 1956).

56 MADRONO [Vol. 25

Cheilanthes feet T. Moore. Keystone Basin, MacNeill s.n. (OM, UC), widespread
throughout Keystone Basin from 1700 m to 2000 m in crevices of often north-
facing limestone cliffs and boulders. Associated with both the bluish-gray and the
white limestones of the Pennsylvanian Bird Spring Formation, but seemingly absent
from a Triassic metamorphosed sandy limestone formation. A single plant was
found on granite. Known from Texas to Iowa and west to Arizona, Nevada, British
Columbia, Washington, and transmontane California in the Providence, New York,
Clark, Panamint, and Inyo-White Mountains.

Cheilanthes wootonit Maxon. Keystone Basin, 1900 m, MacNeill s.n. (OM), bases
of granitic outcrops under oaks. Found only among granitic rocks above 1825 m in
Keystone Basin. Reported by Munz (op. cit.) from New York, Panamint, Inyo-
White, and Providence Mountains in California, but we have seen specimens only
from the two first-named ranges: New York Mts., Fourth of July Canyon, Alex-
and and Kellogg 1411, 1412 (UC); Panamint Mts., Munz 12571 (UC). Lloyd and
Mitchell (A Flora of the White Mountains, California and Nevada. 1973) recorded
it only as “‘to be expected” in the White Mountains. Otherwise known from Baja
California, Sonora, Chihuahua, Arizona to Texas and north to Colorado and
Oklahoma.

Woodsia plummerae Lemmon. Known from California only from a single ravine
in Keystone Basin (Smith, Madrono 22:378. 1974) ; additional collections have now
been made from the same locality (Smith 673, UC; MacNeill s.n., UC). The species
is found at 1900 m in soil at the base of north-facing granitic cliffs and boulders,
often under the canopy of Quercus chrysolepis. Many of these plants show the
characteristic forking or cresting og the blade apex mentioned for the species by
Brown (Beih. Nova Hedwigia 6:106. 1964).

Woodsia oregana D. C. Eaton. Keystone Basin, 1900 m, MacNeill sn. (OM),
097510D9 (UC), in soil at the base of granitic cliffs and boulders, generally well-
shaded. Widespread throughout the higher elevations on intruded granite pluton
that forms the crest of the range. In Keystone Canyon it grows with W. plummerae
and in many other places on north-facing granitic slopes. The species ranges from
British Columbia to Vermont in the north, south to New Mexico, Arizona and
southern California. Nearly all records in California are transmontane. Howell and
Long (op. cit., p. 9) cited only a single certain collection of this species from the
southern Sierra Nevada.

Pellaea truncata Goodding (P. longimucronata of California and Arizona refer-
ences; an illegitimate name, see Cronquist et al., Intermountain Flora 1:202, 1972).
Keystone Basin, 1700 m, MacNeill sn. (OM), mainly among granitic boulders and
fissures in granite cliffs, but also frequent in a Jurassic formation of sheared vol-
canic and metamorphosed sedimentary rocks, and remarkably, the dominant fern
(the only species found) throughout a formation of metamorphosed sandy lime-
stone that may be Triassic. This latter formation seems not to support any of the
several “limestone” ferns of the region. In the Keystone Basin area it ranges, on
appropriate substrates, from 1600 m to 2150 m. It is rare on the Bird Spring lime-
stones. Known with certainty in California only from the New York Mountains
(also Ferris and Bacigalupi 8076, UC, Alexander and Kellogg 1323a, UC) and Provi-
dence Mountains (Pray, Amer. Fern J. 57:52-58. 1967). Its range beyond this part
of California extends from Nevada to Colorado and Texas, Sonora, Arizona, and
Baja California (Tryon, Ann. Missouri Bot. Gard. 44:155. 1957).

Pellaea mucronata D. C. Eaton. Fourth of July Canyon, Alexander and Kellogg
1323, pt. (UC). The locality cited is to the southwest of the region we sampled
and is not far from a station in the Mid Hills (Smith 682, UC). This fern is also
known from the Providence Mountains. It is very closely related to the preceding
and, according to Pray (op. cit.), the two species hybridize in the eastern Mojave.
Pellaea mucronata is one of several species that are much more prominent nearer
the western margins of the desert regions. Indeed, this species is largely cismontane.

1978] NOTES AND NEWS 57

but does reach the desert, extending eastward barely into Nevada; it is not known
from Arizona. Pellaea mucronata is here considered to be one of several ‘“Califor-
nian” elements (discussed below) that extend eastward to meet elements from

Several additional ferns may ultimately be found in the New York Mountains.
These fall into two categories: 7) those with distributions primarily to the south
and east (Sonoran element) ; and 2) those with distributions in the eastern Mojave
and Colorado Deserts and westward (Californian element). There are three such
ferns in the first category:

Notholaena cochisensis Goodding occurs in limestone in the Providence Moun-
tains and on a bluish limestone (much like that in Keystone Canyon) on Clark
Mountain. It is to be expected on limestone in the New York Mountains but at
elevations somewhat lower than the floor of Keystone Basin. Hevly (J. Ariz. Acad.
Sci. 3:205—208. 1965) recognized this as a species separate from WN,. stnuata (Lag. ex
Swartz) Kaulf., a distinction we support on morphological, geographical, and eco-
logical grounds.

Asplenium resiliens Kunze has been reported recently in the Spring Mountains
of Nevada (Fisher, Madronfo 23:72. 1975), where it occurs on Navajo sandstone.
The species may occur on shaded limestone or limey sandstone cliffs in other
ranges of the eastern Mojave Desert at elevations somewhat below those of Key-
stone Basin.

Cheilanthes fendlert Hook. was reported from southern California by Cronquist
et al. (Intermountain Flora 1:205. 1972), but Cronquist indicates (in litt.) that the
inclusion of California was based on misidentified collections. Still, it is another
Sonoran element that may occur in the eastern Mojave.

There are four ferns of the second category that may eventually be found in the
New York Mountains. Notholaena californica D. C. Eaton and Cheilanthes vtiscida
Davenport have not yet been recorded for the eastern Mojave ranges; they might
be expected in the Granite Mountains. Notholaena parryi D. C. Eaton, a wide-
spread fern of moderate elevations in the deserts, can be expected anywhere in the
eastern Mojave ranges below 1550 m. We have seen specimens from the Providence
Mountains (Bonanza Mine, Opler s.n., OM). Cheilanthes covillet Maxon has been
collected in the Providence Mountains (Wolf 10688, UC) and may be present in the
New York Mountains. This species is very closely related to C. wootoni. We ex-
pect all four species to be most prominent near the westernmost parts of the eastern
Mojave ranges—C. Don MacNeErL1t, The Oakland Museum, Oakland, CA 94607,
WILLIAM Bropnuy, Chabot College, Hayward, CA 94545, and ALAN R. SmirH, Uni-
versity Herbarium, Department of Botany, University of California, Berkeley, CA
94720.

DipLoip CLAYTONIA PERFOLIATA FROM SOUTHERN MeExico.—Prior studies of Clay-
tonia perfoliata Willd. [Montia perfoliata (Willd.) T. Hewell] (Miller, Syst. Bot.
1:20-34. 1976; Fellows, Madrofo 23:296-297. 1976; Swanson, Ph.D. Dissertation,
Univ. California, Berkeley, 1964) revealed two morphologically different diploids
(2n = 12), which were called “Channel Islands” or “Coastal” (referable to C.
perfoliata ssp. perfoliata) and “Montane” [= C. rubra (T. Howell) Tidestrom].
These diploid species are easily distinguishable morphologically. The former is char-
acterized by petals 3 to 4 mm long, linear juvenile basal leaves, deltoid mature basal
leaves (with mucronate tips), green herbage, and a perfoliate to only slightly cleft
cauline leaf disc. Claytonia rubra has petals similar in length to those of C. perfoli-
ata ssp. perfoliata; and deltoid mature basal leaves. However, the juvenile leaves
are never linear but instead are rhombic, and the cauline leaves are free or are
united on only one side of the scape. As the name implies, C. rubra is characterized
by livid beet-red foliage coloration, particularly on the abaxial leaf surfaces, al-
though green-leaved morphs may be encountered in some populations.

Un

8 MADRONO [Vol. 25

Mexican populations at 3200 m on the slopes of Popocatepet] and 3000 m on
Cerro Ajusco were examined cytologically (2m =12; México, Distrito Federal,
Slopes of Cerro Ajusco, 2 km E of Estacion La Cima on Hwy. 95, Miller 568 ; Estado
México: Slopes of Popocatepetl, 12.5 km E of Hwy. 115 junction on the road from
Amecameca to Tlamacas, Miller 570; Slopes of Popocatepetl, 5.5 km W of Paso de
Cortez on the road from Tlamacas to Amecameca, Miller 571; Municipio Ameca-
meca, Rodriguez 1460). These populations are morphologically indistinguishable
from diploid C. perfoliata ssp. perfoliata found in coastal California and on the
Channel Islands. In contrast to C. rubra, which is common in drier northern mon-
tane and transmontane coniferous woodlands, diploid C. perfoliata ssp. perfoliata
is more southern in distribution, ranging from coastal and cismontane California,
through the Sonoran Desert, to high elevation coniferous forests of Mexico and
Guatemala. Herbarium specimens examined from Durango, Queretaro, Hidalgo, Ja-
lisco, Distrito Federal, Puebla, México, Morelos, and Cuesta El Caracol in Guate-
mala, indicate relative homogeneity of Mexican and Guatemalan populations, not
only in their striking resemblance to the known diploid populations cited above but
also in their elevational distribution and habitat preference.

Voucher specimens and permanent microslides for the chromosome counts re-
ported here are deposited in OSC. Duplicate cytovouchers are deposited in CAS
and ENCB. I am grateful to Dr. J. Rzedowski and Miss L. S. Rodriguez of the
Escuela Nacional de Ciencias Biologicas for their help with field work and to the
National Science Foundation for financial assistance (Doctoral Dissertation Re-
search Grant DEB 76-06048).—JoHN M. Mirirer, Department of Botany and Plant
Pathology, Oregon State University, Corvallis 97331.

NOMENCLATURAL CHANGES IN SPILANTHES, LYCOPERSICON, AND OPUNTIA FOR THE
GaLApacos IsLanps.—Research on the endemic flora of the archipelago reveals that
the following nomenclatural changes must be made:

(1). Spilanthes diffusa Hook. f. (Trans. Linn. Soc. London 20:214. 1847) is a
later homonym of S. diffusa Poepp. & Endl. (Nov. Gen. Sp. Pl. 3:50. 1843). No
other specific epithet being available for the former taxon, the following is pro-
posed: Spilanthes darwinii D. M. Porter, nomen novum (Holotype: Darwin, end
of Sept. 1835, Charles Island (CGE).].

(2). The widespread Galapagos tomato (Lycopersicon cheesmanii Riley) has
long been recognized to consist of two infraspecific taxa, f. cheesmanii and f. minor
(Hook. f.) Muller. However, recognition at a higher taxonomic rank is warranted,
and the following combination is proposed: Lycopersicon cheesmanii var. minor
(Hock. f.) D. M. Porter, comb. nov. [Basionym: Lycopersicon esculentum var.

minor Hook. f., Trans. Linn. Soc. London 20:202. 1847. Holotype: Darwin, beg. of |

Oct. 1835, James Island (CGE).].

(3). Opuntia megasperma var. orientalis (J. T. Howell) D. M. Porter, status |
novum [Basionym: O. megasperma subsp. orientalis J. T. Howell, Proc. Calif. |
Acad. Sci., ser. 4, 21:48. 1933. Holotype: Stewart 3003, Hood Island (CAS).]. |

Opuntia echios var. gigantea (J. T. Howell) D. M. Porter, status novum [Basi-

onym: O. echois subsp. gigantea J. T. Howell, op. cit. 51. 1933. Holotype: Howell |

9112, Indefatigable Island (CAS).].
These two taxa inadvertantly were included under the varietal rank in I. L. Wig-
gins and D. M. Porter’s Flora of the Galdpagos Islands (Stanford Univ. Press,

Stanford, 1971) by E. F. Anderson and D. L. Walkington in their treatment of the |

Cactaceae, although new status was neither proposed nor effected. Recognition at

the varietal level is desirable in order to conform with the classification of the |
genus in the archipelago. Where infraspecific taxa have been recognized in these |
species and in O. galapageia Hensl., they have been designated as varieties. Such |
trivial nomenclatural problems could be avoided if Raven, Shetler, and Taylor’s |
“Proposals for the simplification of infraspecific terminology” (Taxon 23:828-831. |

1978] NOTES AND NEWS 59

1974), which advocate recognition of a single infraspecific rank (subspecies), were
incorporated into the International Code of Botanical Nomenclature.

A grant from the Penrose Fund of the American Philosophical Society which
enabled me to examine Charles Darwin’s Galapagos collections at Cambridge Uni-
versity and the Royal Botanic Gardens, Kew during the summer of 1976 is grate-
fully acknowledged—Duncan M. Porter, Department of Biology, Virginia Poly-
technic Institute & State University, Blacksburg 24061.

Rare TAXA IN THE LITERATURE,—We were impressed while reading the July, 1977
issue of Madrono to note that three authors discussed three rare California taxa.
However, we were equally impressed by the omission of any reference to the facts
that these taxa are listed in the Inventory of Rare and Endangered Vascular Plants
of the California Native Plant Society and that two of them are listed as candidates
by the Office of Endangered Species, U.S. Fish and Wildlife Service, in the Federal
Register. We respectfully submit that these omissions are serious oversights because
presumably the authors have the best possible data concerning the status of the
taxa they are studying relative to rarity and endangerment. Being very rare is a
critical attribute of a plant possessing it.

As coordinators of the CNPS Rare Plant Project, we depend greatly upon the
botanical community which includes you, the readers and writers of Madrono and
similar publications. Please send your published and unpublished information about
rare plants to either of us for use by the CNPS project —W. RosBeErt Powe Lt, Di-
rector, CNPS Rare Plant Inventory, Dept. of Agronomy and Range Science, Univ.
of California, Davis 95616; and ALicE Q. Howarp, Chair, CNPS Rare Plant Ad-
visory Committee, Botany Herbarium, Univ. of California, Berkeley 94720.

POLLEN SHED AS TETRADS BY PLANTS OF ESCHSCHOLZIA CALIFORNICA (PAPAVER-
ACEAE) —Mature pollen of Eschscholzia, and of the rest of the Papaveraceae, is
normally shed from anthers as single grains (monads). In greenhouse-grown plants
from two populations of Eschscholzia californica Cham. (Clark 492—California.
Alameda Co.: ca. 2 mi SE of Livermore on S Livermore Rd, 24 May 1975; Clark
503—Butte Co.: Butte Canyon Rd, 0.9 mi E of junction with Manzanita Ave and
Centennial Ave, 25 Jun 1975), I observed that pollen was shed not only as monads,
but also as dyads, triads, and intact, generally tetrahedral tetrads. Individual plants
of Clark 492 present all monad pollen, or mixtures of monad, dyad, triad, and tetrad,
or nearly all tetrad pollen. Of the two plants of Clark 503 examined, one produced
all monad pollen and the other a mixture of monad, dyad, triad, and tetrad pollen.

Scanning electron micrographs of intact pollen tetrads are presented in Fig. 1.
Notice that individual grains are held together by bridges of pollen wall material.
Sachar and Mohan Ram (Phytomorphology 8:114—-124. 1958) state that “Wall for-
mation [to form microspores] occurs by furrowing.” In these populations furrow-
ing evidently does not proceed to completion, leaving bridges of pollen wall and
even cytoplasmic connections, which have been seen in light micrographs of pollen
stained with cotton blue. Similar exine bridges have been reported in the Onagra-
ceae (Skvarla et al., Amer. J. Bot. 62:6-35. 1975) and in the fossil Eomimosoidea
(Crepet & Dilcher, Amer. J. Bot. 64:714—-725. 1977), but unlike those of Esch-
scholzia, their tetrads are also bound together at the margins of the apertures.
Dyads and triads apparently result frcem furrowing which detaches only one or two
grains from the tetrad.

Both populations have high pollen stainability in cotton blue; meiosis observed
in Clark 492 was normal. Tetrad pollen appears to be functional—pollen from an
individual of Clark 492 which sheds almost all tetrads was able to effect full seed
set in other E. californica plants. The ability to form tetrads is evidently a heritable
trait, appearing in F, progeny of crosses betweén Clark 492 and other populations of
E. californica and thé tlosely related E. mexicana Greene, but dppearing in none of

60 MADRONO [ Vol. 25

Fic. 1—Scanning electron micrographs of tetrad pollen of Eschscholzia californica

over one hundred plants from 12 other populations of EF. californica and their
hybrids. However, the genetic basis of this inheritance cannot be estimated with the
data available.

Examination of pollen of pressed voucher specimens of Clark 492 (3 plants) and
Clark 503 (2 plants) and of flowers collected in April, 1977, from the approximate
location of Clark 492 revealed only monad pollen. Some greenhouse-grown plants
of Clark 492 present only monads, but it is somewhat surprising that there should
be no evidence of tetrad pollen in the natural populations, since the trait is so
common in their progeny. Perhaps these plants form tetrads in response to some
condition of the greenhouse environment.

I wish to thank Judith A. Jernstedt for invaluable assistance with the scanning
electron micrographs, and D. W. Kyhos for critical review of the manuscript. Pollen
vouchers are deposited in the Department of Botany, University of California,
Davis, and seeds for propagating the populations are available to interested investi-
gators from the author—CurtTis CLARK, Department of Botany, University of
California, Davis 95616.

COMBINATIONS IN THE GENUS ERIOGONUM (POLYGONACEAE) NOT PROPERLY PRO-
POSED BY Munz in “A FLoraA OF SOUTHERN CALIFORNIA.”—A few months after his
death, “A Flora of Southern California” was published by the University of Cali-
fornia Press (1974), and the final work and fitting tribute of Philip A. Munz was
made available to his friends, colleagues, and to the people of California. He and I
had collaborated on a treatment of the genus Eriogonum (Polygonaceae) for his
“Supplement to A California Flora” (University of California Press, 1968), but as
he wished to use the subspecific rank as his major infraspecific category in the 1974
book, I suggested he independently prepare the treatment and allow me to review
it. This was done. I advised him in certain matters regarding the use of subspecies
in Eviogonum, and he thereby avoided making some superfluous new combinations
for taxa which occur beyond the limits of southern California. However, possibly
through an oversight and no doubt aggregated by his illness, some of his proposed
new combinations in the flora were not validly published in accordance with the
International Code of Botanical Nomenclature, mainly because the basionym and
its place of publication were not cited. I cannot take into consideration all of these
various names, but as I communicated and talked with Dr. Munz on the names
associated with Eriogonum, and knew his intentions, I feel the oversights can and
should be corrected so that the names used in his flora are valid. Therefore the fol-

1978] REVIEW 61

lowing combinations are made here: Eriogonum kearneyi Tidestr. ssp. monoense
(S. Stokes) Munz ex Reveal, comb. nov., based on E. nodosum Small ssp. monoense
S. Stokes, Leafl. W. Bot. 3:201. 1943; E. spergulinum A. Gray ssp. reddingianum
(M. E. Jones) Munz ex Reveal, stat. nov., based on Oxytheca reddingiana M. E.
Jones, Bull. Torrey Bot. Club 9:32. 1882. The first name was credited to Stokes by
mistake, and in the second case the basicnym was not cited. Several months before
his death I called Dr. Munz and informed him that the type of E. elatum ssp.
glabrescens S. Stokes was actually a specimen of E. latens Jeps., and that a new
combination would be necessary. He agreed but apparently could not make the
change for his Flora. The proper name, he concurred, would be as follows: E.
elatum Dougl. ex Benth. ssp. villosum (Jeps.) Munz ex Reveal, stat. nov., based
on E. elatum var. villosum Jeps., Fl. Calif. 1:421. 1913.

Supported by NSF Grant BMS75-13063——James L. Reveat, Department of
Botany, University of Maryland, College Park 20742; and National Museum of
Natural History, Smithsonian Institution, Washington, D.C. 20560.

REVIEW

The genus Epilobium (Onagraceae) in Australasia: a systematic and evolutionary
study. By Peter H. Raven and TAMRA ENGLEHORN Raven. With illustrations by
Keith R. West. 321 pp. 1976. New Zealand Dept. of Scientific & Industrial Research
Bulletin 216. Published by A. R. Shearer, Government Printer, Wellington, N.Z.
$20.00 NZ.

This sumptuous botanical monograph may be termed a labor of love both liter-
ally and figuratively: the authors who published on New Zealand Epilobium in
1971 as Raven and Engelhorn are listed in 1976 as Raven and Raven. Although one
hesitates to use the overworked word “mastperiece” for a taxonomic revision in the
20th century, this book with its superb illustrations (partly in color) by Keith
West, attractive format, and meticulously detailed narrative harks back to the tra-
dition of the great illustrated folio volumes by Hooker of the 19th century. The
DSIR and the Government Printer merit applause from the botanical world for
publishing a work of such exemplary merit.

The work consists of four chapters of discussion preceding a systematic treatment
of the 50 species of Epilobium found in Australasia (mainly Australia, New Zealand,
and New Guinea). The essay on crossing and ecological relationships published in
1972 (in Valentine, Taxonomy, Phytogeography, and Evolution) is here expanded
to provide considerably more detail about the 45 native species.

The discussion of breeding systems in Australasian Epilobium owes a great deal
to the work of the New Zealand botanist W. B. Brockie (1897-1972), who made
over 900 crosses between the various taxa; the authors have appropriately dedi-
cated the volume to his memory. Studies by the Ravens on the pollen fertility of
Brockie’s voucher specimens show that, in contrast to the Holarctic species of Epi-
lobium, the Australasian ones readily form interspecific hybrids with a high degree
of fertility. However, there is a surprising difference between the Australian species,
which include a number of strongly outcrossing taxa, and the New Zealand species,
which are predominantly (33 out of 37) autogamous. The authors regard the
extraordinary radiation of Epilobium in New Zealand to be the result of extensive
hybridization between sympatric populations that are kept apart by autogamous
breeding systems and different habitat preferences (the latter graphically indicated
in some detail).

An unusual feature of the Ravens’ taxonomic interpretation of the New Zealand
Epilobiums is their treatment of sympatric populations. No less than 40 of the 45
Australasian species occur sympatrically, and in New Zealand the Ravens found 9
or 10 taxa occurring together in the same local population! In a distinctly heterodox
fashion, they have abandoned the conventional “subspecies displacement” rule, and

62 MADRONO [Vol. 25

permit subspecies as well as species to have overlapping geographical distributions.
This practice is counter to that followed by most taxonomists, who would treat
sympatric variants as either distinct species or as simply genotypes in a variable
population. To some extent, the situation in Epilobium may be unusual in that the
large degree of microgeographic habitat variation in New Zealand can permit geneti-
cally miscible but strongly inbreeding taxa to maintain themselves as distinct. Then,
too, as the authors candidly point out, there is an inherent degree of arbitrariness
involved in assigning rank to these proteanly differentiated populations. In any
event, researchers on other taxa might well reexamine the “sympatry rule” as it
applies to closely related geographically overlapping taxa in regions both in and
outside of New Zealand.

The phylogenetic model proposed by the authors involves a Holarctic origin of
Epilobium and an invasion of Australasia in the Pliocene. The small comose seeds
of Epilobium obviously “preadapt” the genus for long-distance dispersal, so that
there is no difficulty in explaining the spread and differentiation of the genus within
Australasia during the past 5-8 million years. The vagility of Epilobium is under-
scored by the Ravens’ observation that 5 of the 9 Australian species also occur in
New Zealand, and two of these species have made the crossing twice. In comparing
occurrences of taxa in the region, one encounters one of the few deficiencies in this
book: the lack of a table of the distributions. Since this can readily be extracted
from the data, it is provided here (figures in parentheses indicate endemic species).

New Guinea & Moluccas 4. (3) South Island 35 CEY)
Australia 8 (0) North & South I. 22» (9)
Tasmania 9 (1) Chatham I. 12°")
New Zealand 30: (21) Auckland/Campbell I. AG)
North Island 23 21)

What is particularly striking about these figures is the low number of Australian
native and endemic species and the very high number in the South Island of New
Zealand. This seems especially anomalous in view of the fact that the Australian
species, according to the Ravens, are the more primitive, and that New Zealand
was invaded from Australia (and New Guinea). Evidently the autogamous but
hybridizing breeding systems in the New Zealand taxa—as the authors suggest-—
have played a crucial role in adaptive response of the plants to the violent climatic
fluctuations of the Pleistocene. Some curious aspects of these distribution patterns,
in particular the notable focus of endemicity in the South Island, still await explana-
tion and may provide important clues to the Quarternary biogeographic history of
the Australasian region.

Until speciation in other Australasian genera is analyzed as thoroughly as has
been done for Epilobium, it will still be difficult to arrive at cogent generalizations
regarding the recent phyletic history of the antipodal flora. In the meantime, this
exemplary work of the Ravens will serve as a paradigm. of the kind of study which
needs to be done for a range of taxa with differing life forms and reproductive
economies.—Grapy L. WEBSTER, Department of Botany, University of California,
Davis 95616.

1978] BOOKS RECEIVED 63

Books RECEIVED AND LITERATURE OF INTEREST

Redwood national and state parks. By DONALD F. ANTHROP. 70 pp. 1977. Nature-
graph Publishers, Inc., Box 1075, Happy Camp, Cal. 96039. $3.95 paper, $7.95 cloth.

An introduction to the botany of the major crop plants. By A. M. M. Berrie
(ed.). x + 220 pp. 1977. Heyden & Son Ltd., Bellmawr, N. J. 08030.

List of California herbaria and working collections 1977. By THomMas G. FULLER
and G. DoucLas Bars. California Dept. of Food & Agriculture, Div. Plant Indus-
try—Botany Laboratory. 47 pp. This useful publication may be obtained free by
writing to the Botany Laboratory at 1220 N Street, Room 340, Sacramento, Cal.
95841.

Trees and shrubs of the United States: a bibliography for identification. By E.
L. LitrLe Jr. and BarsparA H. Honxata. U.S.D.A. For. Serv. Misc. Publ. 1336:
1-56. 1976.

The flora of Canada. By Homer J. Scoccan. 4 vols., publication commencing in
1978. Available from the Museum of Natural Sciences of the National Museums of
Canada as follows: Part 1 (General survey), 101 pp., 1978, $11.00; Part 2 (Pterido-
phytes, Gymnosperms, and Monocotyledons), 460 pp., 1978, $33.00; Parts 3 and 4
(Dicotyledons), c. 500, 575 pp., 1978, 1979. $41.00, $46.00.

Vascular plants of British Columbia. By Roy L. Taytor and Bruce MacBryper.
xxiv + 754 pp. 1977. Tech. Bull. 4, University of British Columbia Botanical Gar-
den, Univ. Br. Col. Press. An interesting computer-generated checklist.

REVIEWERS OF MANUSCRIPTS
The Editor of Madrono extends sincere thanks to the following persons who
assisted as reviewers of manuscripts and in many other ways in connection with
publication of Volume 24.—B.D.W.

Michael Barbour
Spencer C. H. Barrett
John H. Beaman
James P. Bennett
Bert Brehm

Roy Curtiss Brown
Annetta Carter

Curtis Clark

Lincoln Constance
Beecher Crampton
William B. Critchfield
Robert W. Cruden
Roger del Moral
Lauramay T. Dempster
Melinda F. Denton
Joseph Ewan

William J. Ferlatte
Thomas C. Fuller
Fred Ganders

James R. Griffin

J. Robert Haller

L. R. Heckard
Charles B. Heiser, Jr.
James Henrickson
William M. Heisey
Subodh Jain

Dale Johnson
Marshall Johnston

C. Eugene Jones

Steven P. Lynch
Tom Mabry

Jack Major

Roger McVaugh
Harold Mooney
Reid Moran
Rcdney G. Myatt
Robert Ornduff
Rexford E. Palmer
Robert W. Pearcy
Richard W. Pohl
Charles F. Quibell
John Reeder
Thomas L. Rost
Robert A. Schlising
Dale M. Smith
James P. Smith, Jr.
Richard Spellenberg
John Strother
Dean W. Taylor
Ronald J. Taylor
Alice Tryon

Frank C. Vasek
Grady L. Webster
John A. West
Kenneth Wells

Ira L. Wiggins
Robert Wilburn
David A. Young

64 MADRONO LVol.25

ANNOUNCEMENT OF CONTEST FOR EMBLEM OF
CALIFORNIA BOTANICAL SOCIETY

The California Botanical Society wishes to have an appropriate logo or emblem
on brochures, official correspondence, etc. Persons with artistic ability are encour-
aged to submit a design (the theme need not be restricted to the Madrofio, Arbutus
menziesii) by 2 April 1978, to the address below. Designs should be simple and
have a final size (after reduction) of approximately 7 by 7 or 7 by 10 cm. Designs
will be evaluated by a panel of members of the Society; the winner will receive a
$25 prize and published recognition of his or her work. Good quality xerox copies
of entries are acceptable, and none can be returned.—California Botanical Society,
Dept. of Botany, Univ. of California, Berkeley, CA 94720.

ANNOUNCEMENT OF PROGRAM FOR MEETINGS OF
THE CALIFORNIA BOTANICAL SOCIETY: SPRING, 1978

Jan. 19. Dr. Eduardo Zeiger, Stanford U.
How stomata respond to light: a new perspective.
Feb. 8. Dr. Kenneth V. Thimann, U.C. Santa Cruz.
Annual Banquet address: Some history of the study of plant growth
and growth hormones. (Claremont Hotel, Berkeley).
Mar. 16. Dr. Dan Cheatham, U.C. Berkeley.
Floating down the River Amazon: a 400-mile journey from Iquitos,
Peru to Leticia, Colombia.
Apr. 20. Dr. Michael R. Mesler, Humboldt State U.
The secret of subterranean sex of the Ophioglossaceae.
May 18. Dr. Michael F. Baad, Sacramento State U.
Adaptations of arctic vegetation.
All meetings held in Room 2003, Life Sciences Building, U.C. Berkeley, unless
otherwise noted.

REPORT ON BOTANY GRADUATE STUDENT MEETING

On November 12-13, 1977, the Department of Botany, University of California,
Berkeley, hosted the fourth annual Botany Graduate Student meeting, which is
sponsored by the California Botanical Society. The meeting was attended by nearly
150 botany graduate students and professional botanists from California, Oregon
and Washington. Graduate students presented papers on research projects in a wide
range of botanical disciplines, including physiology, anatomy, morphology, mycol-
ogy, phycology, pollination ecology, physiological ecology and systematics.

Each presentation was judged for content and presentation by a panel of gradu-
ate students. Robert N. Bowman, Botany Department, University of California,
Davis, received an award of excellence for his paper titled “Phylogenetic implica-
tions from Cuticular Wax Analyses in Epilobium canum (section Zauschneria) .”

The next Botany Graduate Student Meeting has been tentatively scheduled for
Fall, 1978. Traditionally, the meetings have been hosted alternately by northern
and southern schools, so a host for the 1978 meeting from the south or south-central
area is being sought. Students or faculty of departments interested in hosting future
meetings should contact L. R. Heckard, Department of Botany, University of
California, Berkeley, 94720 (415-642-2465) —Nancy Mortn, Department of Botany,
University of California, Berkeley, 94720.

Membership in the California Botanical Society is open to individuals ($12.00 per
year, regular; $8.00 per year, student). Members of the Society receive MADRONO
free. Institutional subscriptions to MapROoNO are available ($20.00 per year).

Back issues of Madrofio are available at the following rates (some issues are
out of print):

Vol. 1 (1916-1929) and Vol. 2 (1930-1934, each consisting of 17 numbers: $1.50
per issue and $25.50 per volume for members; $3.00 per issue and $51.00 per volume
for institutions.

Vol. 3 (1935-1936) through Vol. 23 (1975-1976), each biennial, consisting of 8
numbers: $3.00 per issue and $24.00 per volume for members; $5.00 per issue and
$40.00 per volume for institutions.

Vol. 24 (1977) et seq., one volume per year, each consisting of 4 numbers: $3.50
per issue and $14.00 per volume for members; $5.00 per issue and $24.00 per volume
for institutions.

Applications for membership (including dues), orders for subscriptions, requests
for back issues, changes of address, and undelivered copies of MApRONo should be
sent to the California Botanical Society, Inc., Department of Botany, University of
California, Berkeley 94720.

INFORMATION FOR CONTRIBUTORS

Manuscripts submitted for publication in MaproNo should be sent to the Editor.
Membership in the California Botanical Society is requisite for publication in
MaproNo.

Manuscripts and accompanying illustrative materials must be submitted in dupli-
cate and should follow the format used in recent issues of MaproNo. Original illus-
trations should not be submitted until paper is accepted for publication. All manu-
scripts MUST BE DOUBLE SPACED THROUGHOUT, including title, text, tables,
captions, lists, literature cited, etc. Footnotes (which should be avoided wherever
possible), captions, and tables must be typed on sheets separate from the text.
Presentation of nomenclatural matter (accepted names, synonyms, typification)
should follow the format used for Rhus integrifolia in MADRONO 22:288. 1974. All
measurements should be given in S. I. (metric) units. Where appropriate, scales
should be included on figures rather than in captions as estimates of ratio of repro-
duction such as X %4, X 1620, etc. Institutional abbreviations in specimen citations
should follow Holmgren and Keuken’s list (Index herbariorum, Part 1. The herbaria
of the world. Sixth edition. 1974. Regnum Veg. vol. 92). Abbreviations of names of
journals should be those in Botanico-Periodicum-Huntianum (Lawrence, G. H. M.
et al. 1968. Hunt Botanical Library, Pittsburgh). If the correct abbreviation cannot
be determined, the full title of the journal should be used. Titles of books should be
given in full, together with the place and date of publication, names of publisher,
and an indication of the edition, if other than the first.

Short articles such as range extensions and other brief notes are published in con-
densed form under the heading “Notes and News”. Authors of such articles should
follow the format used in recent issues of MapRONO.

Authors are allowed up to 10 pages per year without page charges; charge for
additional pages is $30.00 per page. Subject to approval by the Fditors, articles may
be published ahead of schedule, as additional pages of an issue, provided the author
assumes complete costs of publication.

ADRONO

VOLUME 25, NUMBER 2 APRIL 1978

Rupert C. BArRNEBY, Daleae Imagines, an illustrated revision of
Errazurizia Philippi, Psorothamnus Rydberg, Marina Liebmann,
and Dalea Lucanus emend. Barneby, including all species of
Leguminosae tribe Amorpheae ever referred to Dalea
(Grady L. Webster) 111

SPECIAL NOTICE: SCIENTIFIC COLLECTING IN MEXICO 111

|
| > Contents
| Zz MARITIME CHAPARRAL AND ENDEMIC SHRUBS OF THE MONTEREY BAY
< ReEcIon, CaLrrorniA, James R. Griffin 65
| = CIRCUMSCRIPTION AND GENERIC RELATIONSHIPS OF GALINSOGA
(ComposiraE, HELIANTHEAE), Judith M. Canne 81
| ‘o) ON THE TAXONOMIC STATUS OF FRITILLARIA PHAEANTHERA EASTW.
faa] (Lrzraceaz), Roger M. Macfarlane 93
| A NEw SPECIES OF DRABA (CRUCIFERAE) FROM WYOMING AND UTAH,
[x4 Robert D. Dorn 101
| © NOTES AND NEWS
VALIDATION OF THE NAME JUNCUS BUFONIUS VAR. OCCIDENTALIS,
pl F.J. Hermann 104
| < PLANT ABUNDANCE AND DISTRIBUTION IN RELATION TO TYPES OF
SEED DISPERSAL IN CHAPARRAL, Stephen H. Bullock 104
| Z, GreaT BASIN VEGETATION IN CARBON CouNTY, Montana,
fara Robert D. Dorn 105
KNOBCONE PINE SOUTHWARD RANGE EXTENSION IN THE SIERRA NEVADA,
~) Jon E. Keeley, Sterling C. Keeley, and Janet Lee 106
| (e) SCROPHULARIA LAEVIS (SCROPHULARIACEAE), A LEGITIMATE SPECIES,
| _ Thomas K.Todsen 106
CorRDYLANTHUS MOLLIS SSP. MOLLIS (SCROPHULARIACEAE) , REDISCOVERY
Z, oF Extinct Recorp WiTHIN Napa County, CALIFORNIA,
Stephen P. Rae 107
< TRAGUS RACEMOSUS IN ARIZONA,
\@) John R. Reeder and Charlotte G. Reeder 107
ro REVIEWS
faa NorMAN WEEDEN, A Survival Handbook to Sierra Flora
fx) (Dean W. Taylor) 108
Janice C. BeaTLey, Vascular Plants of the Nevada Test Site and
= Central-Southern Nevada: Ecological and Geographical
< Distributions (Dean W. Taylor) 109
NicHoLas T. Mrrov and JEAN Hassrouck, The Story of Pines
i (Annetta Carter) 110
7p)
fi)
<

JBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY

MaproNo is published quarterly by the California Botanical Society, Inc., and is
issued from the office of the Society, Herbarium, Life Sciences Building, University
of California, Berkeley. Established 1916. Second-class postage paid at Berkeley.
Return requested.

Editor — BarBaraA D. WEBSTER
Department of Agronomy and Range Science
University of California, Davis 95616

Associate Editor — Grapy L. WEBSTER
Department of Botany, University of California, Davis 95616

Board of Editors
Class of:

1978—SHERWIN CaARLQUIST, Claremont Graduate School
Les.iE D. Gortrtiies, University of California, Davis
DeEnnIs R. PARNELL, California State University, Hayward
1979—Puitip W. RunpDEL, University of California, Irvine
ISABELLE TAVARES, University of California, Berkeley
1980—JamMEs R. GrirFin, University of California, Hastings Reservation
Frank A. Lane, Southern Oregon College, Ashland
1981—DanIEL J. CRAWFORD, University of Wyoming, Laramie
JaMEs HEnricKsEN, California State University, Los Angeles
1982—Dean W. Tayvtor, University of California, Davis
RICHARD VoGL, California State University, Los Angeles

CALIFORNIA BOTANICAL SOCIETY, INC.

OFFICERS FOR 1978

President: Dr. MicHAEL F. Baap, Department of Biological Sciences,
California State University, Sacramento, California 95819

First Vice President: Dr. ALAN R. PoLANSHEK, Department of Biological Sciences,
San Jose State University, San Jose, California 95114

Second Vice President: Dr. AMy GitMarTIN, Department of Botany,
Washington State University, Pullman, Washington 99163

Recording Secretary: Dr. CHARLES F. QUIBELL, Department of Biological Sciences,
California State University, Rohnert Park, California 94928

Corresponding Secretary: Dr. RUDOLPH SCHMID, Department of Botany,
University of California, Berkeley, Berkeley, California 94720

Treasurer: Dr. JoHN H. THomas, Department of Biological Sciences,
Stanford University, Stanford, California 94305

The Council of the California Botanical Society consists of the officers listed above
plus the immediate Past-President, WINSLOW R. Briccs, Carnegie Institute of Wash-
ington, Stanford, CA 94305; the Editors of Madronio; three elected Council mem-
bers: James R. GrirrFin, Hastings Reservation, Star Route 80, Carmel Valley, CA
93924 (1976-1978) ; Harry D. THiers, Department of Ecology and Systematic Bi-
ology, California State University, San Francisco, CA 94132 (1977-1979) ; ALAN R.
SmiTH, The Herbarium, Department of Botany, University of California, Berkeley,
CA 94720 (1978-1980) ; and a Graduate Student Representative, Nancy R. Morin,
Department of Botany, University of California, Berkeley, CA 94720.

MARITIME CHAPARRAL AND ENDEMIC SHRUBS OF THE
MONTEREY BAY REGION, CALIFORNIA

JAMES R. GRIFFIN
Hastings Reservation, Carmel Valley, California 93924

Serious plant collecting started near Monterey in 1786, and the re-
gion’s flora is now well known (Howitt and Howell, 1964, 1973; Thomas,
1961). Much has been written about the local manzanitas (Gankin, 1966,
1971; Hildreth, 1976; Hoover, 1964; Howell, 1952; Roof, 1961, 1964,
1976; Wells, 1968). The Monterey Peninsula closed—cone pine and
cypress habitats as well as genetic and physiological features of conifers
on these habitats have also been studied (Cannon, 1913; Duffield, 1951;
Dunning, 1916; Forde, 1966; Lindsay, 1932; McDonald, 1959; Mc-
Millan, 1956; Vogl et al., 1977; Wolf, 1948). However, this knowledge
about trees and shrubs is not matched by vegetative data, particularly
for the maritime (or fog-belt) chaparral around Monterey Bay which
has a conspicuous endemic element.

Chaparral patches dominated by endemic Arctostaphylos taxa along
with Adenostoma fasciculatum appear locally within a coastal scrub-live
oak forest-closed-cone conifer forest mosaic. Unfortunately, urban de-
velopment is rapidly destroying this unique chaparral. For this reason, I
surveyed species composition and relative dominance in the least dis-
turbed maritime chaparral remnants between Larkin Valley, Santa Cruz
County and Malpaso Creek, Monterey County (Fig. 1).

METHODS

While surveying the Monterey Bay region for rare plants (Powell,
1974), I studied distributional patterns of six endemic shrubs (Table 1).
That survey led to this project in which I sampled more carefully regions
with major populations of one or more of the endemics (Fig. 1). Al-
though the project did not specifically include Ceanothus dentatus, my
sampling for the other endemic shrubs essentially covered the distribu-
tion of typical C. dentatus.

Between October 1975 and March 1977, 33 stands were analyzed
with the Braun-Blanquet relevé method adapted from Muller-Dombois
and Ellenberg (1974). Widespread disturbance in the maritime chap-
arral forced me to use stands with less than desirable homogeneity in
aspect, slope, and soil depth. Relevés were placed in the least disturbed
portions of each stand; a given stand covered several hectares. There was
a total of 284 relevés, usually 6-10 per stand. All relevés were 10 m by
10 m squares. Cover-abundance ratings were assigned to all vascular
plants on the first visit, and stands were revisited several times looking

Madrono, Vol. 25, No. 2, pp. 65-112, June 15, 1978

65

66 MADRONO (Vol. 25

‘4
Pay me

9 ul
+

ey
a y
Q
¢O !/e
: eo ped dnett tat
A. hookeri Qo ue
Cemex ARXLAA ni te A
A. montereyensis etait oO.
nS pajaroensis resend sbag dea bs E aL 21 aese ib Lo
Ree) ele eat
A. pumila ee Ee ON _
Grrigiaus Br hsp , Woe sy
E. fasciculata ce oe | HEIR
ea eee oo
* POREST LAKE ae
S
Cy.
‘y
MaOtn Etec oR
Bay cm Ly
Hee \ T/.
$4 ;)
il wa
ea v.
SON fe
e 15 GE era] Vets
PENINSULA / (ee ee 5 Te
IE oe Semin CG Ol ®
ise Seat é mae ‘ey PAR
ee : ifr SE V,
ees Nien ae
a aot as.
eye 5
Cr. On,
GIBSON C Pe
CANYON eave al
& y fo)
zs ‘ L 36.5
oY aS CY
° Sv) Ne af iS
122 ET. ve ye Ves
"te 10, (Ke

Fic. 1. Present distribution of six Monterey Bay endemic shrubs, extinct popu-
lations in urban areas not included; relative positions of the 33 sample stands indi-
cated by the “locality”’ numerals.

for additional herbs and errors in cover estimates. Vouchers for most
taxa are filed at the Pacific Grove Museum of Natural History; some
particularly interesting specimens are at JEPS.

Dr. Dean W. Taylor produced an association table from the stand
data by a modified algorithm method of Ceska and Roemer (1971).
Stands with similar ranges of geological and topographic conditions
were grouped into 15 “localities” for convenience of some analyses and
discussion (Fig. 1, Table 2)

1978] GRIFFIN: MARITIME SHRUBS 67

TABLE 1. SHRUB SPECIES WHICH IN TYPICAL ForM ARE ENDEMIC TO MONTEREY
Bay; in some cases the differentiation between the Monterey taxon and related
coastal taxa is controversial.

Arctostaphylos hookeri G. Don
Excluding ssp. ravenii, A. franciscana, and A. hearstiorum
(Hoover and Roof 1966, Roof 1976, Wells 1968)

A. montereyensis Hoover
THREATENED (Smith. Inst. 1974)

A. pajaroensis Adams (A. andersonit var. pajaroensis Adams ex McMinn)
A, pumila Nutt.
ENDANGERED (Dept. Int. 1976, Howell 1952, Roof 1961)

Ceanothus dentatus T. and G.
Excluding C. hearstiorum and intermediates with C. foliosus and C. papillosus
(Hoover and Roof 1966, McMinn 1961)
C. rigidus Nutt.
THREATENED, excluding C. ramulosus
(Hoover 1970, Nobs 1957, Smith 1976, Smith Inst. 1976, Thomas 1961)
Ericameria fasciculata (Eastw.) McBride (Haplopappus eastwoodae Hall)
ENDANGERED (Dept. Int. 1976)

TABLE 2. AVERAGE SEASONAL PRECIPITATION, AVERAGE SLOPE, ELEVATIONAL RANGE,
DisTaNcE INLAND, ESTIMATED ABSOLUTE TREE AND SHRUB COVER, AND AVERAGE
NUMBER OF SPECIES PER RELEVE FOR MARITIME CHAPARRAL LOCALITIES.

Locality and Seas. / Cover : Ave no.
(no. relevés) prec. Slope Elev. Dist. Tree Shrub sp./relevé
cm % m km % %
1 Larkin V.(20) 55 7 120-145 4-6 1 84 Tia
2 Prunedale(62) 52 i7 75-160 9-14 2 85 8.2
3 Toro Park(10) 13 310 15 0 81 17.4
4 Toro Park(22) Zi 230-290 15 0) 79 Bee
5 Fort Ord(20) 11 60-150 6-11 2 84 10.0
6 Fort Ord(28) 35 2 45-140 3-4 0 85 10.4
7 Airport(17) 13 60-180 4-7 1 95 6.2
8 Airport(29) 31 i 75 2-4 i my) 11.0
9 Penin.(6) 42 38 Z10 3 55 54 11.0
10 Penin.(12) 13 120 3 11 84 15.7
11 Penin.(6) 46 9 105 2 20 68 15.0
12 Penin.(12) di 120 2 40 82 bo:5
_ 13 Gib. C.(10) 53 11 105 1 8 63 ier
| 14 Gib. C.(10) 26 150-210 1 0 90 8.2

| 15 Jacks P.(20) 24 170-210 5 1 83 3.3

68 MADRONO [Vol. 25

-~ENVIRONMENTAL FEATURES

Climate. Abundant precipitation data are available for the Monterey
region. Marginal records for the city of Monterey go back to 1847, and
reliable records started at Forest Lake in 1896 (Fig. 1). The Forest
Lake 80-year seasonal average was 457 mm; the extreme low 191 mm,
extreme high 855 mm. In 1957 the Meterology Department at the U.S.
Naval Postgraduate School, Monterey began compiling rainfall data
from a network of over 40 stations around Monterey. Preliminary sum-
maries from this unpublished data suggest large differences between
study localities (Table 2). The 1975-76 season in which this project
started was dry and most stations received about 50% of average rain-
fall. The airport had only 175 mm of rain in 1975-76.

All localities have summer drought moderated by fog. oun”
data on amounts of fog are limited to the Monterey climatological sta-
tion and the airport. Unpublished records at the Monterey station from
1963-76 averaged 135 foggy days per year. Monthly distribution of fog
during the summer averaged:

days with fog hours fog/fog day

June 14 9
July 20 13
August 22 14
September | ie! ef) 12

At the airport fog is most frequent at 0600, and it remains until 1000 on
25% of the days. Farther inland the fog dissipates sooner. Toro Park
probably has the least summer fog, followed by Prunedale. |

McDonald (1959) estimated fog drip at many Peninsula spots. Maxi- —
mum drip during McDonald’s superficial single-season study was re-
corded on Huckleberry Hill in the central Peninsula where 144 mm |
accumulated during a one-week period in August. I have observed wet |
spots on the ground under Pinus radiata trees in the same area on August |
mornings. But the major fog effect must be the widespread reduction of ©
evaporative stress and not the localized addition of soil moisture. Most |
of the closed-cone forests and maritime chaparral stands do not have
enough fog drip to wet the soil.

Temperature data for specific study localities are not available. From |
1952-76 the Monterey climatological station had a mean annual tem-
perature of 13.5° C. Average minimum and maximum for January, the |
coldest month, were 5.9° and 15.5° C; for September, the warmest |
month, 11.3° and 22.2° C. The daily extremes ranged from — 5.0° to
38.3° C. Temperature fluctuations would be greater at the more inland —
localities.

Geology and soils. General geologic features of the study localities were
shown by Jennings and Strand (1958). Youngest soi] parent materials

1978] GRIFFIN: MARITIME SHRUBS 69

are late Pleistocene dunes which blew off the Monterey Bay shore. These
stabilized ‘“‘pre-Flandrian” dunes extend from near Larkin Valley south
across Fort Ord to the airport (Cooper, 1967). Chaparral covered soils
on the dunes were mapped as Baywood sand; a few spots might be
Oceano loamy sand (Soil Cons. Serv., 1975).

The loosely consolidated “red sand” beds of the mid-Pleistocene Aro-
mas formation (Allen, 1946) are important to this project; 60% of the
relevés were either directly on this material or on a thin layer of pre-
Flandrian said over Aromas standstone. Aromas sandstone extends from
Larkin Valley through the Prunedale hills and Fort Ord to the airport.
Aromas sandstone underlies the pre-Flandrian dunes at Fort Ord and the
airport (Bowen, 1966). The Monterey County chaparral soils on Aro-
mas sandstone were mapped as Arnold loamy sand (Soil Cons. Serv.,
1975). Storie et al. (1944) mapped the chaparral soils near Larkin Val-
ley as Moro Cojo loamy sand, which is now synonymous with Arnold
series in Monterey County.

The lower portion of Toro Park lies on Plio-Pleistocene Paso Robles
formation sand and gravel deposits (Bowen, 1966). Aromas formation
was not mapped at Toro Park, but one sandy swale in the Paso Robles
sediment area resembles Aromas sand. In fact, the chaparral soils on
Paso Robles formation were also mapped as Arnold loamy sand (Soil
Cons. Serv., 1975). The upper Toro Park slopes are on garnetiferous
quartz monzonite (Ross, 1977); chaparral soils on these granitics were
mapped as Cieneba fine gravelly sandy loam.

The basement rocks of the Peninsula and Gibson Canyon are porphy-
ritic granodiorite (Ross, 1977). During the Pleistocene a series of marine
terraces were cut into these granitics (Bowen, 1966; Hart, 1966). The
Peninsula localities were either on these terrace sands or the adjacent
granitics from which the terrace deposits have been eroded away. Gra-
nitic soils with chaparral were mapped as Sheridan sandy loam; soils on
the terrace deposits were Narlon fine sand and often have a strong clay-
pan. (Soil Cons. Serv., 1975).

_ Gibson Creek drains a granitic basin and crosses a small area of
Paleocene Carmelo formation conglomerate-sandstone (Bowen, 1966).
Terrace deposits must have been eroded off the lower slopes here, for all
the relevés had rounded cobbles scattered on the granitic outcrops. The
_ granitic slopes with chaparral were mapped as Cieneba fine gravelly
sandy loam (Soil Cons. Serv., 1975). Soils on the Carmelo formation
were mapped as San Andreas fine sandy loam; however, so much rock
outcrops that there is no “soil” in an agricultural context.

I found maritime chaparral on shale only near Jacks Peak Regional
Park within an area mapped as upper Miocene Aquajito shale (Bowen,
1966). Soils on the shale have been described as Santa Lucia shaly clay
loam (Soil Cons. Serv., 1975), but there are some sandstone strata
within the shale beds with sandy loam soils.

70 MADRONO [Vol. 25
REGIONAL VEGETATION PATTERNS

The unpublished maps and notes of the Vegetation Type Map (VTM)
survey for quadrangles 105 A,B,C,D and 106 B are important vegetation
descriptions. Colwell (1977) discussed the background of this old survey.
Field maps and rough type descriptions were completed between 1930
and 1934.

The study localities are associated with a gap in the Coast Range ridge
and there is also a gap in well developed coastal forests here. Redwood
forest in the Santa Cruz Mountains grew south to Larkin Valley prior to
logging (Gordon, 1974), and scattered Sequoia sempervirens trees still
grow near the Larkin Valley locality. Redwoods are absent across the
lowland, and the northern outpost of the Santa Lucia Range redwood
forest is on San Jose Creek near Gibson Creek.

Closed-cone conifer forest. The Monterey and Point Lobos Peninsulas
lie at the southern end of the lowland and are covered with closed-cone
conifer stands. These stands, which include two pine and two cypress
species, are more extensive and better developed than the other disjunct
closed-cone communities along the central California coast (Vogl et al.,
1977). Conifer stands relevant to maritime chaparral are Pinus radiata
woodland, P. muricata woodland, P. muricata-Cupressus goveniana
dwarfed woodland, and C. goveniana dwarfed woodland. Pinus radiata
local distribution occurs between the airport, where the dunes meet the
steeper Peninsula hills, and Malpaso Creek to the south. Within this
range the woodlands with chaparral understories are mostly associated
with poorly drained Narlon clay-pans or very rocky soils. |

Live oak forest. Depauperate forms of mixed hardwood forest (Sawyer
et al., 1977) occur on north slopes and well-drained canyon bottoms
across the Monterey Bay lowland. Arbutus menziesii is rare and Litho-
carpus densiflorus is absent from these forests; Quercus agrifolia is the
only dominant tree. These live oak stands differ little from live oak com-
munities in other central coast lowlands. The live oak stands are better |
developed in the Prunedale hills than in the Fort Ord area. On the Penin- |
sula, live oaks shift to a subdominant position in the P. radiata commu-
nities. Oaks are negligible in the dwarfed conifer woodlands.

Chaparral. Typical chaparral does not approach the coast closely on
shale, but maritime chaparral penetrates well into the scrub zone on the |
sandhills. Cooper (1922) noted that Monterey Bay chaparral had a high ©
species diversity, and he considered Ceanothus rigidus, Arctostaphylos —
hookeri, A. pumila, and a form of A. tomentosa (A. vestita Eastw.) as
endemic to Monterey. Shreve (1927) discussed five types of chaparral —
in the Santa Lucia Range. His comments relating to Monterey Bay chap- —
arral were: ‘“‘ddenostoma type” was uncommon within 5 km of the coast;

1978] GRIFFIN: MARITIME SHRUBS #1

“Arctostaphylos type” with A. tomentosa was common only aoe the
coast, and near Monterey other manzanitas were added to it; “mixed
Beactal type” had much Toxicodendron diversilobum and Serer Sal-
via mellifera and Ericameria ertcoides in it.

Near Larkin Valley Arctostaphylos tomentosa ssp. crinita dominates
a tall chaparral. To the south less typical forms of this Santa Cruz Moun-
tains taxon are rare on Fort Ord and the Peninsula. The VIM survey
mapped a xeric phase of the ssp. crinita chaparral near Larkin Valley
which included A. hookeri as a local dominant. In areas transitional to
woodland, shrubs such as Corylus cornuta and Ceanothus thrysiflorus
intermingle with the manzanitas. These thickets are similar to Shreve’s
(1927) “Ceanothus type” of chaparral. The Corylus is absent to the
south except for a few shrubs near Prunedale.

In the Prunedale hills the VIM survey recognized two chaparral
types—a dwarfed “chamise” type and a mixed “chaparral” type. The
latter had Arctostaphylos pajaroensis and A. tomentosa ssp. crustacea as
important elements. Both types, which were in the foggy zone, were dis-
tinguished from a widespread inland non-foggy type. The VTM survey
noted that Ceanothus rigidus was widely scattered near Prunedale and
was favored by occasional fires. Davis (1972) studied the structure of
11 A. pajaroensis stands in detail.

Fort Ord chaparral has strong similarities to Prunedale chaparral even
though A. pajaroensis is absent. Several forms of A. tomentosa grow at
Fort Ord, and A. montereyensis and A. pumila are locally important.
Critchfield (1971) published a VIM profile transect across the Peninsula
that included one corner of Fort Ord.

The Toro Park region has chaparral covering several thousand hectares
of steeper terrain rising to 700 m. From a distance this brushland appears
to be typical chaparral, and the VTM survey mapped it all as “Adeno-
stoma type”. However, endemic shrubs are locally common along with
scrub species. The endemic shrubs have not been reported farther inland
than Toro Park. The Toro Park chaparral is separated from Fort Ord
chaparral by a zone of grassland and savanna on Santa Ynez clay-pan
soils (Soil Cons. Serv., 1975).

Maritime chaparral with Monterey endemics does not continue far to
the south of Malpaso Creek. Another endemic Arctostaphylos edmundsu
starts a few kilometers farther south, but this coastal bluff manzanita
does not seem to have as close an ecological connection with the sandhills
as do the other endemic shrubs. Ceanothus rigidus is scattered into typi-
cal chaparral east of Gibson Creek, and the VTM survey mapped it as a
minor chaparral component 8 km inland. Ceanothus rigidus appears in
less typical form as far south as Big Sur. To the south Ceanothus den-
tata merges into C. papillosus.

12, MADRONO [Vol. 25

Coastal scrub. Scrub types have a greater continuity across: the lowland
than either chaparral or live oak forest. To the lee of the active dunes
wind-pruned scrub was almost continuous prior to agricultural: develop-
ment. Farther inland scrub is more conspicuous on shale than on the
sandhills. Elements of both the northern coastal scrub (Heady et al.,
1977) and coastal sage scrub (Mooney, 1977) are included. Shreve
(1927) noted that Artemisia californica, Salvia mellifera, Ericameria
ericoides, and shrubby Lupinus spp. were important in the “xerophytic
scrub” near Monterey. He was impressed with the dense stands of E.
ericoides along the coast, and he noted that this shrub extended inland
on sandy soils. The VIM survey made similar observations. —

At low elevations the VT'M survey often mapped Artemisia californica
as a dominant on south-facing slopes. Higher on the slopes or on north
aspects Baccharis pilularis was important along with Toxicodendron
diversilobum, Mimulus aurantiacus, and Rubus vitifolius. On the VIM
maps Salvia mellifera was often shown as a minor associate of the chap-
arral species.

MARITIME CHAPARRAL STRUCTURE

Tree stratum. Pine stands around Monterey have gone through several
generations since the Spanish came, and current urbanization makes it
impossible to determine the density of “natural” stands. The present
pine woodland on most of the Peninsula localities regenerated after a
severe fire in 1901, and woodcutters have lightly thinned some stands in
recent times. The scattered pines at the Airport are younger, mostly less
than 50 years old. In any case, endemic shrubs currently grow under the
more open pine woodlands; tree canopy did not exceed 40% cover on the
relevés. These shrubs are also absent under the remaining examples of
closed-canopy, rapidly growing pine forest.

The non-conifer localities all had live oak forest nearby, but on my
relevés Quercus agrifolia consisted of tall shrubs or scattered small trees.
Live oaks grew on only 24% of the relevés and seldom formed more than
5% cover. Some live oaks in the pine localities may become medium-
sized trees in time, but the scrubby oaks on the exposed sandstone ridges
show little sign of forming real trees. Several early naturalists commented
on the low stature of live oaks on the sandhills.

Shrub stratum. Collectively the relevés had 31 shrub taxa. The sandy
locality at Toro Park had the greatest shrub diversity with 15 shrub
taxa; one relevé there had 11 shrub species. In contrast, the adjacent
Toro Park locality with shallower, less sandy soils had many relevés with
only two shrub species. Climate, cultural history, and recent fire history
were the same for both localities. One Jacks Peak relevé had only a single
shrub species; Jacks Peak was the only locality involving shale parent

1978] GRIFFIN: MARITIME SHRUBS 73

TABLE 3. RELATIVE DOMINANCE OF COMMON SHRUBS IN MARITIME CHAPARRAL
AS EXPRESSED IN PERCENTAGE OF RELEVES IN COVER-ABUNDANCE CLASSES AND THE
PERCENT OF ALL RELEVES WHICH CONTAINED EACH SPECIES (CONSTANCY)

Cover-abundance class
R + 1 2 3 4 5 Constancy

Jo Yo
SCLEROPHYLLS
Arctostaphylos tomentosa 1 1 a S32 eee 82
Adenostoma fasciculatum 6 20 28 15 8 1 78
Arctostaphylos hookeri 1 7 12 16 52 43
Ceanothus rigidus 5 18 9 3 35
Arctostaphylos pumila 1 3 3 6 6 3 22
Arctostaphylos pajaroensis il 8 Seo 22
Heteromeles arbutifolia 1 Soe lt 3 18
Arctostaphylos montereyensis 1 4 6 4 1 16
NON-SCLEROPHYLLS
Salvia mellifera 2 5 20 16 1 44
Mimulus aurantiacus 1 21 £20 1 43
Baccharis pilularis 3 OF ati 4 |
Ericameria fasciculata 2 7 #11 20
Toxicodendron diversilobum Z 7 8 2 19
Ericameria ericoides ) 7 2 12

materials. No introduced shrubs grew on the relevés, but Cytisus mon-
spessulanus was common on seriously disturbed sites near several stands.

Average absolute shrub cover for all localities was 80% (Table 2),
and only 6 of 284 relevés had less than 50% shrub cover. Sclerophylls
usually dominated over scrub species (Table 3). Collectively the scrub
species probably did not exceed 25% cover on any relevé, and single
scrub species seldom had more than 5% cover (Table 3).

The height and density of the chaparral varied considerably within
stands. On parts of the most exposed ridges old-growth chaparral was
less than 1 m tall with bare spots between shrubs. Such low chaparral in-
cluded similar looking dwarfed Adenostoma fasciculatum-Arctostaphylos
hookeri patches on granitic, Aromas sandstone, and shale outcrops. On
adjacent slopes with deeper soils the manzanitas were often impenetrably
dense and over 3 m tall. I found some A.. montereyensis shrubs about 6 m
tall on Fort Ord, and Davis (1972) reported A. pajaroensis up to 7 m
tall near Prunedale.

Every relevé had at least one manzanita species; some had three. The
burl-forming Arctostaphylos tomentosa complex (Wells, 1968) was the
most widespread and dominant shrub studied (Tables 3, 4). In localities
with A. tomentosa ssp. tomentosa the tomentose twig, tometose leaf f.

[Vol. 25

1/ST €/08 7/6 7/56 Cadsypef st
7/02 1/08 T/0S ¢/O0I ¢/OOl ‘OID FI
7/001 1/OT 7/06 Z/O0T 7/06 ‘J ‘qID EI
¢/L1 ¢/O0T +/OOT Z/O0I uludsg ZI
7/LT T/LT 7/001 ain oS 1/0S =1/ 29 v/00l ‘uluad TT
1/SZ 1/26 ¢/00T 1/726 Z/OOT ¢/O01 ‘ulus OT
¢/O0T ¢/O0T 7/OS 7/€8 ‘UIUdg 6
7/29 +/O0T 1/8¢ 1/8¢ 1/6L 7/£6 yaodiry g
iS ¢/9S ¢/$9 1/21 =F /S¢ 7/88 ¢/OOT yaodiry /
()
A 1/9 T/0S 7/SS ¢/O0T ¢/O0T ‘OHO 9
< 1/19 ¢/9¢ 1/12 1/9b 7/SL 7/68 ¢/6¢ €/Ls ‘Oo ¢
7/08 1/OOT 1/OS ¢/06 ¢/06 "dq 010, +
7/9 I/8T +/OOT 1/81 ‘"qoO10oy ¢
b/OOT ¢/bP =) TE 1/Z¢ 1/9S ¢/09 ‘ounig 2
¢/O0I Z/Ol 1/08 €/S6 ‘A uUriey I
“ULIO “snd “9UI0} AyyeI0'T
‘9A03 "Ipel ‘efed *qUOU ‘tuind “yooy ISR] ISI “ISB] ‘au07 ‘ou07 *9UIO}
‘adng snulg “ply “pry ‘pry “piy “MAD ‘ued ‘UIpy “piy “pry “piy

“ALIIVOO'T HOVY

ey LIV SaIogdSg AGOOAA GYLOATAS AOA SSVIZ) AONVANNAV-AAAO™) TIVGOJ /AONVISNOD Ad daSssdddxY AONVNIWO( NI SAGNAXT OlHdvad0gdy ‘Pp ATIVE,
~~

1978] GRIFFIN: MARITIME SHRUBS 75

tomentosa was most common; the glandular-setose twig, tomentose leaf
f. trichoclada was uncommon; the glabrescent f. hebeclada did not appear
on the relevés. An undescribed setose twig, glabrous leaf form was con-
spicuous near Jacks Peak. Although it becomes locally common south of
Gibson Canyon, ssp. vosez was not on or near the localities.

Adenostoma fasciculatum was the only conventional chaparral species
that was widespread and important in the maritime chaparral (Tables
3, 4), but it exceeded 50% cover only in Toro Park. A number of sclero-
phyll shrubs with broad distributions elsewhere were present in the mari-
time chaparral in minor amounts: Ceanothus papillosus, Dendromecon
rigida, Garrya elliptica, Lepechinia calycina, Pickeringia montana, Quer-
cus wislizentit, Rhamnus californica, and R. crocea.

Salvia mellifera and Mimulus aurantiacus were the most important
scrub species. Salvia was absent in the pine woodland stands; Mimulus
was present everywhere but was most vigorous in the pine woodland.
Artemisia californica grew near several stands but did not appear on any
relevé.

Herb stratum. The herb flora did not develop fully during the 1975-76
drought; only some 80 species were identifiable. The most diverse herb
flora was under the pine woodland on the Peninsula and the sandy lo-
cality at Toro Park. One relevé on the Peninsula had 16 herb species;
one at Toro Park had 13. The most widespread herb was Carex brevi-
caulis which grew in 13 localities but with a cover usually less than 1%.
Eriophyllum confertiflorum, Gnaphalium californicum, and Zygadenus
fremontii were also common in low densities. None of the herbs reached
significant cover in any stand. The rhizomatous grass Agrostis hallit was
the only herb to have 100% constancy in any stand and over 5% cover.

Introduced species were inconspicuous in intact maritime chaparral.
Aira caryophyllea and Vulpia myuros were most common in sandy open-
ings at the airport, but they still had less than 1% cover. Only a few
annual Bromus spp. plants were present. On disturbed spots adjacent to
several stands Carpobrotus edulis and Cortaderia jubata were serious
invaders.

Species Groups. Since my sampling was biased towards stands contain-
ing endemic shrubs, classification of regional vegetation types is inappro-
priate from this data. However, the association table derived from the
maritime chaparral data does reveal several associative tendencies worth
mentioning.

The most conspicuous group of associated species grew on the Penin-
sula stands. Differential species in this closed-cone forest species group
in addition to the pines and cypress included: Vaccinium ovatum, Agro-
stis hallii, Sanicula laciniata, Achillea borealis ssp. californica, and Iris
douglasiana.

76 MADRONO [Vol. 25

Several species groups were rare on the Peninsula and grew on either
the pre-Flandrian dunes, the Aromas sandstone hills, or some combina-
tion of the two sandy habitats. An airport-Fort Ord group included
Ceanothus dentatus, Arctostaphylos pumila, Ericameria ericoides, and
Corethogyne filaginifolia. A more widespread airport-Fort Ord-Prunedale
group had Lotus scoparius, Ericameria fasciculata, Helianthemum sco-
parium, and Horkelia cuneata.

A group of three associated taxa occurred in both the Peninsula and
sandhill habitats: Arctostaphylos tomentosa ssp. tomentosa, Lomatium
parvifolium, and Galium californica ssp. californica.

One species group which appeared in a variety of localities but was
never dominant was Toxicodendron diversilobum, Baccharis pilularis
var. consanguinea, Rhamnus californica, and Pteridium aquilinum.

Dudleya lanceloata, Pellaca mucronata, and Selaginella bigelovit were
too rare to show as a species group on the association table, but were
scattered on rock outcrops.

Of eight well defined endemic herb taxa in the Monterey area only
Cordylanthus littoralis seems to have been associated with the maritime
chaparral localities. This herb was collected in the past in many mari-
time chaparral localities. Now only one stand of typical material remains
near the airport, and it is considered endangered (Dept. Int., 1976).
From historical sources it appears that Cordylanthus littoralis often grew
with or at least near Ericameria fasciculata in sandy habitats.

SUCCESSIONAL TRENDS

Cooper (1922) felt that succession on the Fort Ord sandhills would
lead from scrub, to Adenostoma-Arctostaphylos pumila chaparral, then
to A. tomentosa chaparral, and finally to a “climax” live oak forest.
Gordon (1974) also viewed the “potential natural vegetation” of these
sandhills as live oak forest. On the dunes of the Peninsula, Cooper (1922,
1967) thought that pine would be important in the climax forest, but
McBride and Stone (1976) concluded that with the present degree of
fire protection live oak forest would replace the pine here.

I agree that live oak forest would eventually dominate much of the
sandhill landscape, but the poorly developed soils on the ridges will have
chaparral patches for a long time. Both the VTM survey and Davis
(1972) suggested that chaparral is often climax on south-facing slopes
and ridges. The ridgetops where old-growth Adenostoma grows less than
1 m tall are unlikely to support live oak forest under the present climate.

We have few historical details on relative distribution of live oak
forest and brushland. However, a comment by an early naturalist sug-
gests that chaparral and scrub have long been obvious on the sandhills.
In 1792 Archibald Menzies observed that the hills behind the Fort Ord
area had “‘. .. Clumps of Trees thinly scattered of the Holly-leavd (sic)

1978] GRIFFIN: MARITIME SHRUBS 77

Oak .. . but the greatest part of the Country here was covered with stiff
low Shrubs, many of them evergreen . . . many of these shrubby Plants
... were of a fragrant quality” (Eastwood, 1924). Stiff low shrubs, many
evergreen, many aromatic is a good description of the maritime chapar-
ral-scrub mosaic on the ridges. Many of the presently living Adenostoma
and Arctostaphylos tomentosa burls probably were growing on these
sandhills when Menzies visited.

On the Peninsula, maritime chaparral seems to grow where unfavorable
soils in conjunction with disturbances such as fire, bark beetle and
dwarf-mistletoe attack, or wind throw have kept the pine and cypress
canopies open for long periods. One of the few historical observations on
chaparral within this Peninsula forest was by Hartweg (1848) who men-
tioned a “thick brushwood” of Arctostaphylos, Ceanothus (including C.
rigidus), and Chrysolepis near the P. muricata-C. goveniana dwarfed
woodland.

This dwarfed conifer woodland at the Morse Botanical Reserve (Grif-
fin, 1972) represents the extreme in unfavorable soils on the Peninsula.
Here the Narlon clay-pan closely resembles the aborigine clay-pan of the
Mendocino County coastal pygmy forest (Jenny, Arkley, and Schultz,
1969). The marine terrace and dune deposits have produced the most
mature soils on the Peninsula, and both live oak and P. radiata forests
are excluded from these podsolized soils. Maritime chaparral species can
tolerate the poor drainage and sterile, acid conditions.

When either maritime chaparral or live oak forest is disturbed, scrub
species such as Baccharis pilularis and Ericameria ericoides invade.
Short-lived chaparral species such as Ceanothus dentatus and C. rigidus
may also increase. In time, manzanita sprouts and seedlings will replace
much of the scrub and ceanothus. Salvia mellifera seems to persist in
chaparral longer than other scrub species, particularly in Adenostoma
stands. Although A. hookeri and A. pumila are often thought of as low
mound formers, they can exceed 2.5 m and may survive in mixed chap-
arral for long periods. However, they do not have the height potential of
A, tomentosa, A. montereyensis, or A. pajaroensis. Davis (1972) empha-
sized the role of tall manzanitas in Prunedale hills chaparral succession.

Near San Luis Obispo, Wells (1962) noted that fine textured soils on
shale favored grassland, whereas either deep sand or rockland favored
chaparral and live oak forest. He emphasized that these substrate trends
were reinforced by frequent fires. The Monterey Bay situation is com-
patible with this theme.

CONCLUSIONS AND RECOMMENDATIONS
Maritime chaparral consists of variable sclerophyll shrub communities
within a scrub-live oak forest region that is best developed on sandy
soils within the summer fog zone. This chaparral is frequently dominated

78 MADRONO [Vol. 25

by forms of Arctostaphylos tomentosa plus one or more of four endemic
manzanita taxa. Adenostoma fasciculatum is a common sub-dominant.
These shrubs of the sandhill brushland also form a conspicuous part of
the understory of the closed-cone conifer forest on some marine terraces
and rocklands on the Peninsula. There are close floristic ties between
the chaparral on the sandhills and the closed-cone forest understory
chaparral on the Peninsula.

Maritime chaparral on the airport and Fort Ord sandhills receives sub-
stantially less rainfall than the closed-cone forests on the Peninsula. The
water relations of sclerophyll shrubs on the sandhills should be compared
with those on clay-pan or rockland habitats within the closed-cone coni-
fer forest as well as with the conventional chaparral farther inland out of
the fog zone.

Many aspects of the Monterey Bay maritime chaparral resemble those
of coastal lowlands to the south. Near Morro Bay and the Nipomo Mesa,
San Luis Obispo County and the Point Sal, Orcutt, Burton Mesa areas,
Santa Barbara County are similar sandy soils, foggy climate, and con-
centrations of endemic manzanitas within scrub-live oak forest remotely
connected with closed-cone pine forest. The vegetative and floristic ho-
mologies between Monterey Bay and southern coastal lowlands should
be specifically studied while some intact landscape samples remain in
both regions.

Pressures for development are so great around Monterey Bay that
maritime chaparral stands need legal protection to survive. For example,
areas at the airport were advertised for sale as industrial sites, and
one area was developed during the sampling period. One Prunedale sam-
ple area was bulldozed bare after sampling, and another is within a
rapidly expanding county dump.

The opportunity for new chaparral “‘reserves” is so limited that maxi-
mum support should be given to present areas with any sort of adminis- —
trative protection. The Gibson Creek annex (Point Lobos State Reserve) |
and the S.F.B. Morse Botanical Reserve (Pebble Beach Corporation)
protect samples associated with dwarfed closed-cone conifers. The fate |
of maritime chaparral on the Monterey Presidio (U.S. Army) is not |
known. Human impact at Veterans Memorial Park (City of Monterey) |
is probably too great to maintain natural chaparral. Development at the |
airport and surrounding industrial park is so extensive that chaparral |
can not survive. Despite an increasing level of military activity, Fort
Ord (U.S. Army) provides the best opportunity for maintaining mari- |
time chaparral communities on the sandhills (Griffin, 1976). Toro Re- |
gional Park (Monterey County) protects some relatively interior sam- —
ples. No adequate sample of chaparral near Prunedale has formal protec- |
tion now. The few scraps of maritime chaparral near Larkin Valley also |
have no protection.

1978] GRIFFIN: MARITIME SHRUBS 79

ACKNOWLEDGMENTS

Point Lobos State Reserve, Pebble Beach Corporation, Fort Ord Complex, and
Monterey County Parks Department kindly provided access to important stands.
Dr. Lowell E. Urbatsch, Louisiana State University, furnished unpublished data on
Ericameria; Dr. Robert J. Renard, U.S. Naval Postgraduate School, Monterey,
furnished unpublished climatic data. Appreciation is extended to Dr. Dean W.
Taylor for help in analyzing the stand data and to many colleagues for encourage-
ment and review comments.

LITERATURE CITED

ALLEN, J. E. 1946. Geology of San Juan Bautista SMITE ts, California. California
Div. Mines Bull. 133. 75 pp.

Bowen, O. E. 1966. Stratigraphy, structure and oil possibilities in Monterey and
Salina quadrangles, California. In Symp. Pac. Sec. Amer. Assoc. Pet. Geolo-
gists, Bakersfield, Calif., 1965. pp 48-67.

Cannon, W. A. 1913. A note on chaparral-forest relations at Carmel California.
Plant World 16:36-38.

Ceska, A. and H. Roemer. 1971. A computer program for identifying species-
relevé groups in vegetation studies. Vegetatio 23:255-277.

CoLWwELL, W. C. 1977. The status of vegetation mapping in California today. In M.
G. Barbour and J. Major, [eds.], Terrestrial vegetation of California. Wiley-
Interscience, New York.

Cooper, W. S. 1922. The broad-sclerophyll vegetation of California. Carnegie Inst.
Wash. Publ. 319.

. 1967. Coastal dunes of California. Geol. Soc. Amer. Memoirs 104.

CRITCHFIELD, W. B. 1971. Profiles of California vegetation. USDA Forest Service
Research Paper PSW-76.

Davis, C. B. 1972. Comparative ecology of six members of the Arctostaphylos ander-
soni complex. Ph.D. Thesis, Univ. California, Davis.

DEPARTMENT OF INTERIOR. 1976. Endangered and Heute ica species. Federal Reg-
ister 41(117) :24524—-24572.

DurfiELp, J. M. 1951. Interrelationships of the California closed-cone pines with
special reference to Pinus muricata D. Don. Ph.D. Thesis, Univ. California,
Berkeley.

Dunninc, D. 1916. A working plan for the Del Monte Forest of the Pacific Improve-
ment Company. M.S. Thesis, Univ. California, Berkeley.

Eastwoonp, A..1924. Menzies’ California journal. Calif. Hist. Soc. Quart. 2:265-340.

ForpveE, M. B. 1966. Pinus radiata in California. New Zealand J. Forestry 11:2-42.

GANKIN, R. 1966. Some observations on the growth habit of certain manzanitas.

Leaflets W. Bot. 19:245-250.
1971. Nomenclature and interpretation of a California subspecies in
Arctostaphylos (Ericaceae). Madrono 21:147-148.

Gorpon, B. L. 1974. Monterey Bay area, natural history and cultural imprints. Box-
wood Press, Pacific Grove.

GRIFFIN, J. R. 1972. What’s so special about Huckleberry Hill on the Monterey
Peninsula— Ix B. F. Howitt, [compiler], Forest heritage, a natural history of
the Del Monte Forest. Calif. Native Plant Soc.

. 1976. Native plant reserves at Fort Ord. Fremontia 4(2) :25-28.

Hart, E. W. 1966. Mines and mineral resources of Monterey County, California.
Calif. Div. Mines and Geology, County Report 5.

Hartwec, T. 1848. Journal of a mission to California in search of plants, Part 4.
J. Hort. Soc. London 3:217-227.

80 MADRONO [Vol. 25

Heapy, H. F. et al. 1977. North coastal prairie and northern coastal scrub. In M. G.
Barbour and J. Major, [eds.], Terrestrial vegetation of California. Wiley-
Interscience, New York.

HitpretH, W. R. 1976. A new cultivar: Arctostaphylos X ‘sunset’. Fremontia
3(4) :26.

Hoover, R. F. 1964. A new manzanita from Monterey. Leaflets W. Bot. 10:87-90.

. 1970. The vascular plants of San Luis Obispo County, California. Univ.
California Press, Berkeley.

————., and J. B. Roor. 1966. Two new shrubs from San Luis Obispo County,
California. Four Seasons 2(1) :1-4.

Howe tt, J. T. 1952. The Little Sur manzanita, a new species. Leaflets W. Bot.
6:202-204.

Howitt, B. F. and J. T. Howetv. 1964. The vascular plants of Monterey County,
California. Wasmann J. Bio. 22:1—-184.

. 1973. Supplement to the vascular plants of Monterey County, California.
Pacific Grove Museum of Natural History Association, Pacific Grove.

Jennincs, C. W. and R. G. StraAnp. 1958. Geologic map of California, Santa Cruz
sheet. California Div. Mines.

Jenny, H., R. J. ARKLEY and A. M. ScHuttz. 1969. The pygmy forest-podsol eco-
system and its dune associates of the Mendocino coast. Madrono 20:60—74.
Linpsay, A. D. 1932. Report of Monterey pine in its native habitats. Common-

wealth Forestry Bur., Canberra. Bull. 10.

McBripez, J. R. and E. C. STone. 1976. Plant succession on the sand dunes of the
Monterey Peninsula, California. Amer. Mid. Naturalist 96:118-132.

McDonatp, J. B. 1959. An ecological study of Monterey pine in Monterey County,
California. M.S. Thesis, Univ. California, Berkeley.

McMirtan, C. 1956. The edaphic restriction of Cupressus and Pinus in the Coast
Ranges of central California. Ecol. Monogr. 26:177-212.

McMiny, H. E. 1961. An illustrated manual of California shrubs. Univ. California
Press, Berkeley.

Mooney, H. A. 1977. Southern coastal scrub. In M. G. Barbour and J. Major, [eds.],
Terrestrial vegetation of California. Wiley-Interscience, New York.

Mutter-Domeors, D. and H. ELrenperc. 1974. Aims and methods of vegetation
ecology. John Wiley and Sons, New York.

Noss, M. A. 1957. Experimental studies in species relationships in the genus Ceano-
thus, section Cerastes (Rhamnaceae). Ph.D. Thesis, Univ. California, Berkeley.

Powe Lt, W. R. 1974. Inventory of rare and endangered vascular plants of Cali-
fornia. Calif. Native Plant Soc. Special Publ. 1.

Roor, J. B. 1961. The manzanitas of California’s Point Sur region. Leaflets W. Bot.
9:188-196.

. 1964. The nature and distribution of Arctostaphylos rosei Eastwood.
Four Seasons 1(1):1-15.

Ross, D. C. 1977. Modal and chemical data patterns, granitic rocks, Santa Lucia
Range, Salinian block, California. U.S. Geologic Survey. MF-799.

SAWYER, J. O., D. A. THorNBURGH and J. R. Grirrin. 1977. Mixed evergreen forest.
In M. G. Barbour and J. Major, [eds.], Terrestrial vegetation of California.
Wiley-Interscience, New York.

SHREVE, F. 1927. The vegetation of a coastal mountain range. Ecology 8:27-44.

SmitH, C. F. 1976. A flora of the Santa Barbara region, California. Santa Barbara
Museum of Natural History, Santa Barbara.

Sot CONSERVATION SERVICE. 1975. Soil survey of Monterey County, California: an
interim, unedited report. Salinas, California.

STorIE, R. E. et al. 1944. Soil survey of the Santa Cruz area, California. USDA.
Bureau of Plant Industry. Series 1935, No. 25.

1978] CANNE: GALINSOGA 81

SMITHSONIAN INSTITUTION. 1974. Report on endangered and threatened plant species
of the United States.

Tuomas, J. H. 1961. Flora of the Santa Cruz Mountains of California. Stanford
Univ. Press, Stanford.

Voct, R. J. et al. 1977. The closed-cone pines and cypresses. Jn M. G. Barbour and
J. Major, [eds.], Terrestrial vegetation of California. Wiley-Interscience, New
York.

WELLS, P. V. 1962. Vegetation in relation to geological substratum and fire in the
San Luis Obispo quadrangle, California. Ecol. Monogr. 32:79-103.

. 1968. New taxa, combinations, and chromosome numbers in Arcto-
staphylos (Ericaceae). Madrono 19:193-210.

Wo tr, C. B. 1948 Taxonomic and distribution studies of the New World cypresses,
El Aliso 1:1—250.

CIRCUMSCRIPTION AND GENERIC RELATIONSHIPS OF
GALINSOGA (COMPOSITAE: HELIANTHEAE)

JupitH M. CANNE
Department of Botany & Genetics
University of Guelph, Guelph, Ontario, Canada NIG 2W1

In 1894, B. L. Robinson commented that (p. 325): ““Few genera have
been subject to so much doubt as to proper limitations as Galinsoga. Sat-
isfactory generic limitations can perhaps only be obtained by a mono-
graph including not only the plants hitherto ascribed to Galinsoga, but
several neighboring genera...”

During a recent study of the specific limits and relationships in Galin-
soga, it became apparent that Robinson had indeed assessed the situation
correctly. The genus has been known only superficially and therefore,
has not been well defined from a number of closely allied genera in the
subtribe Galinsoginae. This circumstance has prevailed even though
related genera have been monographed recently: Sabazia Cass., Selloa
H.B.K., and Tricarpha Longpre (Longpre, 1970), Tridax L. (Powell,
1965), and Stenocarpha S. F. Blake (Turner, 1965).

The first objective of this paper is to consider the circumscription of
Galinsoga. 1 believe that the taxonomic changes discussed here, based on
morphological, chromosomal, and geographical criteria, reflect the close
phenetic and inferred phylogenetic relationships of the taxa involved. A
second objective is to discuss and re-evaluate the relationships of Galin-
soga with closely allied genera in the Galinsoginae.

HISTORY OF THE GENERIC CONCEPT OF Gdlinsoga S. STR.
When Ruiz and Pavon (1794) first described Galinsoga (s. str.), they
emphasized the following characters as diagnostic: plants annual, herba-

_ Ceous, with a paleaceous receptacle, pappus of ciliate scales, and ray

82 MADRONO [Vol. 25

florets 4—6, with corollas trifid. Later authors have characterized the
genus by a combination of the above features plus the following char-
acteristics (Candolle, 1836; Bentham and Hooker, 1873; Hoffmann,
1890): heads small (to 8 mm diam.), on short peduncles (to 4 cm long) ;
involucre 2-seriate; and achenes black and somewhat angular. Both Can-
dolle (1836) and Bentham (1844) also noted a characteristic typical of
most Galinsogas: each ray floret is enclosed by a phyllary and two or
three pales. This aggregation falls from the receptacle as a unit at
maturity.

As additional species of Galinsoga were described in the latter half of
the 19th century and into the present decade, the genus came to include
species with eligulate ray florets (G. eligulata Cuatrecasas, 1954) and, in
the case of some specimens of G. subdiscoidea Cronquist (1965), even
discoid plants. However, the generic concept of Galinsoga was not sig-
nificantly altered until the publication by McVaugh (1972) of Galinsoga
mollis. Plants of this species differed notably from previously recognized
Galinsogas in: height to 1.5 m; elongate internodes; heads to 2 cm
diam.; large number of disc florets (125-150, compared with 10-60 in
other species of Galinsoga); ray florets 8 rather than 5 per head; and
disc and ray corollas twice the length of those in other species. Despite
these differences, G. mollis possessed many of the diagnostic features of
the genus mentioned earlier and so was related to the other Galinsogas as
a somewhat anomalous species.

In a recent revisionary study of Galinsoga (Canne, 1977b), several
newly described species were recognized that bridge the morphological
gap between the small-headed, short-statured Galinsogas and G. mollis.
Three species, G. triradiata, G. longipes and G. elata, possess the follow-
ing characteristics that approach the condition found in G. mollis: (1)
heads to 1.5—2.0 cm diam; (2) heads many to few on elongate peduncles
in open, cymose clusters; (3) phyllaries with scarious margins; (4)
plants to 55-90 cm tall with elongate internodes; (5) leaves elliptic-

lanceolate to ovate-lanceolate with serrate to remotely denticulate mar-

gins; and (6) ligules tridenticulate to deeply divided, to 8 in number,
and 8 mm in length.
As revised, Galinsoga includes 14 species in 3 sections. The fact that 6

of these species are described as new is a reflection of an absence of pre- |

vious taxonomic study. The transfer of 3 species to Galinsoga from 3

other genera is an indication, as well, that generic concepts have been |
inadequately defined among these genera of the Galinsoginae. While I —
believe a morphological examination of the new species and combinations |

leaves no doubt that they are indeed Galinsoga, their inclusion does
broaden the limits of the genus somewhat and brings Galinsoga closer to

other members of the Galinsoginae, most notably Sabazia, with regard to |

certain characteristics.

Galinsoga and the genera that are morphologically, chromosomal |

|

1978] CANNE: GALINSOGA 83

and distributionally most similar to it form a rather close-knit unit. Mor-
phological features which in general are characteristic of a given genus
occur regularly in some members of other genera. However, in my view,
the rather confusing mosaic of character state distributions among genera
does not severely alter the integrity of most generic boundaries. Steno-
carpha and Tricarpha, with one and two species respectively, are excep-
tions. Table 1 provides a comparison of the differences and similarities
among the genera discussed here. A key to these taxa is presented below.

1. Annual herbs with slender stems; heads solitary or in cymose
clusters... Z
2. Heads solitary: or Jecastenvilly : in gnarias. Haplleg Bilabiaee or

if absent then disc corollas eae pappus of plumose

bristles. . . . Wee eta
2. Heads in dense to ipeee Smee or cance ames occasionally

solitary in Sabazia; ligules rarely bilabiate, sometimes absent or

disc-like; pappus of scales, rarely of setose bristles, or absent . 3.

3. Heads in cymose panicles; disc corollas white . Cymophora

3. Heads in dense to loose cymes or solitary; disc corollas se
low to occasionally reddish-purple . . . . 4,

4, Plants erect and single-stemmed; heads in spies peachy a
floret usually enclosed within a wenilleiay. pale association;
involucre 1-2 seriate of ovate phyllaries, and a pappus of
obtuse to aristate scales or absent; or involucre 2—3 seriate
and a pappus of setose bristles . . . . . . Galinsoga

4. Plants decumbent to ascending and multistemmed or much
branched near the base; ray florets usually not enclosed
within a phyllary-pale association; involucre 2-3 seriate
and a pappus of obtuse scales; or stems erect, branched or
not; involucre 3—5 seriate and pappus of delicate, linear
eles. ioe ark te Sabazia

1. Shrubs or perennial feabe sath en om Anceened rootstocks or
rhizomes; heads solitary, in loose cymes, or in corymbose or um-
bellate clusters. 5.

5. Herbs: heads radiate: phyflaries parabolical: Ace corol-

las 1.5-4(-5) mm eae disc styles branches 0.3-1.2 mm
long with acute or rhomboid tips . . . 6.

6. Leaves cauline and basal; pales of beret Briedes to

0.2 mm wide; style tips enanabord nee COG

6. Leaves cauline; pales of linear or lanceolate scales

more than 0.2 mm wide; style tips acute . Sabazia

5. Shrubs and herbs; heads radiate or discoid; phyllaries

linear-lanceolate to oblong or obovate; disc corollas

4-10 mm long; disc style branches 1-3 mm long with
obtuse, truncate or subulate tips . . ..... . 7,

84 MADRONO [Vol. 25

7. Herbs; ray corollas bilabiate; achenes densely pi-
lose, rarely glabrous; pappus of plumose bristles,
rarely of deeply fimbriate scales; style tips long
subulate’. io) ee ee eed ax,
. Shrubs and herbs; ray corollas not bilabiate;
achenes strigose or glabrous; pappus usually of
spinulose-fimbriate, linear-lanceolate scales, occa-
sionally of blunt scales or setose; style tips obtuse
to thuncate 2. 62 ae ee Caled

~I

RELATIONSHIP TO Tvicarpha

The 3 new species of Galinsoga mentioned above are remarkable in
their resemblance to Tvicarpha durangensis Longpre. Tricarpha duran-
gensis and a second species, T. purpusi (Brandg.) Longpre, were con-
sidered by Longpre to represent a distinct genus on the basis of scarious-
margined phyllaries, deeply trifid pales, and shallowly lobed ray corollas.
These characters are found not only in numerous species of Galinsoga,
in which scarious-margined phyllaries and trifid pales are common, but
some representatives of the related genus Sabazia [e.g., S. humilis
(H.B.K.) Cass. and S. multiradiata (Seaton) Longpre]| also possess
these characteristics as Urbatsch and Turner (1975) noted. In vegeta-
tive features Tricarpha cannot be distinguished from Galinsoga. The
chromosome number of Tricarpha purpusii is unknown. Tricarpha dur-
angensis was reported to be m = 8 by Keil and Stuessy (1975) and
Turner and Flyr (1966, as Sabazia microglossa DC., Flyr 292, the type
collection for T. durangensis). This is also the base chromosome number
for Galinsoga (Turner and Flyr, 1966; Canne, 1977b).

The similarity in vegetative and floral characters between Galinsoga
and T. durangensis is so great that I have transferred T. durangensis to
Galinsoga (Canne, 1977b). This transfer is contrary to that made by
Urbatsch and Turner (1975) who placed T. durangensis in Sabazia be-
cause, as noted above, a few species of Sabazia possess two of the three
diagnostic characters used as generic markers for Tricarpha. Clearly, the
generic boundaries of Sabazia and Tricarpha overlap, at least with regard
to these characters. A similar overlap exists with Tricarpha and Galin-
soga, and in the case of 7. durangensis (but not T. purpusii) the overlap
is more extensive with Galinsoga than it is with Sabazia. Tricarpha
durangensis shares with nearly all species of Galinsoga those features
that were used as generic markers for Tricarpha. In addition, this species
is a Slender-stemmed annual as is Galinsoga, but most Sabazias are either
weak-stemmed, lax perennials or firm-stemmed, erect perennials with
rhizomes or caudices. Yellow anthers and ovate, scarious-margined phy]l-
laries, characteristic of Galinsoga, are found in 7. durangensis whereas
Sabazias characteristically have red anthers and parabolical, herbaceous
phyllaries. Tricarpha pur pusii, however, differs from Galinsoga in several
features of the head. The 3-4 seriate involucre of sharply acute phyl-

1978 | CANNE: GALINSOGA 85

laries, ray ligules to 11 mm long with tubes to 4 mm long, shiny achenes,
disc corollas to 3.7 mm long and anthers to 1.8 mm long are characteris-
tics not found in Galinsoga. Tricarpha purpusit, a species known only
from the type collection (Purpus 3961) gathered in Baja California Sur
in 1901, has close affinities with Sabazia, as Brandegee (1903) originally
suggested.

RELATIONSHIP TO Stenocarpha

Stenocar pha was described in 1915 by S. F. Blake, based upon Galin-
soga filiformis Hemsley. Blake distinguished Stenocarpha from Galin-
soga by the former’s tall, narrowly conical receptacle, narrow pales,
alternating truncate and aristate pappus scales, and smaller heads. Tur-
ner (1965) concurred with Blake in maintaining Stenocarpha as distinct,
emphasizing that this genus differed from Galinsoga by having heads on
long peduncles arranged in an open capitulescence, and 8 ray florets with
narrow ligules. However, all the features mentioned above, with the
exception of the dimorphism of the pappus, occur commonly in various
species of Galinsoga (Canne, 1977b).

Plants of Stenocarpha filiformis differ most strikingly from Galinsoga
by having linear pales, relatively few cauline leaves, and usually some
basal leaves. Nevertheless, I have observed the similar formation of basal
leaves in G. quadriradiata Ruiz & Pavon grown under greenhouse condi-
tions. Conversely, a specimen of S. filiformis grown from seed at Stanford
from a Breedlove 1668 collection (DS) shows many pairs of well-formed
cauline leaves. The pales of S. filiformis are narrower than those in Galin-
soga but are approached by those of G. mollis and G. triradiata, which
reach only 1 mm in width. The character states that distinguish S. fili-
formis as a species do not warrant generic recognition as well. I agree
with Hemsley’s original decision to place this species in Galinsoga where
it is located in the monotypic section Stenocarpha. Turner and Flyr
(1966) reported » = 8 for G. filiformis while Solbrig et al. (1972)
reported » = 9, indicating the possible existence of intraspecific aneu-
ploidy.

RELATIONSHIP TO Sabazia

In his discussion of generic relationships, Longpre (1970) indicated
the following character states by which Galinsoga was thought to differ
from Sabazia: plants annual in Galinsoga vs. perennial in Sabazia; heads
3—5(—7) mm diam. vs. 6-23 mm diam.; 5 vs. 8-17 ray florets per head;
and ray florets enclosed by a phyllary-pale association in Galinsoga but
not in Sabazia.

All species of Galinsoga are considered to be annuals (although pos-
sibly perennial in G. mollis) but more than three-fourths of the species
of Sabazia are perennial. The large overlap in capitulum size (including
rays) between Galinsoga (3-20 mm diam.) and Sabazia [6-25(-—35) mm
diam.] makes this character of limited use for generic differentiation.

The ray floret number was first noted as a variable character in
Galinsoga by Ruiz and Pavon (1798) when only two species were known

86

TABLE 1. COMPARISON OF SELECTED CHARACTERISTICS OF Galinsoga, Sabazia,
Selloa, Tridax, Cymphora anp Calea ILLUSTRATING DIFFERENCES AND SIMILARITIES

MADRONO

IN MorPHOLOGY AND BASE CHROMOSOME NUMBER.

GALINSOGA
Annual herbs; stem
erect, rarely decum-
bent, from a slender
taproot

Heads in few to
many-headed cy-
mose clusters, never
solitary

Involucre 1-2 (—3)
seriate ; inner phyl-
laries ovate, usually
with 2-3 pales at-
tached at the base
and enclosing a ray
floret

Ligules usually 5 or 8,
quadrate to obo-
vate or less often
rectangular, occa-
sionally bilabiate in
one species

Disc corollas yellow,
1.0-3.2 mm long

Anthers yellow or less
often pale brown,
rarely red; apical
appendage with
glands in two
species

Disc style branches
0.2—0.6 mm long,
tips acute

Pappus of broad, ob-
tuse or aristate,
fimbriate or ciliate
scales, of setose
bristles in one
species, or absent

Base chromosome
number, x = 8

SABAZIA

Perennial or less often
annual herbs; stem
erect, decumbent or
procumbent, from
a taproot or oftena
short rootstock or
rhizome

Heads solitary or in
few-headed cymose
clusters

Involucre (1—) 2-5
seriate ; phyllaries
parabolical, less-
often ovate, usually
free from adjacent
pales

‘Ligules usually 8-16,

less often 5, rectan-
gular, sometimes
quadrate-obovate,
not bilabiate

Disc corollas yellow,
2-5 mm long

Anthers red, rarely
yellow, aglandular

Disc style branches
0.4-1.2 mm long,
tips acute

Pappus usually of
spinulose-fimbriate,
linear-lanceolate
scales or of obtuse,
fimbriate scales,
occasionally absent

Base chromosome
number, x = 8 (4?)

[Vol. 25

SELLOA

Perennial herbs; stem

erect or often de-

cumbent or ascend-

ing, from a short
rootstock

Heads solitary or in

few-headed subcymes

Involucre 2-3 seriate ;
phyllaries paraboli-

cal and free from
pales

Ligules 6-19, rectan-

gular, quadrate-

obovate in one spe-

cies, not bilabiate

Disc corollas yellow
or greenish-yellow,

1.5-4.8 mm long

Anthers reddish,
aglandular

Disc style branches
0.3-1.2 mm long,
tips abruptly
rhomboid

Pappus of setose
bristles or absent

Base chromosome
number, x = 8

1978]

TRIDAX

Annual and perennial
herbs,sometimes
woody near the
base; stem procum-
bent, decumbent, or
erect, from a tap-
root, sometimes
with woody thick-
enings, or rhizo-
matous

Heads solitary or in
few-headed subcymes

Involucre 2-5 seriate ;
phyllaries oblong to
lanceolate or ovate,
with attached pales
in one species

Ligules 0, 3-8(—13),
usually obovate to
suborbicular, or
broadly oblong,
bilabiate

Disc corollas yellow
or greenish-yellow,
(3—) 4-10 mm long

Anthers brown to red,
apical appendage
usually with glands

Disc style branches
1-3 mm long, tips
subulate

Pappus usually of
plumose bristles,
less often of deeply
fimbriate scales,
rarely absent

Base chromosome
number, x = 9, 10

CANNE: GALINSOGA

CYMOPHORA

Annual herbs; stem

erect, from a tap-
root.

Heads in several- to
many -headed cy-
mose panicles

Involucre 1-2 (-3)
seriate ; phyllaries
oblong to ovate and
free from pales or
pales absent

Ligules 0-5, obovate,
bilabiate

Disc corollas white,
2.8-3.5 mm long

Anthers pale brown
to red, apical ap-
pendage with
glands in one species

Disc style branches
0.6-1.2 mm long,
tips subulate

Pappus of obtuse
fimbriate or plu-
mose-fimbriate
scales or absent

Base chromosome
number, x = 8

87

CALEA
Shrubs, rarely peren-
nial herbs; stem erect,
less often prostrate or
climbing

Heads solitary or in few
to many-headed cy-
mose, corymbose or
umbellate clusters

Involucre 3—multi-
seriate, phyllaries ob-
long, oblanceolate,
obovate and free
from pales

Ligules 0-8, 10-25, ob-
long-rectangular to
obovate, not bilabiate

Disc corollas yellow,
orange or white.
ca. 4-7 mm long

Anthers yellow to red-
dish-brown; of the
taxa seen, none with
glands as in Tridax

Disc style branches
1-2 mm long, tips
obtuse or truncate

Pappus usually of
spinulose-fimbriate,
linear-lanceolate
scales

Base chromosome num-
ber, x = 8 and/or 9?

88 MADRONO [Vol. 25

in the genus. In the more common weedy species (e.g., G. parviflora
Cav., G. quadriradiata, and G. mandonii Sch.-Bip.) ray number varies
from 3 to 10, with 5 the usual number, and the characteristic number for
many species. Galinsoga filiformis, G. durangensis, G. mollis, and G.
elata are exceptions and have 6—12 ray florets per head with 8 the usual
number. Eight or more ray florets per head is common in Sabazia but
S. micros permoides Longpre, S. humilis, S. acoma (S. F. Blake) Longpre,
S. sarmentosa Less., S. liebhmanni Klatt (Longpre, 1970), and S. tri-
dacioides Urbatsch & Turner (1975) sometimes have 5 ray florets per
head. In general, then, Galinsoga has 5-8 ray florets per head whereas
in Sabazia there are usually 8 or more.

The phyllary-pale enclosure of the ray florets in Galinsoga is a trait
that has been used by a number of authors (Candolle, 1836; Bentham,
1844; Longpre, 1970; and Urbatsch and Turner, 1975) to distinguish
Galinsoga. McVaugh (1972) has been the only author to point out that
this feature also occurs in Tridax (e.g., T. dubia Rose). I have observed
it in Sabazia sarmentosa, S. pinetorum S. F. Blake, and S. trianae (Hier-
on.) Longpre. It is, however, a fairly consistent feature in Galinsoga
except for its absence in G. glandulosa, G. filiformis, and G. durangensis,
and its weak occurrence in G. subdiscoidea and some representatives of
G. mandonti. Thus, the phyllary-pale association is characteristic of
Galinsoga in general, but does not occur in all species and is not restricted
to Galinsoga.

The morphological closeness of Galinsoga and Sabazia is emphasized
further by the morphology of Sabazia trifida Fay. The annual S. trifida is
of particular interest because in overall vegetative and floral morphology
it is much like G. durangensis, G. elata and G. mollis, differing from
them significantly only by having a setose pappus. Accordingly I have
transferred to this species to Galinsoga (Canne, 1977b).

Little doubt exists that Sabazia is the genus most closely related to
Galinsoga. The genera are alike in chromosome number, distribution and
habitat, and they share a number of morphological features (Table 1).
Characters typical of one genus are found sporadically in the other, but
when combinations of characters are considered the two genera can be
recognized as separate entities. Until additional data can indicate a more
taxonomically sound alternative, it is best to note the morphological
overlap that exists and maintain the genera as distinct.

McVaugh (1945), Gillis (1971) and Grashoff (1975) have noted that
homogeneity within a genus is a most important consideration, even if
in the present situation by including in Galinsoga the new species men-
overlap does occur with other genera. This ‘“‘dictum” is best adhered to
tioned previously and Tricarpha durangensis, Stenocarpha filiformis, and
Sabazia trifida. These taxa are clearly, on morphological grounds, most
closely related to Galinsoga, but within that genus are conveniently rec-

1978] CANNE: GALINSOGA 89

ognized as units at the sectional level. Tvicarpha durangensis, Sabazia
trifida, Galinsoga elata and G. mollis are recognized in section Elata.
Section Stenocarpha accommodates the transfer of Stenocarpha filiformis,
and the remaining nine species are placed in section Galinsoga (see Table
1 for a comparison of Galinsoga and Sabazia).

RELATIONSHIP TO Selloa

Selloa, in terms of overall morphology (particularly shape and size of
the ray corollas, number of heads per plant, decumbent stems, and para-
bolic phyllaries) must be considered closely related to Sabazia, as Long-
pre (1970) maintained. The relationship of Selloa to Galinsoga seems to
be through Sabazia rather than directly. Selloa can be best distinguished
from Galinsoga by the former’s elliptic-obovate leaves, setose pales, and
rhomboid style tips. Additional differences are listed in Table 1.

RELATIONSHIP TO Tridax

Powell (1965) noted in his revision of Tridax that Galinsoga is prob-
ably the genus most closely related to Tridax. Tridax comes closest to
Galinsoga through T. dubia Rose. This species has the habit, head size,
ciliate pappus scales, and phyllary-pale association similar to Galinsoga
but T. dubia differs conspicuously from all Galinsogas by its orange or
golden-yellow ray corollas, brownish to rose-color of the pappus, large
disc corollas (5-6 mm long), densely pilose ray achenes, and = 9 chro-
mosome number.

Galinsoga glandulosa Canne most closely resembles Tvidax. The glands
of the anther appendages noted by King and Robinson (1970) to occur
in many Tridax are also presented in this Galinsoga. The phyllaries of G.
glandulosa are intermediate in shape between those of Tvidax and Galin-
soga and more densely pilose than in other species of Galinsoga. The
tubes of the ray corollas are longer than in other Galinsogas and much
like those in Tridax. Galinsoga differs from Tridax by numerous charac-
teristics (listed in Table 1).

RELATIONSHIP TO Cymophora

Cymophora was established by Rebinson (1907) to accommodate the
single species C. pringleiz. Anderson and Beaman (1968) later submerged
Cymo phora into Tridax, noting its similarities with T. accedens, a species
in which I have a special interest because of its likeness to Galinsoga.
Recently Turner and Powell (1977) have advocated the transfer of
I’. accedens to Cymophora. I concur with these authors that 7. accedens,
C. pringlei, and a species newly described by them (C. hintonii) form a
coherent generic unit, but feel that Tvidax venezuelensis Aristequieta &
Cuatrecasas belongs in Cymophora as well (Canne, 1977a). The zygo-
morphic outer disc corollas, white corolla color, and the paniculate-

90 MADRONO [Vol. 25

cymose capitulescence differentiate Cymophora as a well-defined genus.
On the basis of morphological similarity, Cymophora stands near both
Galinsoga and Tridax. The n = 8 chromosome number of C. pringlei
(Turner et al., 1973) suggests a somewhat closer tie with Galinsoga
(x = 8) than with Tridax (x = 9, 10), as noted by Turner and Powell.
That Cymophora and Galinsoga are erect, small-headed annuals whereas
most Tridax are decumbent, larger-headed perennials supports the view
that the former genera are more closely related to each other than to
Tridax.

RELATIONSHIP TO Calea

The poorly known genus Calea L. was revised in part by Robinson and
Greenman (1896). At that time the genus was estimated to contain
about 85 species, but numerous species have been described since then
and 110-120 species is probably now a more accurate estimate.

Calea seems more distantly related to Galinsoga than the genera dis-
cussed previously and differs from Galinsoga in a number of ways (listed
in Table 1). The two genera also differ in shape of achenes, pubescence
and lobing of the disc corollas, and shape of corolla tubes. The shrubby,
small- and many-headed Caleas seem most closely related to Galinsoga,
while the large- and solitary-headed species with truncate style branches
(e.g., C. pennellii S. F. Blake, C. monocephala Dusen, C. lucidivenia
Gleason & S. F. Blake) may deserve generic ranking apart from Calea.
The base chromosome number for Calea is not readily apparent from the
reported counts of nm —16, 17, 18, 19, 24, 32 (Federov, 1969; Moore,
1973, 1974) and this also is perhaps an indication of the heterogeneity
of the genus. A clear view of how Calea relates to Galinsoga and other
genera of the Galinsoginae must await a comprehensive systematic study
of the former genus.

SUMMARY

A diagramatic interpretation of generic relationships as discussed
above is shown in Fig. 1. Stenocarpha, Sabazia trifida, and Tricarpha
durangensis have been transferred to Galinsoga. Similarities between
Galinsoga (particularly sect. Elata) and Sabazia are numerous, but mor-
phological overlap is insufficient to demand merger of the two genera.

Selloa seems closest to Sabazia but is rather clearly defined. It is
placed in the diagram near Sabazia, Galinsoga, and Calea since two spe-
cies, [S. obtusata (S.. F. Blake) Longpre and S. breviligulata Longpre}],
resemble Galinsoga and Sabazia while the third species, (S. plantaginea
H.B.K.), resemble those species of Sabazia that are closest to Calea.

For the most part, the boundary between Galinsoga and Tridax is a
clear one, although the similarities of T. dubia to Galinsoga and the Tr1-
dax-like features of Galinsoga glandulosa emphasize the closeness of the
two genera.

1978] CANNE: GALINSOGA 91

Selloac3) bd Sabazia
(14 Jeatinsoss
Cymophora

Fic. 1. Relationships among Galinsoga, Sabazia, Selloa, Tridax, Cymophora, and
Calea. Numbers and relative sizes of ellipses designate present number of species
per genus. Relative placement of ellipses infers closeness of relationship. Overlap
indicates uncertainty of generic boundaries.

The relationship between Trvidax and Sabazia seems more distant than
that between these genera and Galinsoga. Several species of Tridax and
Sabazia possess large solitary heads and are decumbent in habit but
otherwise differ in most vegetative and floral characteristics.

The relationship of Tridax and Calea is unclear. Tridax is more dis-
tinct from Calea in total morphology than is Sabazia. However, similari-
ties in pale and phyllary shape among some species suggest a not too
distant relationship.

Urbatsch and Turner (1975) discussed the relationship of Sabazia
and Calea and noted that Caleas such as C. caracasana O. Ktz., C. integ-
rifolia Hemsl., C. scabra Robins., and C. colimensis McVaugh resemble

92 MADRONO [Vol. 25

Sabazia. The uncertainty of the Calea-Sabazia relationship is indicated
in Fig. 1 by the overlapping and broken boundaries of the two taxa.

Finally, the small genus Cymphora is similar to Tridax and Galinsoga
and so is positioned in Fig. 1 close to both genera. Morphology and
chromosome number suggest a somewhat closer relationship of Cymo-
phora to Galinsoga than to Tridax.

ACKNOWLEDGMENTS

This paper is based upon a portion of a dissertation submitted to the Graduate
School of The Ohio State University in partial fulfillment of the requirements for
the degree Doctor of Philosophy. Appreciation is offered to Dr. Tod F. Stuessy for
his suggestion of the project and his guidance throughout the study. Thanks are
extended to Dr. David J. Keil, Ms. Kathleen Hruschak, and Sr. Jose Schunke for
their aid on collecting trips that helped to form a basis for this study. Field work
was supported by National Science Foundation Grant GB30240. Appreciation is
offered to the curators and directors of the following herbaria for loan of speci-
mens: A, ARIZ, BM, CAS, DAV, DUKE, ENCB, F, FSU, GH, K, LD, LL, M,
MEXU, MICH, MO, MSC, NY, OAC, OS, P, TEX, UC, US, USM, WIS, Z, and ZT.

LITERATURE CITED

ANDERSON, C. E., and J. H. BEAMAN. 1968. Status of the genus Cymophora (Com-
positae). Rhodora 70:241-246.

BENTHAM, G. 1844. The botany of the voyage of H.M.S. Sulphur, pp. 119-120.
London.

, and J. D. Hooker. 1873. Galinsoga, p. 390. Genera Plantarum, Vol. 2,
Reeve and Co., London.

BLAKE, S. F. 1915. Plantarum novarum in Herbario Hortii Regii Conservatarum.
Bull. Misc. Inform. No. 7:348-349.

BRANDEGEE, T. S. 1903. Notes and new species of Lower California plants. Zoe 5:162.

CANDOLLE, A. P. DE. 1836. Galinsoga, pp. 677-678. Prodromus, Vol. 5, Paris.

CanneE, J. M. 1977a. A new combination in Cymophora (Compositae: Heliantheae:
Galinsoginae). Madrono 24:190-191.

.1977b. A revision of the genus Galinsoga (Compositae: Heliantheae). Rho-
dora 79:319-389.

Cronouist, A. 1965. Studies in Mexican Compositae. I. Miscellaneous new species.
Mem. New York Bot. Gard. 12:286-292.

Cuatrecasas, J. 1954. Notas a la flora de Colombia. XIII. Revista Acad. Colomb.
Ci. Exact. 9:233-249.

Feporov, A. A. [ed.]. 1969. Chromosome Numbers of Flowering Plants. Acad. Sci.
U.S.S.R., Leningrad.

Gittis, W. T. 1971. The systematics and ecology of poison-ivy and the poison-oaks
(Toxicodendron, Anacardiaceae). Rhodora 73:72-159, 161-237, 370-443, 465-
540.

GrasHorr, J. L. 1975. Metastevia (Compositae: Eupatorieae): A new genus from
Mexico. Brittonia 27:69-73.

HoFrrMann, O. 1890. Compositae, pp. 87-391. In A. Engler and K. Prantl [eds.], Die
Naturlichen Pflanzenfamilien, Vol. 4(5). Berlin.

Kei, D. J., and T. F. Srurssy. 1975. Chromosome counts of Compositae from the |
United States, Mexico, and Guatemala. Rhodora 77:171-195. |

Kinc, R. M., and H. Rogrnson. 1970. The new synantherology. Taxon 19:6-11.

LoncprE, E. K. 1970. The systematics of the genera Sabazia, Selloa and Tricarpha |
(Compositae). Publ. Mus. Michigan State Univ., Biol. Ser. 4(8) :283-384.

1978] MACFARLANE: FRITILLARIA 93

McVaucH, R. 1945. The genus Triodanis Rafinesque, and its relationships to Specu-
laria and Campanula, Wrightia 1:13-52.
. 1972. Compositarum Mexicanarum Pugillus. Contr. Univ. Michigan Herb.
9(4) :359-484.
Moore, R. J. [ed.]. 1973. Index to plant chromosome numbers 1967-1971. Regnum
Veg. 90:379.
. Led.]. 1974. Index to plant chromosome numbers for 1972. Regnum
Veg. 91:72.
PowE Lt, A. M. 1965. Taxonomy of Tridax (Compositae). Brittonia 17:47-96.
Rosinson, B. L. 1894. Notes upon the genus Galinsoga. Proc. Amer. Acad. Arts
29:325-327.
—. 1907. New or otherwise noteworthy spermatophytes, chiefly from Mexico.
Proc. Amer. Acad. Arts 43:21-48.
, and J. M. GrREENMAN. 1896. Revision of the Mexican and Central Ameri-
can species of the genus Calea. Proc. Amer. Acad. Arts 32:20-30.
Ruiz, H., and J. A. Pavon. 1794. Florae Peruvianae et Chilensis Prodromus, p. 110.
Madrid.
, and —————. 1798. Systema Vegetabilium Florae Peruvianae et Chilen-
sis. Vol. 1. pp. 198-199. Madrid.
Sotsric, O. T., D. W. Kyuos, A. M. Powe tt, and P. H. Raven. 1972. Chromosome
numbers in Compositae VIII: Heliantheae. Amer. J. Bot. 59:869-878.
TURNER, B. L. 1965. Taxonomy of Stenocarpha (Compositae-Heliantheae-Galin-
soginae). Southw. Naturalist 10:238-240.
, and D. Fryr. 1966. Chromosome numbers in the Compositae. X. North
American species. Amer. J. Bot. 53:24-33.
, and A. M. Powe Lt. 1977. Taxonomy of the genus Cymophora (Astera-
ceae: Heliantheae). Madrono 24:1-6.
, and T. J. Watson. 1973. Chromosome numbers in Mexican
Peteracene. Amer. J. Bot. 60:592-596.
Urpatscu, L. and B. L. Turner. 1975. New species and combinations in Sabazia
(icianthene. Galinsoginae). Brittonia 27:348-354.

ON THE TAXONOMIC STATUS OF
FRITILLARIA PHAEANTHERA EASTW. (LILIACEAE)

RoGER M. MACFARLANE
1801 Waverley, Palo Alto, California 94301

The identity and geographical distribution of Fritillaria phaeanthera
Kastw. and Fritillaria phaeanthera Purdy have been the subject of some
confusion. Several issues involved have been discussed by Furse (1969).
It is pointed out here that these two homonyms apply to different taxa,
and the question of priority is resolved by renaming the later homonym
F. phaeanthera Eastw., to which I give the name F. eastwoodiae. The
probable origin of F. eastwoodiae as a hybrid between F. recurva Benth.
and F. micrantha Heller is explored. Pollination experiments were carried
out to confirm the interfertility of the latter two species, and seeds of the
cross F. recurva (2) X F. micrantha (3) have been germinated.

94 MADRONO [Vol. 25

The confusion began when Purdy (1932) published a photograph
labelled F. phaeanthera, together with a brief diagnostic description,
following one of F. lanceolata. No type was cited, so the type of F.
phaeanthera Purdy is here designated as the photograph and original
description. Both photograph and description (“F. phaeanthera, a slen-
der related species, in which the anthers are red, and the petals rather
twisted’) are those of F. lanceolata Pursh var. gracilis Wats. (Watson,
1879) from Napa Co., California. As Furse has noted, the publication of
Purdy’s description precedes that of Eastwood (1933).

A study of Purdy’s correspondence to Eastwood, in the archives of the
California Academy of Sciences, reveals the reason for this confusing
situation. On 22 March, 1932, Purdy sent a specimen of Fritillaria from
east of Middletown (Lake Co., California) to Eastwood for identification.
A few weeks later, the type material of F. phaeanthera Eastw. was col-
lected by Mrs. J. H. Morrison near Durham (Butte Co.), and Eastwood
recognized the Durham plant as a new species. She apparently wrote to
Purdy, incorrectly identifying his plant as F. phaeanthera, an as yet un-
published name. Although Eastwood’s letter making the identification
cannot be found, its content can be inferred from Purdy’s reply of 7
May, 1932, acknowledging her identification and noting that he had a
good photograph of his plant which he intended to publish in an article
in the R.H.S. Lily Year Book (Purdy, 1932). A possible reason for East-
wood’s misidentification is that one of the diagnostic characters used by
her (indeed the source of the epithet ‘paeanthera’) is the dark red color
of the anthers, a character which is also found in F. lanceolata var.
gracilis.

The taxonomic confusion was continued by Beetle (1944) and Munz
(1959) who cite locations for F. phaeanthera Eastw. in Butte Co. (the
type locality of F. phaecanthera Eastw.) and in Napa Co. (a locality of
F.. lanceolata var. gracilis Wats.). In the text of Beetle’s monograph she
gives a further location as Plumas Co. but does not cite a specific speci-
men. However, a specimen from Quincy, Plumas Co. (Luke Gill s.n.,
DS) was annotated by Beetle as F. phaeanthera Eastw. but is F. atro-
purpurea Nutt. with linear, ascending, scattered leaves and open cam-
panulate flowers.

Eastwood’s plant (F. phaeanthera Eastw.) is a distinct species belong-
ing to a rather difficult taxonomic unit, intermediate between F. recurva
Benth. and F. micrantha Heller. It has been suggested by Beetle (1944)
that it arose as a hybrid between F. recurva and F.. micrantha. Although
there is little experimental or definitive cytological evidence to establish
this as yet, the circumstantial evidence in favor of the hypothesis is com-
pelling and will be reviewed below. Since the epithet phaeanthera was
previously used by Purdy, I propose a new name for Eastwood’s plant,
and include an expanded description. +

1978] MACFARLANE: FRITILLARIA 95

Fritillaria eastwoodiae Macfarlane = Fritillaria phaeanthera East-
wood, Leafl. West. Bot. 1:55, 1933, not F. phaeanthera Purdy, Roy.
Hort. Soc. Lily Year Book 1:97, 1932.

Perennial herb. Bulb 1.5—2.5 cm diam., a sub-conical, enlarged stem
base, surrounded by several fleshy scales and numerous rice-grain bulb-
lets. Stem 20-80 cm tall, simple, terete, glaucous, flecked with purple
near the base. Leaves on upper 24 of stem, in 1-2 whorls of 3—5 below,
scattered above, blades linear to narrowly lanceolate, 4-9 mm wide, 5—10
cm long, often glaucous. Flowers (2)3—5(7), racemose, nodding, cam-
panulate, ca 15 mm long and as broad, pale greenish-yellow, apricot, to
red. Perianth segments six, narrowly elliptic, 3-5 mm wide, 10-17 mm
long, flaring to partially recurving at the tips, sepals narrower and more
acute than petals. Nectary lanceolate, less than 13 the length of tepals.
Androecium: filaments contiguous near their base, attenuate to the tips,
stamens equalling or exceeding pistil before anther dehisence, but not
after; anthers cultrate, ca 5 mm long before dehiscence, red; pollen rust
to yellow in color. Gynoecium: style divided in three less than ™% its
length, style branches recurving somewhat; pistil often absent in upper
flowers. Capsule 6-angled, winged, truncate, ca 15 mm tall and as wide.
Chromosome numbers: 7 = 18, near Cherokee, Butte Co. RM 202536
(Beetle 1944); m = 12, 12 +f, near Magalia, Butte Co. JEPS 55690,
n= 17, 18 + f, 16 + 2f, Pinkston Canyon, Butte Co. (Cave 1970);
2n = 24 + 2f, near Magalia, Butte Co. (Marchant, pers. comm.).

Type: Near Durham, Butte Co., California, 17 Apr 1932, Mrs. J. H.
Morrison (Holotype CAS 194148!; isotypes: CAS!, K!, GH!).

DISTRIBUTION AND HasitaT: California, Butte Co. and S Shasta Co.
Dry slopes, chaparral, foothill woodland, in sun to partial shade, in stony
soil with leafmold, on serpentine formations, 400m—1000m. Flowering
April-May.

SPECIMENS EXAMINED: CALIFORNIA: De Sabla May 1917, Helen M. Edwards
(NY); Feather R. near Yankee Hill, Butte Co., 23 Apr 1922, A. A. Heller 13618
(DS, WTU) ; occasional about shrubbery in leafmold in the Yellow Pine forest .. .
above the reservoir beyond Magalia, 1 May 1938, A. A. Heller 15058 (DS, MO,
WTU); on a serpentine outcrop about 4 mi from Magalia on the Coutolenc Rd.,
Butte Co., 8 Apr 1939, A. A. Heller 15367 (DS, MO, NY, PH, UC, WTU); near
Cherokee (pres. Butte Co.), a few in clusters, serpentine soil, 1 Mar 1942, V. Holt
(RM); in ponderosa pine woods S of Shingletown, Shasta Co., 21 Apr 1959, R.
Bacigalupi 7019 (JEPS); Pinkston Canyon, Big Bend Rd to Yankee Hill, Butte
Co., 1800 ft, with Arctostaphylos viscida, Berberis dictyota, Dichelostemma volubile,
Brodiaea californica, 18 Jun 1960, G. E. See and H. M. Beard s.n. (JEPS); near
Magalia, Butte Co., 15 Apr 1967, Margaret Williams sn. (RENO); near Magalia,
Butte Co., E slope amongst bunch grass in the open at the S end of cypresses (a
few scattered Douglas Fir and pines), growing in sand and humus amongst serpen-
tine gravel, 16 Apr 1967, W. Roderick s.n. (JEPS) ; N side of Butte Ck. near Honey
Run covered bridge, Butte Co., in wet area, soil a sandy clay, associates Meconella
californica, Brodiaea laxa, 26 Mar 1971, T. C. DeWitt 8 (HSC).

96 MADRONO [Vol. 25

F. micrantha

F. eastwoodiae

F. recurva

Fic.1. Floral characters of Fritillaria micrantha, F. eastwoodiae and F. recurva
showing (left to right) a) longitudinal section of flower with dehisced and unde-
hisced anthers; b) sepals with nectary in outline; c) petals with nectary in outline.

It will be seen from the above that the first collection of F. eastwoodiae
was in 1917 at De Sabla, about 20 mi NE of the type locality (it was
then identified as F. coccinea). In view of the confusion between East-
wood and Purdy, it is interesting to note that the next collection of F.
eastwoodiae, by Heller in 1922, was identified by him as F. lanceolata
var. gracilis Wats.

The name Fritillaria eastwoodiae commemorates Alice Eastwood, long
time curator of the California Academy of Sciences herbarium, and |
author of Fritidlaria brandeget, F. eximia, F. purdyi, and F. striata.

Without flowers there are few reliable morphological characters sepa-
rating the species F. recurva, F. micrantha, F. eas:woodiae and F. lanceo-

1978] MACFARLANE: FRITILLARIA 97

lata var. gracilis. A diagnostic key is given below based on floral charac-
ters, some of which are illustrated in Fig. 1.

Nectary less than 14, the length of tepals, flowers campanulate to nar-
rowly campanulate, style divided less than ™% its length, style
branches not strongly recurving.

Tepals 20-25 mm long, red with or tesselation, recurving at

their tips . . . . et IF vecurva
Tepals 10-15 mm long, a) mane or ale green, rarely showing
tesselation, straight to Aine at their tips . 2. F. eastwoodiae

Nectary greater than 1, the length of tepals, flowers open-campanu-
late, style divided more than /% its length, style branches strongly
recurving.

Flowers inconspicuously tesselated, sepals narrower than petals
3. F. micrantha
Flowers ponepicuonelay feceelted eepals broader or approximately
equal to petals . . . . . . 4.F. lanceolata var. gracilis

The tepal color in both F. micrantha and F. lanceolata var. gracilis is
purple-brown or rarely greenish, but the nectary of F. lanceolata var.
gracilis is longer and generally more conspicuous.

We return now to the question of the hybrid origin of F. eastwoodiae.
Its geographical distribution lies at the intersection of those of F. recurva
and F. micrantha (see the distribution map in Fig. 2). It should be
stressed that in this region it occurs as a series of stable populations, not
as sporadic individuals of recent hybrid origin. Pollination experiments
in the field show that the plants are self-sterile but produce normal cap-
sules when unprotected from visits by pollinators. This is the case in all
other N. American species studied, viz. F. atropurpurea, F. lanceolata,
F. micrantha, F. pluriflora, and F. recurva. It was also noted by Rix
(1976) in E. Mediterranean species, and it is believed that in general
Fritillaria are obligate outcrossers. The regular production of seed cap-
sules in populations of F. eastwoodiae containing diploids therefore sug-
gests that these are normal outbreeding populations.

Typical F. eastwoodiae is found on serpentine and it is therefore plaus-
ible that the mode of speciation involved stabilization of the hybrid, via
apomixis, in a habitat not occupied by either of the alleged parents.
Not all regions of overlap provide such a habitat, e.g., near Nevada
City, Nevada Co., F. recurva and F. micrantha are found, but not F.
eastwoodiae.

There are several other characteristics which strongly suggest hybrid
origin. First, the floral characters, which are the main points of distinc-
tion in this group of species, place F. eastwoodiae intermediate between
F. recurva and F.. micrantha—e.g., the flower and tepal shape and color,
and the style division (see Fig. 1). Fritillaria eastwoodiae shows a rather

98 MADRONO [Vol. 25

A F. micrantha

4 F. recurva

@ F. eastwoodiae

Fic. 2. Distribution map showing the occurrence of F. eastwoodiae in California,
at the intersection of the distributions of F. micrantha and F. recurva. 1 denotes a
location with possible backcrosses between F.. eastwoodiae and F. recurva; and 2,
between F. eastwoodiae and F. micrantha.

large variation in flower color and shape even in one population, but the
flowers are smaller than in F. recurva and do not have strongly recurv-
ing tepals or prominent yellow tesselation. Second, F. eastwoodiae shows
irregularities in chromosome pairing at meiosis (Cave, 1970) and fre- |
quent cases of female sterility (ovary and style absent). Many individu- |
als which have been studied cytologically are triploids. Somatic cell |
counts give 2m = 36 and pollen grains in a given individual show vari- _
able chromosome numbers distributed around m = 18 (Cave, 1970). |
This behavior is expected as a result of meiosis in the triploids. As noted —

|

1978] MACFARLANE: FRITILLARIA 99

earlier, diploid plants also occur and seed capsules are regularly produced.
Finally, controlled pollination of F. recurva (2) by F. micrantha (2),
and also the reverse cross, yielded capsules about 80% of the seeds of
which had normally developed embryos. Seeds from a single capsule of
F. recurva (2) F. micrantha (3) showed better than 50% germina-
tion, suggesting that interspecific sterility barriers do not exist between
these species. In particular, the endosperm of the hybrid seed does not
appear to contain embryo growth inhibitors such as have been found
in Lilium (Emsweller et al, 1962) and suspected in the case of E. Medi-
terranean Fritillaria (Rix, 1971).

The diversity of flower form found in F. eastwoodiae is probably due
in part to the frequent occurrence of triploids with their associated sexual
sterility, and to a stabilization of the diversity by vegetative propagation
which exists via the rice-grain bulblets produced in great abundance.
The mechanism for producing triploids is not clear. No tetraploid indi-
viduals of F. recurva or F. micrantha which potentially could be parents
in a triploid-diploid hybrid are known. Indeed, tetraploidy is rare in N.
American Fritillaria, only one case having been reported (in F. lanceo-
lata by Beetle, 1944). This is true of the genus as a whole, and is prob-
ably due to the large size of the chromosomes, which might inhibit cell
division in the tetraploid condition (Darlington, 1932; Grant, 1971). If
it were not for this, allotetraploidy would provide a convenient route to
bypass hybrid sterility. A more likely hypothesis is that triploids have
been formed many times from unreduced gametes, and these individuals,
because of their well-developed habit of vegetative apomixis, have sur-
vived as clones. The triploid character apparently confers some consti-
tutional advantage, since the triploids have competed successfully with
diploid parents and siblings, which have both sexual and vegetative chan-
nels available to them. The possibility of agamospermy is being investi-
gated. Chromosome counts have been done on only a few individuals of
the F. eastwoodiae complex (Cave, 1970), but it is interesting that the
only diploids reported (from Magalia, Butte Co., Cave, 1970; Marchant,
pers. comm.) are also the most intermediate morphologically and geo-
graphically between F. micrantha and F. recurva. Individuals closer to
F. recurva (e.g., R. Bacigalupi 7019, JEPS 25404 from Shasta Co.) are
triploid, and possibly represent backcrosses to F. recurva. Similarly, tri-
ploid plants have been found in El] Dorado Co. (W. Roderick s.n., JEPS
55692) which are close morphologically and geographically to F. mi-
crantha and perhaps represent backcrosses to the latter (see Fig. 2).
Whether unreduced gametes function more effectively in the backcrosses
remains, however, a matter of speculation, since populations of typical
F.. eastwoodiae contain both diploid and triploid individuals. The origin
of F. eastwoodiae needs further investigation, e.g., by a study of the de-
gree of heterozygosity using chromosome banding patterns (La Cour,
1951; Dyer, 1963).

100 MADRONO [Vol. 25

Finally, the probable hybrid origin of F. eastwoodiae raises an inter-
esting question regarding pollinators, since one of the supposed parents,
F. recurva, is hummingbird pollinated (Grant and Grant, 1968; D. San-
tana, pers. comm.) whereas the other, F. micrantha, with its open brown-
ish-green flowers and much less abundant nectar, almost certainly is not.
This suggests that hummingbirds are not the exclusive pollinators of
F. recurva.

ACKNOWLEDGMENTS
I thank Dr. Peter Raven and Dr. John Thomas for helpful discussions on nomen-
clature and for their critical reading of the manuscript; Dr. Chris Marchant for
providing a chromosome count of F. eastwoodiae; and Mrs. Margaret Campbell,
archivist at the California Academy of Sciences, for help in locating correspondence
between Alice Eastwood and Carl Purdy.

LITERATURE CITED

BEETLE, D. E. 1944. A monograph of the North American species of Fritillaria
Madrono 7:113-159.

Cave, M. S. 1970. Chromosomes of the Californian Liliaceae. Univ. Calif. Pub. Bot.
57:1-48.

DariincTon, C. D. 1932. Recent Advances in Cytology. Churchill, London.

Dyer, A. F. 1963. Allocyclic segments of chromosomes and the structural hetero-
zygosity that they reveal. Chromosoma 13:545—576.

Eastwoop, A. 1933. A new California Fritillaria. Leafl. West. Bot. 1:55.

EMSWELIER, S. L., S. ASEN and J. UHRING 1962. Tumor formation in interspecific
hybrids of Lilium. Science 136:266.

Furs, P. 1969. Fritillaria phaeanthera Eastwood 1933. Roy. Hort. Soc. Lily Year
Book 32:108-111.

Grant, K. A. and V. Grant. 1968. Hummingbirds and their Flowers. Columbia
Univ. Press, New York.

Grant, V. 1971. Plant Speciation. Columbia Univ. Press, New York.

La Cour, L. F. 1951. Heterochromatin and the organization of nucleoli in plants.
Heredity 5:37—-50.

Mouwz, P. A. and D. D. Keck. 1959. A California Flora. Univ. of California Press,
Berkeley.

Purpy, C. 1932. Western American Fritillaries. Roy. Hort. Soc. Lily Year Book
1:95-98.

Rix, E. M. 1971. The taxonomy of the genus Fritillaria in the Eastern Mediterra-
nean Region. Ph.D. Thesis, Cambridge Univ., Cambridge, U.K.

Watson, S. 1879. Contributions to American botany IX. Revisions of the North
American Liliaceae. Proc. Amer. Acad. Arts and Sci. 14:213-288.

A NEW SPECIES OF DRABA (CRUCIFERAE) FROM
WYOMING AND UTAH

. ROBERT D. DorN
Box 1471, Rawlins, WY 82301

In 1953 Rollins described Draba pectinipila with the type from Clay
Butte, Park County, Wyoming, at an elevation of 3050 meters. Two other
specimens from Daggett County, Utah, were cited. Rollins noted that
these plants are similar to Draba oligosperma but the petals are white
instead of yellow, the fruits are more elongate and bear doubly pectinate
hairs rather than simple hairs, and the pedicels and scapes are pubescent
instead of glabrous and are longer. He further noted that the Utah speci-
mens are very similar to the type material but differed in having a slight-
ly coarser pubescence and siliques that are tapered both above and below
rather than just below. He commented that the northeastern Utah-
northwestern Wyoming disjunct distribution may seem unusual but cites
the parallel distributions of Draba apiculata and Parrya nudicaulis.

Two points were apparently overlooked by Rollins (1953). First, the
type of Draba pectinipila is from an alpine habitat while the Utah speci-
mens are from the juniper-pinyon zone. Populations of Parrya nudicaulis
and Draba apiculata are from alpine or subalpine habitats in both lo-
calities. Second, one of the Utah collections, Williams 476, is labeled
“Fils. yellow.” I have since confirmed the yellow petal color in popula-
tions in Sweetwater County, Wyoming, near the Utah border.

The Utah and southern Wyoming populations grow among Pinus
edulis Engelm. and Juniperus osteosperma (Torrey) Little or just with
the juniper when the pine drops out to the north. Some plants occur in
sagebrush (Artemisia spp.) which is adjacent to or among the juniper.
In contrast, Draba oligosperma grows on exposed rocky slopes and ridges.

The juniper-pinyon populations are sufficiently different from the al-
pine population and from Draba oligosperma to warrant specific status.
Differences between the taxa are summarized in Table 1.

Draba juniperina Dorn, sp. nov. Herba perennis caespitosa. caulibus
erectis pubescenttibus 3-15 cm longis; foliis linearibus vel lineari-ob-
lanceolatis dense pubescentibus 2-14 mm longis, 0.5—1.5(2) mm latis;
pedicellis tenuibus divaricatis pubescentibus 3-10 mm longis; petalis lu-
teis spathulatis 4-5 mm longis; siliquis ovatis vel ellipticis pubescentibus
4—7 mm longis, 2-3 mm latis; stylis 0.7—-1.5 mm longis.

Caespitose, perennial herb; caudex much branched; leaves all near
base, linear to linear-oblanceolate, 2-14 mm long, 0.5—-1.5(2) mm wide,
pubescent with appressed doubly pectinate hairs, opposite pairs of leaves
connate to form a sheath; scapes erect, 3-15 cm long, pubescent with
doubly pectinate hairs; pedicels divaricately ascending, straight to slight-

101

102 MADRONO [Vol. 25

Table 1. DIFFERENCES BETWEEN THREE SPECIES OF Draba.

Character D. oligosperma D. pectinipila D. juniperina

Habitat Exposed rocky Alpine slopes Juniper-pinyon
slopes and ridges and sagebrush

Habit Compactly Loosely caespitose Compactly
caespitose caespitose

Scape Glabrous Pubescent Pubescent

Petal length 3-5 mm Mostly 3-4 mm Mostly 4-5 mm

Petal color Yellow White Yellow

Silique tip Tapered or rounded Rounded

Valve surface With simple, rarely With doubly With doubly
forked hairs pectinate hairs pectinate hairs

Style length 0.1-1 mm 0.3-0.7 mm 0.7-1.5 mm

ly curved upward, usually pubescent, 3-10 mm long; sepals broadly ob-
long to elliptic, hyaline-margined, 2—3.5 mm long, pubescent; petals yel-
low, spatulate, mostly 4-5 mm long; siliques elliptic to ovate, tapered to
tip, 4-7 mm long, 2-3 mm wide, flattened parallel to replum, pubescent
on valve surfaces with appressed doubly pectinate hairs; styles 0.7-1.5
mm long; stigma slightly greater in diameter than style; seeds 2—5 in
each locule, wingless, ca. 1.5 mm long, % to 2 as wide.

Type: Wyoming, Sweetwater Co., T12N R107W Sec 4, ca. 2135 me-
ters, juniper forest, June 18, 1973, Dorn 1837 (holotype RM, isotype
Wile):

Additional Collections: UtTaH: Daggett Co.: vicinity of Flaming
Gorge ca. 1675 meters, dry hillsides, June 1, 1932, Williams 476 (GH,
RM); same location, 24 km SE of Manila, June 3, 1938, Rollins 2275
(GH). Wyominc: Sweetwater Co.: T12N R107W Sec 9, ca. 2135 me-
ters, juniper-pinyon forest, May 29, 1977, Dorn 2895 (RM); T12N
R106W SE% Sec 12, ca. 2350 meters, sagebrush, May 29, 1977, Dorn
2898 (RM); Richards Gap, ca. 1950 meters, sagebrush at edge of juni-
per forest, May 29, 1977, Dorn 2902 (RM); T15N R102W Sec 18, ca.
2135 meters, juniper forest, May 29, 1977, Dorn 2903 (RM).

Specimens will be distributed to additional herbaria.

LITERATURE CITED
Rottiins, R. C. 1953. Draba on Clay Butte, Wyoming. Rhodora 55: 229-235.

1978] DORN: DRABA 103

5em

| Fig. 1. Draba juntperina. A, habit. B, silique. C, doubly pectinate hair. From
Dorn 1837.

104 MADRONO [Vol. 25

NOTES AND NEWS

VALIDATION OF THE NAME Juncus bufonius var. occidentalis. — In the publication
of Juncus bufonius var. occidentalis as “nom. et stat. nov.” for J. sphaerocarpus
auct. Amer., non Nees, (USDA Forest Service General Technical Report RM-18,
page 14. Oct. 1975) the mistaken assumption was made that reference to a previous-
ly published description was adequate for validation. Dr. Edward G. Voss reminds
me, however, that even in this instance Article 36 of the International Code requires
accompaniment by a Latin description or previously published Latin description, and
Article 37 requires the indication of a nomenclatorial type. The omission, therefore,
is remedied herewith.

Juncus bufonius var. occidentalis F. J. Hermann, var. nov. Plantae plerumque
minus quam 15 cm altae; tepala 24 mm longa; fructus subglobsus usque late ovoi-
deus, 2-3 mm longus. Type: near Camp by Grizzly Butte, Crook Co., Oregon, J. B.
Leiberg 256, June 16, 1894 (US; isotype MICH). — F. J. Hermann, Forest Service
Herbarium, Rocky Mountain Forest and Range Experiment Station, 240 W. Pros-
pect St., Fort Collins, Colorado 80521.

PLANT ABUNDANCE AND DISTRIBUTION IN RELATION TO TYPES OF SEED DISPERSAL IN
CHAPARRAL. — Recent attempts to relate the structuring of local communities to
dispersal mechanisms warrant examination of diverse vegetation types, and casual
explanation. Tree species diversity in the Great Smokey Mountains decreases signifi-
cantly with increases in the proportion of trees that are wind-dispersed (Beyer,
Amer. Natur. 109:103-104. 1975). The relationship does not hold for Ohio wood-
lands (Tramer et al., Amer. Natur. 110:500-501. 1976). The California chaparral
provides a strong contrast to the eastern forests in both physiognomy and dynamics.
Adaptation for wind dispersal is rare in chaparral, but a major division may be made
between dry- and fleshy-fruited species. The former have no obvious means of dis-
persal as far as three diameters of the parent shrub; the latter are presumably dis-
persed in animals and to relatively large distances. Then, is species diversity in
chaparral inversely related to the proportion of individuals that have dry fruits?

Chaparral in the San Gabriel Mountains was extensively sampled by Hanes (Ecol.
Monogr. 41:27-52. 1971) using 10m transects on which each individual plant was
noted by species and length of intercept. Hanes has kindly allowed me to analyze
287 of the transects. H. G. Baker kindly provided much of the data on fruits and
seeds. Sixty species occurred in the sample, of which 40% have fleshy fruits.

The correlation coefficient between species diversity —Z2p; In pi, and the propor-
tion of individuals on‘a transect that have dry fruits is only —.268, although this is
significant (p<.01). In fact, both species and individuals with fleshy fruits are
sparsely distributed. Over 80% of the transects had only 0 or 1 fleshy-fruited spe-
cies, while 90% had 2-4 dry-fruited. species. Also, given that a species ocurs on a
transect, the mean number of individuals is 1.6 for fleshy-fruited species, but 2.8 for
dry-fruited species (considering only species occurring on at least 10 transects;
p<.01, Mann-Whitney U-test). The fleshy-fruited species may be regarded as inter-
stitial, persisting in small cracks in the general matrix of dry-fruited shrubs.

IT suggest the following two hypotheses as explanations of the above observations.
1) Seeds dispersed by frugivorous animals tend to be deposited in compact groups,
and therefore have a very low dispersion within an area of the magnitude of 10m
diameter, even though the groups may be deposited at considerable distances from
the mother plant. 2) For any of a variety of reasons, seeds of fleshy-fruited species
cannot remain dormant beyond the first winter, yet only occasionally (in space)
does the moment of germination correspond with the moment when the site is suit-
able for establishment. Reasons for the lack of dormancy may include the follow-
ing: a) The seeds may be susceptible to fungal/bacterial, attack in fecal material;
b) Other seeds of the group might germinate first and preempt the site; c) Seeds of

{

1978] NOTES AND NEWS 105

these species tend to be heavier than those of dry-fruited species (mean weight of
23.49mg compared to 6.45mg, p<.01, excluding three species of each fruit type with
seeds >100 mg), making these seeds more favorable targets for seed predators. —
StePHEN H. Buriock, Biology Department, University of California, Los Angeles
90024. ie

GREAT BASIN VEGETATION IN CARBON County, Montana. — Great Basin vegeta-
tion is well represented in only two counties in Montana, Beaverhead and Carbon.
Both counties are along the southern edge of Montana, the former bordering Idaho
and the latter: bordering Wyoming. This vegetation is best represented in Carbon
County where it covers about 1800 sq. km. The greatest diversity of species occurs
in an area of less than 260 sq. km. in the south foothills of the Pryor Mountains
and along the Big Horn Canyon.

The Pryor Mountains and Big Horn Canyon have a predominant substrate of
limestone, sandstone, shale, and some gypsum. Several formations with red soils are
also well represented. These form extensive areas of nearly barren hills. The con-
spicuous vegetation of the area contains extensive stands of Utah juniper, Juniperus
osteosperma (Torrey) Little, and black sagebrush, Artemisia nova A. Nels. Several
species of Atriplex are also common.

Field work by the author in 1976 turned up several interesting plants from this
area including some first records for the state. The most interesting collections are
listed below. A few are not strictly Great Basin plants but are significant for other
reasons. Apparent first state records are indicated by an asterisk (*). Specimens are
deposited in MONT with some duplicates in RM and my personal herbarium.

BORAGINACEAE — Cryptantha ambigua (Gray) Greene, Dorn 2633; C. cana
(A. Nels.) Payson, Dorn 2546*, 2627; C. flavoculata (A. Nels.) Payson, Dorn 2543*,
2626.

CAPPARACEAE — Cleome lutea Hook., Dorn 2630.

COMPOSITAE — Artemisia pedatifida Nutt., Dorn 2534; A. spinescens Eaton,
Dorn 2552; Erigeron allocotus Blake, Dorn 2625* (This species is endemic to the
area and was known previously from only Big Horn Co., Wyoming.) ; Hymenopap-
pus fiilifolius Hook. var. luteus (Nutt.) B. L. Turner, Dorn 2660* (These plants
were growing with Hymenopappus polycephalus Osterh. with no evidence of inter-
grading.) ; Hymenoxys torreyana (Nutt.) Parker, Dorn 2549* ; Malacothrix torreyi
Gray, Dorn 2677; Platyschkuhria integrifolia (Gray) Rydb., Dorn 2638, 2670;
Sphaeromeria capitata Nutt., Dorn 2564; Tetradymia spinosa Hook. & Arn., Dorn
2684; Townsendia incana Nutt., Dorn 2628*, 2658; Wvyethia scabra Hook., Dorn
2629; X ylorhiza glabriuscula Nutt., Dorn 2535, 2686.

CRUCIFERAE — Malcolmia africana (L.) R. Br., Dorn 2540* ; Physaria austra-
lis (Payson) Rollins, Dorn 2547, 2559; Stanleya tomentosa Parry, Dorn 2652; Strep-
tanthella longirostris (Wats.) Rydb., Dorn 2570*, 2653.

HYDROPHYLLACEAE — Phacelia ivesiana Torrey, Dorn 2567*, 2632.

LEGUMINOSAE — Astragalus grayi Parry ex Wats., Dorn 2642; A. hyalinus
Jones, Dorn 2774.

LOASACEAE — Mentzelia pumila Nutt. ex T. & G., Dorn 2775*.

ONAGRACEAE — Cammissonia andina (Nutt.) Raven, Dorn 2568; C. minor
(A. Nels.) Raven, Dorn 2654* ; C. scapoidea (T. & G.) Raven, Dorn 2662.

POLEMONIACEAE — Gilia leptomeria Gray, Dorn 2631* ; Gilia tweedyi Rydb.,
Dorn 2566* ; Ipomopsis pumila (Nutt.) Grant, Dorn 2639.

POLYGONACEAE — Erigonum brevicaule Nutt. [E. pauciflorum Pursh var.
canum (Stokes) Reveal], Dorn 2690, 2773.

ROSACEAE — Physocarpus monogynus (Torrey) Coult., Dorn 2778.

SAXIFRAGACEAE — Sullivantia hapemanii (Coult. & Fish.) Coult., Dorn
2674* (This species is otherwise known from only a few localities in Wyoming. It
was found on dripping limestone.) .

106 MADRONO [Vol. 25

SCROPHULARIACEAE — Penstemon caryi Pennell, Dorn 2669*,:2777 (This
species is endemic to the area and was known previously only from Big Horn Co.,
Wyoming.) ; P. laricifoliuns Hook. & Arn., Dorn 2651.

TAMARICACEAE — Tamarix chinensis Loureiro, Dorn 2673.

UMBELLIFERAE — Musineon vaginatum Rydb., Dorn 2683 (This record from
the Pryor Mountains adds to the previous known range of the Big Horn and
Bridger mountains.). — RosBert D. Dorn, Box 1471, Cheyenne, Wyoming 82001.

KNOBCONE PINE SOUTHWARD RANGE EXTENSION IN THE SIERRA NEVADA. — Pinus
attenuata Lemm. has previously been reported as reaching its southern Sierra Neva-
da limit in Yosemite National Park (Griffin, J. R., and W. B. Critchfield, The dis-
tribution of forest trees in California, 1972; specifically, along the fire road to Deer
Camp, Arno, S. F., Discovering Sierra trees, 1973). We report here the existence of
a population near Bass Lake, for a range extension of ca. 35km south, out of Yo-
semite Park and into Madera County. The population is distributed west of the
Beasore Rd. 2.0-2.5 km north of Malum Ridge Rd. T7S, R22E, 10 (J. Keeley 7014,
Occidental College, Los Angeles). At this site it occurs in close association with
Arctostaphylos viscida and, on the periphery, mixes with Pinus ponderosa, Pinus
lambertiana. Libocedrus decurrens and Quercus chrysolepis. The population is an
uneven-aged stand of several hundred trees centered on a knoll at ca. 1200 m eleva-
tion. Several smaller populations occur 1.5—2.5 km further north on the Beasore Rd.

Pinus attenuata has previously been reported south of Yosemite Park (Munz,
P. A., Supplement to a California flora, 1968), but Munz’s report was apparently
based on knobcone pine planted along the Mineral King Rd. (Griffin and Critch-
field, op. cit.). The Beasore Rd. population is apparently indigenous. This is sug-
gested by the large size of the population and confirmation by a Sierra Nevada For-
est spokesperson (J. F. Underwood, Timber Management Officer, pers. comm. 14
Sept. 77) that knobcone pine has not been planted in this area. — Jon E. KEELEY,
Department of Biology, Occidental College, Los Angeles, Ca. 90041, STERLING C.
KEELEY, Department of Biology, California State University, Northridge, Ca. 91330,
and Janet LEE, Department of Botany, University of Kansas, Lawrence 66044.

SCROPHULARIA LAEVIS (SCROPHULARIACEAE), A LEGITIMATE SPECIES — Wooton and
Standley (Contr. U. S. Nat Herb. 16: 173. 1913) described S. laevis based on col-
lections from the Organ Mountains without flowers. The key in their “Flora of New
Mexico” (Contr. U. S. Nat Herb. 19, 578. 1915) states the flowers to be dull-green-
ish, apparently without basis.

Shaw, having seen only the type specimen in the National Herbarium, indicated
in his monograph (Aliso 5(2): 172. 1962) that S. laevis was synonymous with S.
montana. In the same monograph, (Ibid., 173) he listed the type specimen under
S. parviflora. His distribution map (Ibid., 173) showing both S. montana and S.
parviflora locations does not show the Organ Mountain location at all. That it was
probably omitted unintentionally is indicated since there are more New Mexico
voucher specimens cited than there are locations plotted. Later, based on his obser-
vation of plants grown from Organ Mountain seed, Shaw (private communication

with R. Roy Johnson) stated the plants resembled his hybrid, S. macrantha x S. |

parviflora.
All confirmed locations of S. parviflora are west of the Continental Divide, while

all those of S,. montana are east of the Divide. The location of S. laevis in the Organ |

Mountains is about 80 km from the nearest station of S. montana and over 200 km
from the nearest station of S. parviflora.

Plants of S. laevis collected at several sites in the Organ Mountains over the past |

15 years show little variation, indicating the stability and homogeneity of the popu-

lation. They differ from both S. montana and S. parviflora in their smaller stature, |

eS

1978] NOTES AND NEWS 107

carmine corolla, orbicular sterile filament, longer relative length of petiole to leaf
blade, non-fascilate inflorescence, and possible obligate substrate. The plants
further differ from S. montana in their slender habit, single type of pubescence, and
doubly serrate leaf margins and from S. parviflora in distribution, larger corolla,
completely deflexed lip, and limited pubescence.

Based on the above, S. laevis should be recognized as a valid species. The descrip-
tion of Wooton and Standley should be supplemented to include the following:

Glabrescent perennial, 4-10(12) dm tall, stems simple or sometimes with several
weak branches after cropping; leaf blades lanceolate to ovate, acute, proximally
doubly and distally simply serrate with acute teeth, obtuse to cordate at base, gla-
brous except sparsely glandular-puberulent along the main veins; larger leaves with
blades 5—7 cm long, 2—3.5 cm wide, on slender petioles 2-3 cm long; panicle sparse,
short, consisting of 2-5 pairs of few-flowered cpen corymbs; peduncles and pedicels
slender, glandular-puberulent at anthesis; sepals 3-4 mm long, triangular to lanceo-
late; corolla 7-12 mm long, pale carmine below, bright carmine above, the upper
lobes dark carmine, the reflexed lower midlobe white to pink, the throat glabrous;
sterile filament orbicular above, green, 1 mm broad; capsule narrowly ovoid, 8-11
mm long.

Distribution: Canyons, northern part of the Organ Mountains, Dona Ana Coun-
ty, New Mexico, on quartz monzonite substrate at 2250-2600 m elevation. —
Tuomas K. Topsen, Biology, New Mexico State University, Las Cruces, NM 88003.

CORDYLANTHUS MOLLIS SSP. MOLLIS (SCROPHULARIACEAE), REDISCOVERY OF EXTINCT
RECORD WITHIN NApa County, CALIFORNIA. — On 5 August 1976, Craig Thomson
and the author, during an examination of the salt marsh plant community along
Fagan Slough due west of the Napa County Airport about 1.5km east of the Napa
River (USGS Cuttings Wharf quadrangle: UTM 10/05610/42300, 10/05611/42300;
SEY SW% Sec. 3 and NEY% NW% Sec. 10 T4N R4W M.D.B. & M.), observed a
vigorous population of Cordylanthus mollis Gray ssp. mollis (soft bird’s beak) en-
compassing 250-300 individuals in a 30 x 70m area in association with Salicornia
virginica, Jaumea carnosa, Limonium californicum, and Cuscuta salina. Cordylan-
thus mollis ssp. mollis is officially listed as Endangered by the U. S. Fish and Wild-
life Service in 1976 (40 FR 24566). This taxon, identified by the Code COMOM, is
also listed by the California Native Plant Society as “Possibly Extinct” (Powell,
W. W. 1974. Inventory of Rare and Endangered Vascular Plants of California,
C. N. P. S. Spec. Pub. 1, p. 18), and was last reported in 1966 although previously
collected in Marin, Merced, Napa, Solano and Sonoma Counties. Voucher specimens
have been deposited by the author at the California Academy of Sciences. —
STEPHEN P. Rar, Box 66, Napa, California 94558.

TRAGUS RACEMOSUS IN ArIzoNA. — The genus Tragus comprises six or seven spe-
cies of annual grasses native to warmer regions of the Old World. Two of these,
T. berteronianus Schult. and T. racemous (L.) All., have been introduced into the
Americas, where the former species is often a common weed, being found from the
southern United States to Argentina. Tragus racemosus seems to be encountered
less frequently.

Hitchcock’s Manual (U.S. Dept. Agric. Misc. Publ. 200, 2nd. ed. revised by Agnes
Chase, 1951) gives the range of Tragus racemosus as: ‘Waste ground and on ballast
at a few places from Maine to North Carolina; Texas to Arizona; introduced from
the Old World.” Swallen, in his treatment of the grasses for Kearney & Peebles’
Arizona Flora (Univ. of Calif. Press, 1951), indicates for this species; “Campus
of the University of Arizona (Pima County), probably only cultivated.” In the sup-
plement to this work (1960) there is no further note regarding T. racemosus. Gould
(Grasses of Southwestern U. S., Univ. of Arizona Press, 1951) also states that this

108 MADRONO [Vol. 25

species is known in Arizona only from collections made on the University of Arizona
campus, but elaborates further: “This species was grown in the early grass garden
of the University as stated on the label of a Toumey collection made in 1892. Collec-
tions by Griffiths and Thornber made in 1901, 1902, and 1903, are labeled ‘Campus,
U. of A., Tucson’.” He also comments: “If this species actually did become estab-
lished in Tucson outside of the grass garden it is highly improbable that it has per-
sisted for nearly fifty years without detection.”

Examination of the Tragus racemosus folder at ARIZ in the summer of 1977 re-
vealed that the only specimens from Arizona were those mentioned above by Gould.
However, while botanizing in the Chiricahua Mountains (Cochise County) in Sep-
tember of that year, we encountered 7. racemosus in abundance in Rucker Canyon.
In this area it grows thickly as a weed in the sandy soil of the roadside from the top
of the pass (ca. 1750 m) for a distance of some ten miles as one descends toward the
east where the elevation is ca. 1525 m. Tragus berteronianus occurs here also, but
appears to be rather rare. In fact, we were able to find only two or three small colo-
nies in the entire ten mile stretch. Curiously, the two species did not appear to form
mixed stands.

As an additional check of our determinations, we collected young inflorescences of
plants considered to represent the two different species. These were preserved in the
standard 3:1 absolute alcohol-acetic acid solution for cytological study. Subsequent
examination of acetocarmine squashes revealed a chromosome number of 2” = 20
for 7. berteronianus, and 2n = 40 for T. racemosus. This is in accord with informa-
tion in the literature which indicates that the former species is a diploid, whereas
T. racemosus is tetraploid. Voucher specimens, as cited below, are deposited in
ARIZ, with duplicates at US. Collection numbers are those of the authors.

Tragus racemosus (L.) All., 6875, 2n = 40; T.. berteronianus Schult., 6880 &
6882, 2n = 20.

That Tragus racemosus is not a recent invader of Rucker Canyon is attested to by
a specimen collected there more than 30 years ago, incorrectly identified as T. ber-
teronianus, and placed in that folder at ARIZ where we found it in 1977. The col-
lectors are F. W. Gould & H.S. Haskell 4514. The label reads: “In sand along broad
wash, Juglans-Cupressus-Platanus woodlands; altitude 5700 feet; entrance to Ruck-
er Canyon recreational area. Chiricahua Mountains. Oct. 5, 1946.” — JouNn R.
REEDER & CHARLOTTE REEDER, Herbarium, University of Arizona, Tucson 85721.

REVIEW

A Survival Handbook to Sierra Flora. By NORMAN WEEDEN. 1975. iv. + 406, illus.
Interface California Corporation. $5.95. ISBN 0-915580-03-9.

In the past few years there has been a veritable explosion in the number of ‘popu-
lar’ wildflower guides at the disposal of the interested amateur botanist. Most pro-
fessionals would view this book as part of the exploitation of this market. Weeden’s
flora (W), however, is potentially of interest to the practicing botanist.

Essentially W is a series of keys to montane, subalpine and alpine Sierran plants,
supposedly including all species listed by Munz and Keck (M&K) from above 1066
m and 2438 m on the western and eastern slopes of the range respectively. Erigeron,
Carex, Cryptantha and Plagiobothrys are not treated by W at the specific level, and
the keys presented are not all entirely new, being mostly in the Abrams-M&K mold.
Illustrations are provided for most of the genera, and are useful although a few
border on primitive-art (i.e., Poaceae) and are not at all helpful. A glossary of 350
terms is provided. Brief habitat and morphological descriptions are also given for
most of the taxa. Numerous infraspecific taxa are omitted.

The pretension of survival in the wilderness by consuming wild plants is one un-
fortunate intimation of the book’s title. Weeden does present information on the
edibilitvy of many taxa, but many are cast aside with an “edibility unknown”.

1978] REVIEWS 109

I have carried W along with M&K in the field for the past two seasons to test the
usefulness of the former, and I have generally found W’s keys useful, but at times
ambiguous. The most frequent problem with W’s keys is the improper simplification
of morphological terminology. There are few errors, and only one glaring misspelling.

From this field comparison, and making the calculations below, I would say that
W deserves a place in the botanist’s backpack if weight is costly. One is most likely
to go astray with W’s keys when botanizing near the lower boundaries of his stated
elevational ranges: numerous species common within his limits are not keyed. If we
take P to be the probability of keying an unknown, P; being a correct determination
and Pw an incorrect determination, and if we assume P =.95 for M&K and P = .65
for W (my estimate !), then P; gram! for M&K is .0006, and P; gram-! for W is
0023. Clearly, then, W wins on a weight basis if determination error is tolerable.
However, Py gram-! for W is .0012, compared to Pw gram-! for M&K of .0003, so
that M&K is more exactly accurate on a weight basis.

Botanists do tire from lugging around heavy books in the field, and we do need
accurate field guides to introduce the objects of field botany to the people. Weber’s
Rocky Mountain Flora is exemplary in this regard (P; gram-! = .0016; Pw gram-!
= .0001 !). Weeden’s book does not quite approach this ideal, but it does serve a
distinct need. — Dean Wm. Taytor, Department of Botany, University of Califor-
nia, Davis 95616.

Vascular Plants of the Nevada Test Site and Central-Southern Nevada: ecological
and geographical distributions. By JANIcE C. BEATLEY. 1976. vii + 308, 28 figs.
Technical Information Center, U. S. Energy Research and Development Administra-
tion, Springfield, Virginia 22161. $9.75. ISBN 0-87079-033-1.

Inaccessible botany is often the product of the distance which botanists are able
to travel in their mostly random wanderings during vacations. In the case of the
area covered by this floristic volume, long distances from major botanical centers
and governmental access restrictions have conspired to make the flora of the 5100
km2 of the Nevada Test Site and vicinity poorly known. Janice Beatley, assisted in
the field at times by several other botanists, has amassed a significant number of
collections (25000) in the past fifteen years, and has produced several previous plant
checklists for the area. Culminating this effort is the release of this much-needed
and reasonably priced book.

The area covered lies on the phytogeographically important transition zone be-
tween the Sonoran and Great Basin floristic regions. An introduction presents the
background on the previous lack of floristic work in the area. Maps giving the
physiographic and. political features of the area comprise the first 3 figs., and the
fourth gives a generalized vegetation map. Unfortunately, several of the categories
in the legend to the latter map are nearly indistinguishable due to poor reproduction.
Figs. 5-28 are well chosen photographs of plant habitats.

The bulk of the book is divided into 2 parts: 1) Desert Environment and Vegeta-
tion (66 pp.) ; and 2) Vascular Plants (190 pp.). The first part presents a detailed
description of the habitat types in southern Nevada, and is perhaps our most com-
prehensive description of such to date. Numerous site data are reviewed, including
climatic and soil parameters. Kinds of vegetation of the area are discussed in a
semi-hierarchial classification: Mojave, Transition and Great Basin deserts subdi-
vided into kinds of sites (bajadas, mountains, arroyos, springs) or plant associations,
the latter being typified by phytosociologically uninformative ‘genus-genus’ or ‘ge-
nus-common name’ epithets. The second part is a catalogue of the flora arranged
alphabetically, listing 1093 taxa, describing habitat, local range, and phenology. Keys
and descriptions of the taxa are not given. The author justifies this omission by
stating that these identification tools “are (or will be) available for nearly all of the
taxa in the various floras of adjacent areas.” This omission is unfortunate. Keys in

110 MADRONO [Vol. 25

local floras can be more useful than those of the regional manuals in that they sim-
ply involve fewer choices, and are often the product of extensive field experiences
with the plants in hand.

Following are additional statistics on the flora which might be needed by users of
this book. The largest 10 families are (native taxa only): Asteraceae (17.4%), Poa-
ceae (6.8), Polygonaceae (5.8), Fabaceae (5.5), Brassicaceae (4.9), Polemoniaceae
(4.8), Scrophulariaceae (4.2), Boraginaceae (3.8), Onagraceae (3.5), Chenopodiaceae
(3.4). Of the 125 introductions, 28.8% are Poaceae, and the remainder: Asteraceae
(14.4%), Brassicaceae (10.4), Farabeae and Chenopodiaceae (5.6), Tamaricaceae
and Polygonaceae (3.2). A Life Form Spectrum for native taxa is: Phanerophytes
(2.9%), Chamaephytes (16.7), Hemicryptophytes (34.4), Geophytes (11.8), Thero-
phytes (31.4), Eiphytes (0.4), Succulents (2.1). Half of the introductions are thero-
phytes, and a fourth are hemicryptophytes.

The layout of the book is of adequate utility, and is mostly free of printing error.
However, taxa are not numbered in the catalogue of plants.

All factors considered, the book appears to this reviewer to be a good contestant
in the scramble competition for personal and university library dollars. — DEAN
Wo. Taytor, Department of Botany, University of California, Davis 95616.

The Story of Pines. By Nicuoras T. Mirov and JEAN HAsBrRoUCcK. xi + 148 pp.,
including colored frontispiece, forty halftones, five line drawings and one map. Indi-
ana University Press, Bloomington, Indiana. 1976. $7.95.

If one chose to write a book for the layman about a single group of plants, what
could be more appropriate than a genus containing the oldest known living “higher”
plant, the bristlecone pine? But not only are some pine individuals long-lived, the
group also has a long geological history. The chapter, “The First 200 Million Years”
discusses pines from the Jurassic to the present.

Dr. Mirov has worked with and loved pines for more than fifty years. Readers
may imagine themselves chatting with the Mirovs in the study or in front of their
fireplace, and hearing, in simple language, about fragrance of pines (one of Dr. Mir-
ov’s favorite subjects) — why one species exudes a different fragrance than another;
or the legend of how the Black Sea Pinus pityusa was named for the Greek wood
nymph Pitys; or reminiscences on how geneticists developed the science of breeding
pines, a science which had its beginning in 1925 at the Eddy Tree Breeding Station
in Placerville, California (now the Institute of Forest Genetics). The chapter on the
Pine Forest includes discussion of the ecosystem of which the forest is a part, and
also takes up the importance of fire to pines, “Fire never exterminated a pine forest;
only climate can do that.” [that is if man’s activities are excluded].

There is all this ‘and much much more” — structure, physiology, economic im-
portance, natural distribution of pines in the northern hemisphere and their success-
ful introduction into parts of the southern — all in a welcome, readable style. As-
pects of pines that still require research are touched upon and these may intrigue
the scientists who pick up this volume.

Anyone who feels the exhilaration of being among pines, whether they be on a
coastal plain, on a mountain peak, or in the foothills between, will enjoy this book
and will undoubtedly look upon pines with a deepened understanding. — ANNETTA
CarTER, Herbarium, Department of Botany, University of California, Berkeley 94720.

Daleae Imagines, an illustrated revision of Errazurizia Philippi, Psorothamus Ryd-
berg, Marina Liebmann, and Dalea Lucanus emend. Barneby, including all species
of Leguminosae tribe Amorpheae ever referred to Dalea. By RupERT C. BARNEBY.
891 pp., including 142 plates drawn by the author. Memoirs of the New York Botan-
ical Garden, vol. 27. 1977. $50.00 (paperbound!).

1978]

REVIEWS 111

Sage BARNEBY, who’s had his share of fame,
With this new work may greener laurels claim;
We’ve seen some monographs of equal length,
But few that mix such elegance with strength;
So often merit’s antonym to size

That epic length we’re tempted to despise

(Thus STANDLEY wrote, with great facility,
Long works of flawed reliability,

And RypbsBErG, who penned much with firm decision,
Was cursed, like SMALL, with brash pedantic vision).
So now we’re grateful for this splendid book
Which justifies the decade Rupert took;
Amorpheae, as Barneby defines,

Includes eight genera in its confines;

We’re startled that the Dalea we knew

Was not erected by A. L. Jussieu

(But after all, we got into this bind

Because the great LinnarEus changed his mind) ;
From Dalea two taxa are set free:

To Psorothamnus goes the fair Smoke Tree,
Marina comes back from obscurity ;

The prairie clovers (Petalostemon)

Regain their petals — but their rank has gone.
The species groups are many and compound
But their new circumscriptions look quite sound;
We find that the descriptions and the keys

Are well designed, and can be used with ease.
The many illustrations set this book apart
Through exquisite detail of patient art:

The author’s pencil draws each plant’s Gestalt
As BAUER might have done, without a fault;
These species portraits, polished and unique
(Though one regrettably is forced to seek

Each picture far removed from its own text),
Have captions discursive and multiplex.
Although it would have made the book more weighty,
We miss the maps and indexed exsiccatae ;

Still, these are but inconsequential flaws

Which need not damp our chorus of applause:
For Barneby, with flair and art precise,

Has wrought a masterpiece of awesome price;
This noble guidebook to the Daleae

Will find botanic immortality.

— Gravy L. WEssTER, Department of Botany, University of California, Davis

95616.

SCIENTIFIC COLLECTING IN MEXICO

In a recent interview with Lic. Juan Soto Fierro of the Departamento de la Cons-
ultoria Juridica del Consejo Nacional de Ciencias y Tecnologia, on the matter of
collectors, the following rules are presented that should be observed in collecting
plants, animals, and geological specimens. This message is directed especially to the
Members of the Society who reside outside of Mexico. — Signed:

Biol. Magdalena Pena de Sousa, Presidente,
Sociedad Botanica de Mexico.

112 MADRONO [Vol. 25

As an aid to foreign scientists and the Mexican scientific community, the Consejo
can help in their applications to the appropriate authorities when it is proposed to
make scientific collections in Mexico. In all cases investigators should apply to the
Mexican Consulate [nearest them] for entry as “No-immigrante visitante”.

COLLECTING OF PLANTS, ANIMALS, AND GEOLOGICAL SPECIMENS

Application should be made three months in advance of the proposed collecting
trip to Consejo Nacional de Ciencia y Tecnologia (CONACYT) (Consultoria Juri-
dica, Lic. Margarita Peimbert) [Insurgentes Sur 1677, Apdo. Postal 20-033, Mexico
20, D.F., Mexico, telephone 5-34-80-80, Ext. 164]. CONACYT processes applications
and issues permits. Applications should give the following information:

1. Name, nationality, and special interest.

2. Letter from their institution stating approval of the proposed investigation;
thus an authority of the institution accepts the responsibility of sending a
follow-up report.

3. Curriculum vitae.

4. General program of the work, specifying the scientific boundaries to be pur-

sued in the project.

. Proposed itinerary for the investigation, noting the places and routes to be
followed.

6. Intended dates of the investigation.

7. Number and scientific names of the species to be collected.

Animals, Including Insects

Those collecting animals, including insects, should send in addition:

8. Two photographs of passport size.

9. A check for $30.00 (U. S.) to “Direccion General de la Fauna Silvestre” for
permission, which will be valid only for the year of the expedition.

10. The place and approximate date that they will leave Mexico in order to advise

the Delegado Forestal y de Fauna in order to issue a certificate of compliance.

After completing the expedition, the investigator should do the following:

1. Send a report of the field work to CONACYT [see above] and to the Secre-
taria de Agricultura y Recursos Hidraulicos [Ave. Insurgentes Sur 476, Mexico
7D He :

2. Send a representative series of the collections obtained.

Plants

Thos collecting plants should send a specimen of each species or variety collected,
with labels including scientific name, vernacular name, locality, habitat, altitude, and
date of collection, plus an ecological description of the locality, to each of the fol-
lowing institutions:

1. Herbario Nacional del Instituto de Biologia, U.N.A.M., Apdo. Postal 70-367,

Mexico 20, D.F., México.

2. Instituto Nacional de Investigaciones Forestales, Progreso No. 5, Coyoacan,
México, D.F., México.

Also send to both of the above institutions a copy of the final report as well as

any publications that result.
*** Massive collection of plants is prohibited. ***
Both Plants and Animals .

Check at point of departure from Mexico with Delegado Forestal y de la Fauna
and obtain certificate of having complied with the stipulations of the permit.

The Secretaria de Agricultura y Ganaderia stipulates that failure to comply with
any of the mentioned requisites will be sufficient to prevent granting of further per-
mission either to the person in question or to the institution who remommended the
investigator. | j

[Above translated from Macpalxichitl, the Boletin Bimestral de la Sociedad Bo-
tanica de México, Sep-Oct 1977, by Annetta Carter. ]

OL

Membership in the California Botanical Society is open to individuals ($12.00 per
year, regular; $8.00 per year, student). Members of the Society receive MApDRONO
free. Institutional subscriptions to MApRONO are available ($20.00 per year).

Back issues of Madrofio are available at the following rates (some issues are
out of print):

Vol. 1 (1916-1929) and Vol. 2 (1930-1934, each consisting of 17 numbers: $1.50
per issue and $25.50 per volume for members; $3.00 per issue and $51.00 per volume
for institutions.

Vol. 3 (1935-1936) through Vol. 23 (1975-1976), each biennial, consisting of 8
numbers: $3.00 per issue and $24.00 per volume for members; $5.00 per issue and
$40.00 per volume for institutions.

Vol. 24 (1977) et seq., one volume per year, each consisting of 4 numbers: $3.50
per issue and $14.00 per volume for members; $5.00 per issue and $24.00 per volume
for institutions.

Applications for membership (including dues), orders for subscriptions, requests
for back issues, changes of address, and undelivered copies of MApDRONO should be
sent to the California Botanical Society, Inc., Department of Botany, University of
California, Berkeley 94720.

INFORMATION FOR CONTRIBUTORS

Manuscripts submitted tor publication in MaproNo should be sent to the Editor.
Membership in the California Botanical Society is requisite for publication in
MaproNo.

Manuscripts and accompanying illustrative materials must be submitted in dupli-
cate and should follow the format used in recent issues of MaAproNo. Original illus-
trations should not be submitted until paper is accepted for publication. All manu-
scripts MUST BE DOUBLE SPACED THROUGHOUT, including title, text, tables,
captions, lists, literature cited, etc. Footnotes (which should be avoided wherever
possible), captions, and tables must be typed on sheets separate from the text.
Presentation of nomenclatural matter (accepted names, synonyms, typification)
should follow the format used for Rhus integrifolia in MapRONO 22:288. 1974. All
measurements should be given in S. I. (metric) units. Where appropriate, scales
should be included on figures rather than in captions as estimates of ratio of repro-
duction such as X %, X 1620, etc. Institutional abbreviations in specimen citations
should follow Holmgren and Keuken’s list (Index herbariorum, Part 1, The herbaria
of the world. Sixth edition. 1974. Regnum Veg. vol. 92). Abbreviations of names of
journals should be those in Botanico-Periodicum-Huntianum (Lawrence, G. H. M.
et al. 1968. Hunt Botanical Library, Pittsburgh). If the correct abbreviation cannot
be determined, the full title of the journal should be used. Titles of books should be
given in full, together with the place and date of publication, names of publisher,
and an indication of the edition, if other than the first.

Short articles such as range extensions and other brief notes are published in con-
densed form under the heading “Notes and News”. Authors of such articles should
follow the format used in recent issues of MapRONO.

Authors are allowed up to 10 pages per year without page charges; charge for
additional pages is $30.00 per page. Subject to approval by the Editors, articles may
be published ahead of schedule, as additional pages of an issue, provided the author
assumes complete costs of publication.

Editorship of Madrofo. — Terms of current editors of Madrofio end 31 June
1978. The Board of Directors of the California Botanical Society have appointed
James Hickman to serve as Editor. After 15 June 1978, manuscripts intended for
publication in Madrofio should be sent to Dr. James C. Hickman, Department of
Botany, University of California, Berkeley 94720.

de

ils3
2 °
por.

VOLUME 25, NUMBER 3 JULY 1978

>» Contents
Z SYSTEMATICS OF MIRABILIS SUBGENUS QUAMOCLIDION, (NYCTAGINACEAE) ,
< George E. Pilz 113
fase LreaF ANGLE AND LIGHT ABSORPTANCE OF ARCTOSTAPHYLOS SPECIES
oO (ERICACEAE) ALONG ENVIRONMENTAL GRADIENTS, Gaius R. Shaver 133
ca FLORA AND CHOROLOGY OF THE PINUS ALBICAULIS — VACCINIUM
SCOPARIUM AssociATION, Frank Forcella 139
fi THe GENUS TRICHOSTEMA (LABIATAE) IN Mexico, Harlan Lewis and
‘e) Jerzy Rzedowski 151
GOSSYPIUM TURNERI (MALVACEAE), A NEW SPECIES FROM SONORA,
mn] Mexico, Paul A. Fryxell 155
<i CHROMOSOME NUMBERS IN ASTERACEAE, A. Michael Powell and
7 Shirley A. Powell 160
fa NOTES AND NEWS
a) RANUNCULUS CALIFORNICUS, A NEw RECORD FOR THE STATE OF
WasHincTon, Melinda F. Denton 132
© NoTEs ON Two RarE, ENDEMIC SPECIES FROM THE KLAMATH
— REGION OF NORTHERN CALIFORNIA, PHACELIA DALESIANA
(HyDROPHYLLACEAE) AND RAILLARDELLA PRINGLEI (COMPOSITAE),
Z William J. Ferlatte 138
<i HEMIZONI CONJUGENS (COMPOSITAE) : DISTRIBUTION,
S) CHROMOSOME NUMBER, AND RELATIONSHIPS, Barry D. Tanowitz 159
nt NEw PLanT DISTRIBUTION RECORDS FROM THE SOUTHWESTERN
UNITED STATES AND NORTHERN Mexico, Richard Spellenberg . 169
oe IRREGULARITY OF PINYON CONE PRODUCTION AND ITS RELATION TO
i Pinyon Cone Motu Prepation, Frank Forcella 170
= NOTES ON THE FLora oF East-CENTRAL IDAHO,
< Douglass M. Henderson 172
REVIEWS
= MicHacEt G. BARBOUR AND JACK Major, Terrestrial Vegetation of
(x California (W. D. Billings) 174
(Peter Stekel and Edward A. Cope) 175
B BOOKS RECEIVED 154
<< SPECIAL ANNOUNCEMENTS 150, 176, back cover

UBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY

Maprono is published quarterly by the California Botanical Society, Inc., and is
issued from the office of the Society, Herbarium, Life Sciences Building, University
of California, Berkeley. Established 1916. Second-class postage paid at Berkeley.
Return requested.

Editor — James C. HicKMAN
Department of Botany
University of California, Berkeley 94720
(415) 642-1079

Board of Editors
Class of:

1978—SHERWIN CaARLOQUIST, Claremont Graduate School
LeEsLiE D. Gorttiies, University of California, Davis
Dennis R. PARNELL, California State University, Hayward
1979—Puitie W. RuNDEL, University of California, Irvine
IsABELLE TAvaAREs, University of California, Berkeley
1980—JamMEs R. GriFFin, University of California, Hastings Reservation
Frank A. LAnc, Southern Oregon College, Ashland
1981—DanIEL J. CRAwWForD, Ohio State University, Columbus
JaMEs HENRICKSEN, California State University, Los Angeles
1982—DeEan W. Taytor, University of California, Davis
RICHARD VocL, California State University, Los Angeles

CALIFORNIA BOTANICAL SOCIETY, INC.

OFFICERS FOR 1978

President: Dr. MicHAeEtL F. Baap, Department of Biological Sciences,
California State University, Sacramento 95819

First Vice President: Dr. ALAN R. PoLANSHEK, Department of Biological Sciences,
California State University, San Jose 95114

Second Vice President: Dr. AMy Gi~MartTIN, Department of Botany,
Washington State University, Pullman 99164

Recording Secretary: Dr. CHARLES F. QUIBELL, Department of Biological Sciences,
California State University, Rohnert Park 94928

Corresponding Secretary: Dr. RupoLF ScHmip, Department of Botany,
University of California, Berkeley 94720

Treasurer: Dr. JoHN H. THomas, Department of Biological Sciences,
Stanford University, Stanford, California 94305

The Council of the California Botanical Society consists of the officers listed above
plus the immediate Past-President, WINSLOW R. Briccs, Carnegie Institute of Wash-
ington, Stanford, CA 94305; the Editors of Madrono; three elected Council mem-
bers: James R. GrirrFin, Hastings Reservation, Star Route 80, Carmel Valley, CA.
93924 (1976-1978) ; Harry D. TuHriers, Department of Ecology and Systematic Bi-
ology, California State University, San Francisco, CA 94132 (1977-1979) ; ALAN R.
SmitH, The Herbarium, Department of Botany, University of California, Berkeley,
CA 94720 (1978-1980) ; and a Graduate Student Representative, Nancy R. Morin,
Department of Botany, University of California, Berkeley, CA 94720.

i

SYSTEMATICS OF MIRABILIS SUBGENUS QUAMOCLIDION
(NYCTAGINACEAE)

GEORGE E. Piz
Department of Biology, The Polytechnic, Ibadan, Nigeria

Monographs and revisions by Standley (1909, 1911, 1918) and Hei-
merl (1889, 1897, 1934) form the basis of the present taxonomic knowl-
edge of a predominantly North American family. After having studied
the family for over two decades, Standley (1931a) stated that he knew
few groups of plants in which specific differences were more unstable or
baffling. After studying South American material, Standley (1931b)
transferred to Mirabilis all those plants he had previously placed in
Oxybaphus, Hesperonia, Quamoclidion, and Allioniella. Heimer] (1934)
followed Standley’s circumscription of Mirabilis but retained Quamo-
clidion with the single species of Q. triflorum. Subsequent treatments of
Mirabilis (sensu Standley, 1931b) have dealt only with small parts of
the genus as they are represented in regional floras.

Plants referable to Mirabilis subgenus Quamoclidion are characterized
by a gamophyllous involucre that is only slightly accrescent after anthe-
sis and surrounds three or more flowers. An initial purpose of this study
was to compare the broad, ovate, distinct bract subtending each individu-
al flower of Hermidium alipes with the gamophyllous involucre of Quam-
oclidion. As a taxonomic result of this study I am transferring Hermidt-
um alipes to Mirabilis subgenus Quamoclidion.

Mirabilis subgenus Quamoclidion comprises six species that grow pre-
dominantly in western North America between 20°-45° N. Open habi-
tats at the lowermost edge of juniper woodland are where most popula-
tions and the greatest number of plants per population are found.
Quamoclidion is not restricted, however, to areas where Juniperus occurs.
Milabilis triflora inhabits a scrub vegetation dominated by Acacia, Opun-
fia, and an occasional Ficus in Baja California del Sur and in Jalisco. At
the other geographic extreme M. marfarlanei shares dry sand bars with
Celtis on the floors of the Snake and Salmon River canyons.

MorPHOLOGY

My circumscription of Quamoclidion is based on characters that are
stable and rather uniform within the subgenus but which differ widely in
other members of Nyctaginaceae. These include inflorescence and pollen
morphologies as well as chromosome number and habitat preference.
Floral and fruit morphologies possess the most diagnostic characters for
circumscription of species within Quamoclidion because they are stable
at the population level but may vary considerably within a species. Leaf

Maprono, Vol. 25, No. 3, pp. 113-176, September 14, 1978.
113

114 MADRONO [Vol. 25

shape and vestiture vary considerably even within populations. Unfor-
tunately, characters used by other authors (Heimerl, 1934; Standley,
1918) to delimit species of Mirabilis are those which I have found to
vary widely within populations due to extreme seasonally and environ-
mentally induced morphological variation.

INFLORESCENCE. Flowers of Quamoclidion are arranged in soli-
tary, determinate, involucrate heads at ends of branches and in axils of
upper cauline leaves. As in Bougainvillea, each flower is intimately asso-
ciated with a subtending bract. Most taxa of Quamoclidion have a six-
flowered head consisting of a naked, solitary, central flower surrounded
by five flowers borne on the bases of gamophyllous involucral bracts.
Occasionally the central flower is not naked. In all taxa of Quamoclidion,
IT have observed heads that possess central flowers subtended by a dis-
tinct bract similar to one of the involucral bracts. Barneby (1942) stated
that this occasional occurrence suggests that heads of “advanced” mem-
bers of Quamoclidion may have evolved rather recently from the “‘primi-
tive bougainvilleoid” flower head of Mirabilis alipes, in which every
flower has its own independent, subtending floral bract. Although distinct
bracts are usual in this species, heads with the five outermost bracts
united by their margins to one-half their length are common. These in-
volucres are campanulate and have the aspect of the gamophyllous invo-
lucres of other taxa within Quamoclidion.

Exceptions to the six-flowered head of most taxa of QOuamoclidion
occur in the consistently three-flowered Mirabilis triflora and in M.
greenei with as many as 16 flowers in some involucres. When a second or
third whorl of flowers is present, each member of the whorl is borne on
the midvein of one of the five bracts, each one distal to the flower of the
first whorl. Flowers on the second and third whorls often abort before
anthesis but many others may produce fruit. |

POLLEN, Nowicke (1970), utilizing light microscope and scanning
electron microscope techniques, found that pollen of Nyctaginaceae is
structurally variable; however, features within genera are remarkably
uniform. This is certainly true for the uniformly very large, spheroidal,
and pantoporate pollen grains, with a spinulose sexine pattern, of Quamo-
clidion. Pollen-grain diameters form a continuum (100-160 pm) and are
not taxonomically useful for separating taxa within the subgenus. The
range, standard deviation, and mean of pollen-grain diameters from 54
collections of Quamoclidion may be found elsewhere (Pilz, 1974).

_ Pollen viability is typically very high. If immature flowers or flowers _
from plants that are obviously under water stress are observed, the pol- —
len viability sometimes drops below 90%. The relatively small pollen- —
grain diameter reported by Nowicke (1970) for Mirabilis alipes is prob-
ably due to the selection of an immature flower. A notable exception,
M. pudica, often has 30-50% of its grains ill-formed and noticeably —

1978] PILZ: MIRABILIS Ps

smaller than the mean grain size of other species. These smaller grains
also fail to stain in aniline blue-lactophenol. Only one collection of M.
pudica (Beatley 8211, RSA) has nearly 100% viable pollen, and this is
coupled with a significantly smaller grain diameter than is typical of the
species. This aspect of the reproductive biology of M. pudica requires
further study because other collections made by Beatley in the same
area have the low viability typically found in the species throughout its
geographic range.

FRUITS. The basal portion of the perianth begins to enlarge rapidly
soon after fertilization, reaching a size slightly larger than that of the
mature fruit. It is initially bright green and contains an ovary that is
about one-fifth that of the ultimate size of the fruit. The developing
embryo enlarges and fills the cavity of the persistent perianth, and the
mature fruit is termed an anthocarp: “a fruit formed by the union of
floral organs or part of them, with the fruit itself’ (Jackson, 1928). A
thin pericarp envelops the single uncinate embryo and the mealy peri-
sperm. The endosperm is completely absorbed during growth and ma-
turation of the embryo, except for a small amount forming a cap or col-
lar around the apex of the radicle. Cells of the nucellus become packed
with starch and form a mealy perisperm, which constitutes the storage
tissue of the mature seed (Cooper, 1949).

Anthocarps are generally ellipsoidal, although they range from nar-
rowly ovoid to subglobose (Fig. 1). The proximal end is constricted and
the distal end may be so to a lesser extent. Anthocarps are brown to
nearly black except in Mirabilis alipes, which has olive-green fruits that
turn brown only if they are wetted after maturity and allowed to dry
again. Ten slender, tan, longitudinal ribs are evident on anthocarps of
most species, particularly those of M. multiflora. These are remnants of
vascular traces that earlier extended to the upper portion of the perianth.

The surface of mature anthocarps is sparsely pubescent in Mirabilis
triflora and in some individuals of M. multiflora var. glandulosa. Other
members of Quamoclidion have glabrous anthocarps. The anthocarp
surface is smooth or rugulose, a difference that is correlated with produc-
tion of mucilage by the fruit when wetted. The rugulose anthocarps of
M. triflora, M. alipes, M. greenei, M. macfarlanei, and M. multiflora
var. glandulosa produce mucilage in variable amounts. Mirabilis triflora
produces a light mucilage that surrounds the fruit and is two or three
times the volume of the anthocarp. Other mucilage-formers produce a
thick heavy substance that is usually less than 1 mm thick. Plants of
those taxa that have smooth anthocarps, on the other hand, produce no
mucilage when wetted. Mirabilis pudica, M. multiflora var. multiflora
and M. multiflora var. pubescens are included in this second group.

CHROMOSOME NUMBERS AND REPRODUCTIVE BIOLOGY
Chromosome numbers have been reported for only 19 of the more than

116 MADRONO [Vol. 25

hy

i fe

“hie I! \ iy vA
ee aheen

Wr,
vx" Ny /.
MRS TAL

eae

on 5 TARY,

SS oes

NSS

= Sy
: ro S =
Saar a
Boal Sy Sears
ro idee ———

SN
SYS

Ve

SSss3“°““~u

* DDFs IFAs FIFTFY

Ss

BYS DSN He era pot

Seta NEY

ne)
1)
ol ow

a z =>
coke shal Die 8
renee ae = ue
SLs oO (S

Fic. 1. Representative flowers and anthocarps of Mirabilis subg. Quameclidion.
Collection numbers are mine unless otherwise indicated. Voucher specimens are in
UC. A. M. alipes 788, A’. 1125; B. M. macfarlanei Constance et al. 1579. B’. 1282;
C. M. triflora 1270, C’. 949; D. M. pudica Ripley & Barneby 4403, D’. 1132; E. M.
greenei 998, E’. 990; F. M. multiflora var. multiflora 1235, F’. 869; G. M. multiflora
var. glandulosa 1151, G’. Osterhout 6559; H. M. multiflora var. pubescens 831.

1978] PILZ: MIRABILIS Ll?

250 taxa of Nyctaginaceae (Pilz, 1974). Many counts were determined
from serial sections of paraffin-imbedded material, a technique that can
easily lead to errors in interpretation (Epling et al., 1962). Showalter
(1935), using paraffin-imbedded material reported n = 29 for Mirabilis
multiflora. This report is probably due to incorrect interpretation of sec-
tioned material. Using slightly modified aceto-carmine squash techniques
(Beeks, 1955; Pilz, 1974) for anthers and root tips I have consistently
found the gametic chromosome number m = 33 in the taxa under con-
sideration throughout their geographical ranges (Table 1). This consis-
tency is significant in view of the variation existing within the family as
a whole. Species of the same genus within the family tend to have identi-
cal chromosome numbers although considerable variation occurs within
both Boerhavia and Bougainvillea that have both aneuploid and poly-
ploid series (Pilz, 1974).

Except for Mirabilis triflora all species of Quamoclidion are night-
blooming. The ephemeral flowers usually open about three hours before
dusk and close about mid-morning. Cloudy days may extend the hours
that flowers remain open in the morning; however, once closed, the thin
delicate perianth lobes become flaccid and do not reopen.

Visitations to flowers by several kinds of animals have been observed,
but only hummingbirds and hawkmoths are effective pollinators. Ants,
bees, butterflies, and even flies commonly visit the flowers and collect
pollen (Cruden, 1970), but the exserted style, 2-30 mm beyond the peri-
anth tube, and its position at one side of the flower make the transfer of
pollen to the stigma by these organisms rare.

I have observed Hylocharis xanthusi, the Black-Fronted Humming-
bird, a species restricted to the southern half of Baja California, to be
the principal pollinator of Mirabilis triflora in that region. While hover-
ing before the long red tubular flower and taking nectar, the birds brush
the exserted stamens and stigma with the tops of their heads, which be-
come yellow with pollen. Hummingbirds show very low species consisten-

TABLE 1. CHROMOSOME NUMBERS IN Mirabilis susc. Quamoclidion. Collection
numbers are mine. Voucher specimens are in UC.

M. pudica Barneby. Nevada, Lincoln Co., 4 mi W of Crystal Spring, 975 2n = 33 II.

M. greenet S. Wats. California, Glenn Co., 7 mi N of Stonyford, 1274, 2n = 33 II.

M. multiflora (Torr.) Gray var. multiflora. New Mexico, Grant Co., 3 mi. W of San
Lorenzo, 724, 2n = 33 II. New Mexico, Catron Co., 10 mi S of Luna, 726,
2n = 33 II. Arizona, Coconino Co., 6 mi SE of Desert View, 1236, 2n = 66.
New Mexico, Rio Arriba Co., 8 mi W of Abiquiu Dam, 1245, 2n = 66.

M. multiflora (Torr.) Gray var. pubescens S. Wats. California, Kern Co., Caliente,
081, 2n = 66. Arizona, Mchave Co., 23 mi E of Kingman, 910, 2n = 33 II.
California, Kern Co., Caliente, 961, 2n = 33 II. Nevada, Lincoln Co., 13 mi S
of Caliente, 1143, 2n = 33 II.

M. multiflora (Torr.) Gray var. glandulosa (Standley) Macbr. Nevada, White Pine
Co., 1 mi E of Connor’s Pass, 1144, 2n = 33 II.

M. triflora Bentl. Baja California del Sur, 2 mi N of La Burrera, 947, 2m = 33 II.

118 MADRONO [Vol. 25

cy when visiting flowers unless a single nectar-producing species marked-
ly predominates (Baker, 1961). Mirabilis triflora often grows in large,
dense populations in the Laguna Mountains of Baja California del Sur,
and this, along with the abundance of hummingbirds observed visiting
the flowers, may account for the very high percentage of seed set (see
further discussion below).

Pollinators of taxa other than Mirabilis triflora are less certainly
known. Baker (1961) reported that M. multiflora var. pubescens (M.
froebeli) is visited by hawkmoths, Celerio lineata, at the University of
California Botanical Garden in Berkeley, but this site is outside the nat-
ural range of the species. He also noted that hummingbirds make their
visits in the morning and evening when the flowers are open, and he hy-
pothesized that the sharing of flowers by both hawkmoths and humming-
birds may be a widespread phenomenon (Baker, 1961). Cruden (1973)
stated that flower opening and anther dehiscence occurs too late in the
day in M. multiflora for visits by hummingbirds and bees “prior to visi-
tation by hawkmoths, the coevolved pollinators’’. In contrast, I have ob-
served hummingbirds and hawkmoths sharing flowers of M. multiflora
with dehisced anthers at dusk in Grant County, New Mexico. Thus at
least some sharing of floral resources by these two pollinators is occur-
ring late in the day soon after the flowers open. No pollinators of other
taxa have been reported, and I have not observed any.

The amount of seed set on a plant is often extremely low. Plants iso-
lated from others by more than 100 m rarely produce fruit. This probab-
ly is due to their self-incompatible breeding system, known in at least
three taxa (Pilz, 1974), coupled with the lack of compatible pollen trans-
fer to these isolated plants. Many populations fail to produce fruit in a
given year. I have recorded abortion, presumably due to lack of available
water, of all flower buds on all plants of some populations of Mirabilis
alipes, M. macfarlanei, M. greenei, and M. multiflora. Absence of polli-
nator visits during periods of strong winds also contributes to low seed
set recorded for populations of M. alipes.

Establishment of new plants within a population is presumably infre-
quent in this group of long-lived perennials. In three years of field work
I have seen only two first-year seedlings. After the first-year stage it is
difficut if not impossible to estimate the age of plants, so I have little
idea of the age-distribution of plants within populations.

TAXONOMIC TREATMENT
MIRABILIS L. subg. QUAMOCLIDION (Choisy) Jepson, Flora Calif.
1(4):457. 1974. — Ouamoclidion Choisy in DC. Prodr. 13(2):429.
1849. — Mirabilis sect. Ouamoclidion (Choisy) Gray, in Torr. Bot.
Mex. Bound. Surv. 173. 1859 [This combination has been erroneously
attributed to Heimerl, Nat. Pflanzenfam. 3(1b):24. 1889]. Type:
Mirabilis triflora Benth.

1978] PILZ: MIRABILIS 119

Hermidium S. Wats., Bot. King’s Exped. 5:286. 1871. Type: Hermidium
alipes S. Wats.

Mirabilis sect. Paramirabilis Heimerl, Notizbl. Bot. Gart. Berlin-Dahlem
11(106):453. 1932. Type: Heimerl (1932) included two species in
this section; I here designate as lectotype, Mirabilis multiflora (Torr.)
Gray.

Herbaceous perennials from a long spongy cylindrical taproot, the
stems branching alternately or subdichotomously from an expanded
multicipital caudex; branches erect or ascending to nearly decumbent,
the primary branches stout, more slender above, swollen at nodes; herb-
age green, occasionally purplish-red on surfaces exposed to direct sun-
light, densely pubescent to glabrous; leave opposite, petiolate, thin to
succulent, decurrent at base onto petiole, margins entire to repand; low-
ermost leaves orbicular to widely ovate; middle cauline leaves very
widely ovate to ovate; uppermost leaves ovate to lanceolate, highly re-
duced near involucre; leaf stomata anomocytic; peduncles axillary and
in terminal cymes, erect to abruptly reflexed; flowers involucrate, heads
3—16-flowered; involucral bracts connate to distinct, slightly accrescent
in age; perianth tubular, campanulate, or funnelform-salverform, ma-
genta to creamy white, constricted above the ovary, the limb 5-lobed;
stamens 5, circinate before anthesis, filaments unequal, capillary, pu-
bescent to glabrous, connate at base into a cup enveloping the ovary;
anther of paired locules, yellow, versatile, loculicidal; pollen grains
spheroidal, pantoporate, sexine spinulose, spinules 1 »m long; ovary
ellipsoidal, green, style filiform, glabrous, stigma capitate, papillose, ma-
genta; anthocarp ovoid to ellipsoidal, smooth to rugulose, glabrous to
sparsely pubescent; seed with light brown testa, adherent to pericarp;
embryo uncinate, perisperm enclosed on three sides by cotyledons and
bounded by hypocotyl and descending radicle on the other. Chromosome
number 7 = 33.

Involucrate heads 5—16-flowered; involucral bracts 12-35 mm long;

fruit 5-11 mm long.

Free portion of perianth campanulate.

Head erect on peduncle; leaves horizontally oriented; fruit 5-7 mm

long, olive-green, rugulose. . . - 2 2 Le alipes
Head pendent upon recurved peduncle; leaves conspicuously ascend-
ing; fruit 7-8 mm long, dark brown, smooth . 2. M. pudica.

Free portion of perianth funnelform-salverform.

Perianth 15-25 mm long; involucral bracts 15-20 mm long .

. . . 3. M. macfarlanei.
Perianth age 60 mm longs Peeieerl bracts 20-35 mm long.

Fruit 5-angled . . . . . . . « 4, MM. greene.
Fruit ovoid to subglobose, net anvaleale . 5. M. multiflora.
Involucrate heads 3-flowered; involucral bracts 7-10 mm long; fruit
7 MMylONS” 5 KPa we OS a eed oe 2h Oe Meio:

120 MADRONO [Vol. 25

1. MIRABILIS ALIPES (S. Wats.) Pilz, comb. nov.—Hermidium alipes
S. Wats., Bot. King’s Exped. 5:268. 1871. Type: Nevada, Humboldt
Valley, May 1868, S. Watson 968 (erroneously cited as “1860” and
“060” by Standley, 1909). Holotype: US! isotypes; GH!, NY!

Hermidium alipes S. Wats. var. pallidium Porter, Rhodora 54:158. 1952.
Type: Utah, Uintah Co., ‘on the Wasatch formation, 5 miles south of
Vernal’, 3 June 1950, C. L. Porter 5308. Holotype: RM!; isotypes:
CAS!) MO! RSA SMU! FE x wc)

Plants from taproot 1-2 m long, 3—5 cm in diameter; branches erect or
ascending to nearly decumbent, 2-4 dm high, forming hemispheric
clumps 6—8 dm in diameter, the primary branches stout, 4-6 mm thick,
more slender above; herbage pale green, glabrous to very sparsely puber-
ulent; leaves horizontally spreading, succulent, glaucescent; lowermost
leaves suborbicular to widely ovate and occasionally asymmetrical, 4—7
cm long, 3.5—-6.5 cm wide, base rounded, apex obtuse to rounded and
occasionally apiculate; middle cauline leaves ovate to widely ovate and
often asymmetrical, 4.5-5.0(-9) cm long, 3.5—4.0(—-5) cm wide, base
rounded, apex obtuse (rarely acute) and occasionally apiculate; upper-
most leaves narrowly ovate and symmetrical, 2.5—4.0 cm long, 1-2 cm
wide, base obtuse to rounded, apex acute; petioles often narrowly winged,
5—7 mm long on lower leaves, the uppermost leaves subsessile; leaf sto-
mata 30-36 wm long, 18-24 »m wide; peduncles 3-10 mm long, 1 mm in
diameter, erect to ascending; involucre consisting of distinct bracts, or
bracts united by their margins to one-half their length, the tube to 15 mm
long or lacking and the free portions 15-30 mm long, 10-25 mm wide,
ovate, apex acute to obtuse (rarely rounded), often apiculate; involucre
6—8 (—9)-flowered, the individual flowers borne on pedicels 0.2-1.0 mm
long on midvein of involucral bract, 1 flower on each midvein; perianth
campanulate, 15-16 mm long, magenta, occasionally creamy white, glab-
rous, tube 6—7 mm long, 5—6 mm broad, throat gradually widening, 5—6
mm long, the 5 lobes very widely ovate, 1-2 mm long, 7-10 mm broad,
apex emarginate, longitudinal nerves 5, extending to the sinus, apex of
nerve pubescent; stamens equalling perianth, filaments pubescent to
middle, glabrous above, pollen grains 106-133 wm in diameter; style
exserted 2-10 mm; anthocarp ellipsoidal with 10 slender tan longitudi-
nal ribs, slightly constricted at both proximal and distal ends, 5.5—7.0
mm long, 3.0-4.5 m wide, mottled olive-green, rugulose, glabrous, pro-
ducing a thick heavy mucilage when wetted.

DISTRIBUTION (Fig. 2). Scattered throughout much of Great Basin;
transmontane California (Inyo and Mono counties) across Nevada and
Utah to Rio Blanco Co., Colorado. On gravelly-sandy soil of mesas,
washes, and alluvial fans between 1200 and 2000 m in Shadscale Scrub
and Sagebrush Scrub, or, rarely, Pinyon-Juniper Woodland. Flowers
early May through mid-June.

1978]

PILZ: MIRABILIS

e WM. alipes

o M. pudica

oe Sa : a MM. macfarlanei

o 6M. «6greenei

p's a OM. triflora

O 500 cies C

Fie;. 2. Distribution and date of collection of Mirabilis,

122 MADRONO [Vol. 25

The status of Mirabilis alipes as a monotypic representative of a dis-
tinct genus, Hermidium, has always been based on the possession of dis-
crete involucral bracts (Heimerl, 1934; Standley, 1918). Although per-
fectly distinct bracts are typical, heads with the five outermost bracts
united by their margins to one-half their length are common. These in-
volucres are campanulate and have the aspect of the gamophyllous in-
volucres of other taxa within Quamoclidion. Flowers, leaves, and fruits
of M. alipes closely resemble those of other taxa within Quamoclidion.
Much more morphological variation occurs within Mirabilis (sensu
Standley, 1931b) than exists between Quamoclidion and Hermidium.
Thus the inclusion of M. alipes within subgenus Quamoclidion places it
with its closest relatives and represents the most “natural” classification
for these species.

I have observed several populations of plants in Pershing, Washoe,
and Nye counties, Nevada (Pilz 1113, 1127, 1128, 1276-79; all in UC),
which possess creamy white perianths and are referrable to Porter’s
(1952) var. pallidum from northeastern Utah. These occur, in this re-
gion, in populations with plants possessing purple perianths and a wide
range of intermediate colors. They do not appear distinct enough to war-
rant varietal status.

2. MIRABILIS PUDICA Barneby, Leafl. W. Bot. 3:175. 1942. Type: Barne-
by (1942) designated two cotypes; I here designate as lectotype
(CAS!): Nevada, Lincoln Co., “3 miles W. of Crystal Springs”, 10
May 1942, Ripley and Barneby 4403. Isotypes: GH!, K!, NY!,
RSA!, UC!

Plants from a taproot 1-2 m long, 3—5 cm in diameter; branches erect
or ascending, 3—5(-—6) dm high, forming upright columnar clumps 3-5
dm in diameter, the primary branches 3-5 mm thick, more slender
above; herbage pale green to whitish, glabrous to densely pubescent;
leaves conspicuously ascending, very often vertically oriented, succulent,
glaucescent; lowermost leaves ovate to widely ovate and symmetrical,
1.5-3.0 cm long, 1.0—-2.5 cm wide, base obtuse to acute, apex acute to
rounded and apiculate; middle cauline leaves ovate to narrowly ovate
and symmetrical, 3.5-5.0 cm long, 2.0-2.5 cm wide, base acute to
rounded (rarely cordate), apex acute; uppermost leaves narrowly ovate
to lanceolate and symmetrical, 1.4—3.7 cm long, 0.5—1.8 cm wide, base
rounded to acute, apex acute; petioles short, often narrowly winged, 3—5
mm long on lowermost leaves, the uppermost leaves subsessile; leaf sto-
mata 27-36 um long, 24-30 um wide; peduncles 6-10 mm long, 1 mm
in diameter, abruptly reflexed to produce a pendent involucre; involucre
campanulate to nearly rotate in fruit, 12-21 mm long, the bracts united
by their margins, the tube 5-10 mm long, unequally 5-lobed, free portion
of bract 6-12 mm long, 9-17 mm wide at base, triangular to widely ovate,
apex acute, apiculate; involucre 6-flowered, the flowers borne on pedicels
0.2-1.0 mm long, on midvein of involucral bracts, 1 flower on each mid-

1978] PILZ: MIRABILIS 123

vein, the central flower solitary and naked (rarely subtended by a dis-

tinct bract) ; perianth campanulate, 12-14 mm long, creamy white, glab-

rous to densely pubescent, the tube 5-6 mm long, 6-7 mm broad, the
throat gradually widening, 5—6 mm long, the five lobes very widely ovate,

2-3 mm long, 6-9 mm broad, apex emarginate, the five longitudinal

nerves extending to the sinus, apex of nerve densely pubescent; stamens

exserted 2-4 mm, filaments villous to middle, glabrous above, pollen
grains 110-145 »m in diameter; style exserted 3-5 mm; anthocarp ellip-
soidal to widely ellipsoidal with 10 slender tan ribs occasionally evident

at base, occasionally constricted at both proximal and distal ends, 7—8

mm long, 4.5—5.0 mm wide, dark brown, smooth, glabrous, not producing

mucilage when wetted. _

DISTRIBUTION (Fig. 2). On calcareous alkaline hills and sandy playas
of Lincoln and Nye counties, Nevada, in Shadscale Scrub and Creosote
Bush Scrub between 1000 and 1500 m. Flowers early May to mid-June.

This is a distinctive species within Quamoclidion because of its heads,
which are pendent upon recurved peduncles, and its conspicuously as-
cending, often vertically oriented, leaves.

3. MIRABILIS MACFARLANEI Constance & Rollins, Proc. Biol. Soc. Wash.
49:148. 1936. Type: Oregon, Wallowa Co., ‘““Lower Cottonwood Land-
ing, between mouth of Somer’s Creek and Pittsburg Landing, Snake
River Canyon”, 15 May 1936, Constance, Rollins, Clements and Dil-
lon 1579. Holotype: WS!; isotypes: CAS!, DS-2 sheets!, GH!,
JEPS!, K!, MO!, NY!, POM-2 sheets!, RM!, UC-2 sheets!, US-2
sheets!, WIS!, WS!

Plants from taproot 1-2 m long, 3-5 cm in diameter; branches erect
or ascending to nearly decumbent, 6—-8(—10) dm high, forming hemi-
spheric clumps 8—10(—12) dm in diameter, the primary branches stout,
6-8 mm thick, more slender above; herbage green, glabrous to sparsely
puberulent; leaves horizontally spreading, succulent, glaucescent on low-
er surface; lowermost leaves orbicular to very widely ovate, 2.0—5.5 cm
long, 2.5—-6.5 cm wide, base obtuse to cordate and often asymmetrical,
apex rounded to broadly obtuse (rarely short apiculate); middle cauline
leaves suborbicular to widely ovate, 3.5—4.5 cm long, 3.0—-4.5 cm wide,
base obtuse to cordate and symmetrical, apex obtuse to rounded (rarely
acute) and often short apiculate; uppermost leaves ovate and symmet-
rical, 1.5-3.0 cm long, 1.0—-2.5 cm wide, base rounded to obtuse (rarely
cordate), apex acute; petioles stout, 17-25 mm long on lowermost leaves,
the uppermost leaves subsessile; leaf stomata 27-30 wm long, 21-24 ym
wide, peduncles 4—-8(—25) mm long, 1 mm in diameter, erect to ascend-
ing; involucre campanulate to nearly rotate in fruit, 13-20 mm long, the
bracts united by their margins, the tube 6-12 mm long, unequally 5-
lobed, the free portion of bract 6-12 mm long, 11-14 mm wide at base,
triangular to very widely ovate, the apex acute to acuminate; involucre
6-flowered, the flowers borne on pedicels 1-3 mm long, on midvein of

124 MADRONO [Vol. 25

involucral bracts, 1 flower on each midvein, the central flower solitary
and naked (rarely subtended by a distinct bract); perianth funnelform-
salverform, 15-25 mm long, magenta, glabrous, the tube 7-12 mm long,
3-4 mm broad, the throat gradually widening, 5-7 mm long, the five
lobes very widely ovate, 2-3 mm long, 6-9 mm broad, the apex emargi-
nate, the five longitudinal nerves extending to the sinus, apex of nerve
densely pubescent; stamens exserted 3-4 mm, filaments glabrous, pollen
grains 115—122um in diameter; style exserted 4-5 mm; anthocarp el-
lipsoidal with 10 slender tan ribs, constricted at the proximal and often
the distal end, 6-7 mm long, 3-4 mm wide, brown to dark brown, slight-
ly tuberculate, glabrous, mucilaginous when wetted.

DISTRIBUTION (Fig. 2). On dry exposed slopes bordering the Snake
and Salmon rivers of Oregon and Idaho, between 450 and 500 m. Flowers
throughout May.

Mirabilis macfarlanei is most notable for its geographic isolation in the
Snake and Salmon river canyons. These are much more arid than the
surrounding canyon walls and mountain slopes, which support Yellow
Pine Forest. The closest relative of this species is probably Murabilis
multiflora var. glandulosa with which it shares a funnelform-salverform
perianth, general anthocarp characteristics, and a similar though more
depauperate flavonoid profile (Pilz, 1974).

This taxon is named for E. B. MacFarlane, for 30 years a pilot of
boats on the Snake River, who pointed out the location of the plant to
the authors. It has been said that Harold St. John had pointed out the
plant to MacFarlane on an earlier trip up the river, but St. John did not
publish his find (Constance, pers. comm.).

4. MIRABILIS GREENEI S. Wats., Proc. Amer. Acad. Arts 12:253. 1876
(1877). — Quamoclidion greenei (S. Wats.) Standley, Contr. U. S.
Natl. Herb. 12:358. 1909. Type: California, Siskiyou Co., “mountain
sides about Yreka”, 20 June 1876, E. L. Greene 876. Holotype: GH!
mounted with V. Rattan 56, June 1884; isotype: NY! mounted with
T. Howell 1389, July 1889.

Plants from taproot 2-4 m long, 3—-5(—12) cm in diameter; branches
erect or ascending to nearly decumbent, 4-8 dm high, forming hemi-
spheric clumps 6-10 dm in diameter, the primary branches stout, 5-13
mm thick, more slender above; herbage green, glabrous to very sparsely
puberulent; leaves horizontally spreading, succulent, glaucescent; lower-
most leaves orbicular to suborbicular, 3.0-5.5 cm long, 2.5—5.0 cm wide,
base obtuse to cordate and often asymmetrical, apex rounded to obtuse;
middle cauline leaves widely elliptic to ovate, 4.0-7.5 cm long, 3.0—4.5
cm wide, base obtuse and often asymmetrical, apex acute or acuminate;
uppermost leaves narrowly ovate to elliptic and symmetrical, 2.7—4.5 cm
long, 1.0-2.5 cm wide, acute at base and apex; petioles stout, 10-27 mm
long on lowermost leaves, the uppermost leaves subsessile; leaf stomata
30-36 um long, 24-30 »~m wide; peduncles 25-85 mm long, 1.5—-2.0 mm

1978] PILZ: MIRABILIS 125

in diameter, erect or ascending; involucre campanulate, 26-36 mm long,
the bracts united by their margins, the tube 13-20 mm long, unequally
5-lobed, the free portion of bract 10-17 mm long, 10-13 mm wide at
base, triangular to widely ovate, the apex acute to acuminate; involucre
6(—16)-flowered, the flowers borne on pedicels 1—4 mm long, on midvein
of involucral bract, as many as 3 flowers borne along a single bract, the
central flower solitary and naked (rarely subtended by a distinct bract) ;
perianth funnelform-salverform, 40-50 mm long, magenta, glabrous, the
tube 25-30 mm long, 5—7 mm broad, the throat gradually widening, 12-
15 mm long, the five lobes very widely ovate, 3-5 mm long, 10-14 mm
broad, the apex emarginate, the five longitudinal nerves extending to and
1 mm beyond the sinus, apex of nerve densely pubescent; stamens ex-
certed 1-5 mm, filaments glabrous, pollen grains 130-150 um in diameter;
style exserted 3-5 mm; anthocarp widely obovoid to widely ellipsoidal,
5-angulate, constricted at both proximal and distal ends, 7.0—-7.5 mm
long, 4.0-4.5 mm wide, light brown, tuberculate, very sparsely puberu-
lent to glabrous, mucilaginous when wetted, the mucilage most abundant
on ribs (10 visible when wet).

DISTRIBUTION (Fig. 2). Scattered on eastern flank of North Coast and
Klamath ranges from Colusa Co. to Siskiyou Co., California. Growing on
steep talus slopes and gravelly flats with junipers from 400 to 1000 m.
Flowers early May to mid-June.

Mirabilis greene is easily recognizable within Quamoclidion because of
its distinctive 5-angled fruit, an involucre that often surrounds more than
six flowers, and a novel flavonoid profile (Pilz, 1974).

5. MIRABILIS MULTIFLORA (Torr.) Gray, im Torr. Bot. Mex. Bound.

Surv. 173. 1859.

Plants from taproot 1-2 m long, 2—5 cm in diameter; branches erect or
ascending to nearly decumbent, forming hemispheric clumps 6-8 dm in
diameter, the primary branches stout, 5-12 mm thick, more slender
above; herbage green, densely pubescent to glabrous; leaves horizontal-
ly spreading, succulent, glaucescent; lowermost leaves orbicular to very
widely ovate and often asymmetrical, (3—)5—-12 cm long, (4—)5-15 cm
wide, base rounded to cordate, apex rounded to obtuse, occasionally mu-
cronate; middle cauline leaves ovate to very widely ovate and often
asymmetrical, 5-10 cm long, 4-8 cm wide, base cordate to rounded, apex
obtuse to acuminate and often apiculate; uppermost leaves widely ovate
to narrowly ovate and only slightly asymmetrical, 2-7 cm long, 1-5 cm
wide, base cordate to obtuse, apex acute to acuminate and often apicu-
late; petioles slender to stout, 20-40 mm long on lowermost leaves, the
uppermost leaves subsessile; leaf stomata 24-36 mm long, the bracts
peduncles 4-75 mm long, 1-2 mm in diameter, erect to ascending; in-
volucre campanulate to broadly campanulate, 22-35 mm long, the bracts
united by their margins, the tube 11-25 mm long, unequally 5-lobed, the
free portion of bract 6-14 mm long, 8-17 mm wide at base, triangular to

126 MADRONO [Vol. 25

very widely ovate, apex acute to rarely obtuse in fruit, apiculate; invo-
lucre 6-flowered, the flowers borne on pedicels up to 2 mm long on mid-
vein of involucral bracts, one flower on each midvein, the central flower
solitary and naked (rarely subtended by a distinct bract); perianth fun-
nelform-salverform, 40-60 mm long, magenta, occasionally the tube
green, puberulent to glabrous, the tube 27-40 mm long, 5-10 mm broad,
the throat gradually widening, 10-20 mm long, the five lobes very widely
ovate, 2-7 mm long, 10-20 mm broad, apex emarginate, the five longi-
tudinal nerves extending to the sinus, apex of nerve densely pubescent;
stamens exserted 1-10 mm, filaments glabrous to pubescent, pollen grains
118-150 «wm in diameter; style exserted 3-13 mm; anthocarp ellipsoidal
to widely ellipsoidal, often constricted at both proximal and distal ends,
6-11 mm long, 4.0-5.5 mm wide, brown with 10 slender tan longitudinal
ribs alternating with 10 raised dark brown (often interrupted) ribs to
nearly solid black, rugulose to smooth, glabrous to pubescent, the muci-
lage production diverse.
Mirabilis multiflora may be divided into three varieties as follows:
Fruit smooth to only slightly tuberculate, producing no mucilage when
wetted; involucral bracts acute.
Fruit dark brown to black, the ribs inconspicuous
ae 5a. M. multiflora var. multiflora:
Fou fen pron with 10 slender tan longitudinal ribs asa: with
10 brown (often interrupted) ribs .
5b. M. pve var. We Besceral
Fruit deamiae: eipereniarey producing mucilage when wetted; involucral
bracts obtuse . . . . . . 5c. M. multiflora var. glondulosa.
The nature of the mature fruit is the most consistent character for dis-
tinguishing the varieties of Mirabilis multiflora. There are many recog-
nizable anthocarp types that have discrete geographic limits, but these
fall into three major groups as indicated by the key to varieties. In some
areas where the varieties of M. multiflora occur together they are quite
distinct morphologically, as in Colorado National Monument, Mesa
County, Colorado. In this region var. glandulosa has obtuse bracts and
the apices of most leaves are obtuse to rounded, while var. multiflora has
narrowly acute bracts and leaf apices. In addition var. glandulosa flow-
ers in May and June while var. multiflora usually flowers in July and
August. In contrast the plants of southwestern Utah and northwestern
Arizona show a collage of characteristics normally typical of the different
varieties.
5a. MIRABILIS MULTIFLORA (Torr.) Gray var. con Orie
multiflorus Torr., Ann. Lyceum Nat. Hist. New York 2:237. 1827.
— Allionia multiflora (Torr.) Eaton, Man. Bot. ed. 5 Addenda: 2.
1829. — Nyctaginia? torreyana Choisy, in DC. Prodr. 13(2):430.
1849, illegitimate superfluous name. — Quamoclidion multiflorum
(Torr.) Torr. ex Gray, Amer. J. Sci. Arts II. 15:321. 1853. Type:

1978] PILZ: MIRABILIS 127

“Forks of the Platte” (label), ““About the Forks of the Platte” (pro-
tologue), 1820, Dr. E. James s.n. Holotype: NY! ; isotype: K!

DISTRIBUTION (Fig. 3). Occasional in Chihuahua, Coahuila, Nuevo
Leon, and San Luis Potosi, Mexico. More often collected in western Tex-
as, through New Mexico, Arizona, and Colorado, United States of Ameri-
ca. On gravelly-sandy or loose soils of mesas, washes, and open hillsides
between 300 and 2300 m in Pinyon-Juniper Woodland and Yellow Pine
Forest. Flowers mid-May through mid-October.

Among the plants collected by Edwin James, M.D., Assistant Surgeon
in the United States Army, in the summer of 1820, was a member of
Nyctaginaceae collected ‘About the Forks of the Platte” (Torrey, 1827).
The specimen is incomplete and Torrey (1827) stated that the “country
was traversed with great rapidity, .. . and little opportunity was afforded
of making observations, or even of recording all the stations of the
plants.” This collection was published as Oxybaphus multiflorus by Tor-
rey. After crossing the Platte River at Forks (now Lincoln County, Ne-
braska, fide McKelvey, 1955) the party, commanded by Major Stephen
H. Long, continued up the South Platte River to the base of the Rocky
Mountains. I have not seen plants belonging to Torrey’s O. multiflorus
from Nebraska. The closest locality lies over 400 km southwest in Pueblo
County, Colorado. Given Torrey’s remark that James’ records were not
always complete, it seems probable that the collection of O. multiflorus
was made later in James’ journey, perhaps somewhere on the eastern
flank of the Rocky Mountains where the plants are known to occur.
5b. MIRABILIS MULTIFLORA (Torr.) Gray var. PUBESCENS S. Wats., Bot.

Calif. 2:2. 1880. Type: In the protologue Watson stated, ‘The varie-
ty is peculiar to S. California, from near Fort Tejon (Wallace, Ken-
nedy) to San Diego County, Palmer.” Of the three collections cited
by Watson only Wallace’s is at the Gray Herbarium (GH), and it
consists solely of two detached involucres and flowers. This specimen
says simply “California, Wallace”. On the same sheet there are two
other collections of Mirabilis multiflora var. pubescens. The label of
one of these, W. Matthews, 1877, from Owen’s Valley, is in Watson’s
handwriting, and he has determined the specimen to be M. multiflora
var. pubescens. The only specimen of the Kennedy collection I have
seen is at Field Museum (F), and there is no indication on the sheet
that Watson saw this specimen. The label of the Palmer specimen at
New York Botanical Garden (NY) is inscribed in Watson’s hand-
writing “Mirabilis multiflora” but it is not designated var. pubescens.
In addition the specimen is glabrate. I therefore designate as lecto-
type (GH! ): California, Wallace s.n.

Oxvbaphus froebelii Behr, Proc. Calif. Acad. Sci. 1:69. 1855. — Mira-
bilis froebelii (Behr) Greene, Bull. Calif. Acad. Sci. 1:124. 1885. —
Mirabilis multiflora (Torr.) Gray var. froebelit (Behr) Jones, Contr.
W. Bot. 10:49. 1902 (illegitimate superfluous name, since Jones cited

128 MADRONO [Vol. 25

“var. pubescens Wats.” as a synonym). — Quamoclidion froebelii
(Behr) Standley, Contr. U. S. Natl. Herb. 12:359. 1909. Type: Cali-
fornia, San Diego Co., “Warner’s Ranch” (protologue), 1855, J.
Froebel, not seen, perhaps destroyed at CAS in the fire of 1906.

Quamoclidion froebelii (Behr) Standley ssp. glabratum Standley, Contr.
U.S. Natl. Herb. 12:360. 1909. — Mirabilis froebelii (Behr) Greene
var. glabratum [sic] (Standley) Jepson, Flora Calif. 1(4) :458. 1914.
Type: California, San Bernardino Co., “Providence Mts.”, 25 May
1902, T. Brandegee s.n. Holotype: UC!; isotype: NY!

DISTRIBUTION (Fig. 3). Northern Baja California, Mexico, through
southern California, southeastern Nevada, southwestern Utah, and west-
ern Arizona, United States of America. On dry gravelly-sandy soil of
mesas, washes, and open hillsides between 50 and 2100 m in Oak Wood-
land, Pinyon-Juniper Woodland, Chaparral, Sagebrush Scrub, Creosote
Bush Scrub, and Shadscale Scrub. Flowers late April through July.
5c. MIRABILIS MULTIFLORA (Torr.) Gray var. GLANDULOSA (Standley)

Macbr., Contr. Gray Herb. 49:49. 1917. — Quamoclidion multiflo-
rum (Torr.) Torr. ex Gray ssp. glandulosum Standley, Contr. U. S.
Natl. Herb. 12:359. 1909. Type: Colorado, “Grand Junction, dry
mesa’, 12 May 1894, C. Crandall 423. Holotype: US!; isotypes:
MO!, NY!, RM!

Quamoclidion multiflorum (Torr.) Torr. ex Gray ssp. obtusum Standley,
Contr. U. S. Natl. Herb. 12:359. 1909.— Mirabilis multiflora (Torr.)
Gray var. obtusa (Standley) Macbr., Contr. Gray Herb. 49:49. 1917.
Type: Nevada, ‘Kernan, rocky ledges”, 29 April 1902, L. Goodding
653. Holotype: RM!; isotypes: F!, GH!, MO!, NY!, POM!,
Vel wus!

Quamoclidion cordifolium Osterh., Bull. Torrey Bot. Club 55:75. 1928.
Type: Colorado, Mesa Co., “six miles from Grand Junction, in the
hills across the Colorado River” (protologue), ‘Hills across the Colo-
rado River from Grand Junction” (label), 18 June 1926, G. Oster-
hout 6559. Holotype: RM!; isotypes: GH!, NY!, POM!, RM-3
sheets!

Note: two varietal names are available for this taxon, var. obtusa and
var. glandulosa. The latter name has been chosen because Crandall 423
is a more representative collection for my circumspection of the taxon,
since it possesses mature fruits on some sheets while Goodding 653 has
only immature fruits.

DISTRIBUTION (Fig. 3). Scattered from Inyo Co., California, across
southern Nevada and Utah to western Colorado. On gravelly-sandy soil
of mesas, washes, and open hillsides between 900 and 2500 m in Sage-
brush Scrub, Shadscale Scrub, and Pinyon-Juniper Woodland. Flowers
May through July.

6. MIRABILIS TRIFLORA Benth., Pl. Hartw. 23. 1839. — Quamoclidion

nyctagineum Choisy, in DC. Prodr. 13(2):429. 1849 (superfluous

1978] PILZ: MIRABILIS 129

e var. multiflora
var. pubescens

var. glandulosa

Fic. 3. Distribution and date of collection of Mirabilis multiflora.

name). — Quamoclidion triflorum (Benth.) Standley, Contr. U. S.

Natl. Herb. 12:358. 1909. Type: Jalisco, ‘‘Bolafos”, 1837, T. Hart-

weg 197. Holotype: K!; isotypes: G-2 sheets!, GH!, P!; photographs

of holotype: DS!, MICH!

Plants from taproot 1-2 m long, 2—5(—8) cm in diameter; branches
ascending to nearly decumbent, 1—5 m long, forming loose trailing clumps
I—6 m in diameter, the primary branches 5—7 mm thick; herbage green,
sparsely to densely glandular-pubescent, occasionally glabrate; leaves
horizontally spreading, thin; lowermost leaves ovate and often asym-
metrical, 3-9 cm long, 2—6 cm wide, base cordate, apex acute to attenu-
ate; middle cauline leaves similar to lowermost leaves in all respects;
uppermost leaves narrowly ovate to ovate and only slightly asymmetri-
cal, 1-3 cm long, 0.5-1.5 cm wide, base rounded to cordate, apex acute

130 MADRONO [Vol. 25

to attenuate; petioles slender, 20-30 mm long on lowermost leaves, 5-10
mm long on uppermost leaves; leaf stomata 21-27 pm long, 15-21 um
wide; peduncles 2-15 mm long, 1 mm in diameter, erect or ascending;
involucre campanulate, laterally compressed, 7-10 mm long, the bracts
united by their margins, the tube 3-5 mm long, very unequally 5-lobed,
the free portion of bract 3-7 mm long, 3—5 mm wide at base, triangular
to very widely ovate, apex acute to attentuate; involucre 3-flowered, the
flowers borne on pedicels up to 1 mm long on midvein of the two largest
involucral bracts only, one flower on each midvein, the central flower
solitary and naked; perianth tubular, 20-25 mm long, deep cardinal red,
puberulent on both surfaces, the tube 20-25 mm long, 4-5 mm wide, a
throat lacking, the five lobes very widely ovate, 1 mm long, 2.0—-2.5 mm
broad, apex rounded, the five longitudinal nerves extending to the sinus,
apex of nerve densely pubescent; stamens exserted 7-14 mm, filaments
densely pubescent, pollen grains 115-135 um in diameter; style exserted
7-14 mm; anthocarp ellipsodial, slightly constricted at proximal end,
occasionally constricted at distal end, 4-5 mm long, 2.0-2.5 mm wide,
dark brown, tuberculate, sparsely pubescent, producing copious amounts
of clear mucilage when wetted.

DISTRIBUTION (Fig. 2). Scattered in scrub vegetation of Baja Califor-
nia del Sur and Jalisco, Mexico, occurring at 300 to 1200 m. Flowers
October through April.

This is certainly the most remote taxon, both geographically and taxo-
nomically, of subgenus Quamoclidion. Heimer] (1934) proposed that
Quamoclidion be resurrected as a genus composed solely of Mirabilis
triflora. He separated Quamoclidion from Mirabilis, as a genus, on the
following basis: 4-lobed versus 5-lobed involucre and filaments with
spreading hairs versus glabrous filaments. He further stated that only
Hermidium in subtribe Mirabileae-Boerhaaviinae had similarly pubes-
cent filaments. In fact, many taxa of Mirabilis have pubescent filaments
and one of the four involucral bracts of M. triflora is deeply bifid and
supplied by two vascular traces rather than the usual single midvein. Al-
though M. triflora is quite distinctive within Quamoclidion, I prefer to
keep it here until the other members of Mirabilis are better known, since
I have not seen any Mirabilis that resembles M. triflora more than do
plants of subgenus Ouamoclidion.

EXCLUDED NAMES
Quamoclidion angulatum Choisy, in DC. Prodr., 13(2):429. 1849 (Nyc-
tago angulata DC.; Choisy, in DC. Prodr. 13(2):429, as synonym.
1849.) Choisy doubtfully referred this species to the genus. Choisy’s
specimens for this taxon were collected in Mexico by Mocifo and
Sessé. Standley (1911) remarks that Choisy’s description does not
agree with the collectors’ drawing of the specimen in all particulars,
and Standley (1918) later concluded that “the identity of the plant is

1978] PILZ: MIRABILIS 131

problematical.” I have not been able to locate the collection made by
Mocifio and Sessé, but I have seen a copy of the collectors’ drawing
(MO). The plant represented may belong to Mirabilis section Allionia,
but it is definitely not a Quamoclidion.

Quamoclidion laeve (Benth.) Rydberg, Bull. Torrey Bot. Club 29:687.
1902, = Mrrasitis LAEvis (Benth.) Curran, Proc. Calif. Acad. IT.
1255). 1855.

Quamoclidion oxybaphoides Gray, Amer. J. Sci. Arts IT. 15:320, 1853. ==
MIRABILIS OXYBAPHOIDES (Gray) Gray, in Torr. Bot. Mex. Bound.
Surv. 173. 1859.

ACKNOWLEDGEMENTS

I am grateful to Lincoln Constance for his advice and numerous valid criticisms
during all phases of this study. I also thank Robert Ornduff and William Libby for
criticizing portions of this study. Special thanks go to John L. Strother for his help
as well as his companionship in the field. Thanks are due to the curators of the fol-
lowing herbaria for loans of specimens: A, ARIZ, BM, CAS, COLO, DS, ENCB, F,
G, GH, IA, JEPS, K, MEXU, MICH, MO, NY, OKLA, P, POM, RM, RSA, SMU,
TEX, UC, US, WIS, and WS. This research was supported in part by NSF grant
GB 36647.

LITERATURE CITED
Baker, H. G. 1961. The adaptation of flowering plants to nocturnal and crepuscular
pollinators. Quart. Rev. Biol. 36:64—73.
Barnesy, R. C. 1942. A new species of Mirabilis, with remarks on Hermidium and
related genera. Leafl. W. Bot. 3(8) :175-179.
Cooper, D. C. 1949. Flower and seed development in Oxybaphus nyctagineus. Amer.
J. Bot. 36:348-355.
CrupEn, R. W. 1970. Hawkmoth pollination of Mirabilis (Nyctaginaceae). Bull.
Torrey Bot. Club 97:89-91.
. 1973. Reproductive biology of weedy and cultivated Mirabilis (Nyctagi-
naceae). Amer. J. Bot. 60:802—809.
Eprinc, C., H. Lewis, and P. H. Raven. 1962. Chromosomes of Salvia: section
Audibertia. Aliso 5:217-221.
HEIMERL, A. 1889. Nyctaginaceae. Jn A. Engler and K. Prantl, Die natiirlichen
Pflanzenfamilien. III. 1b:14—32.
. 1897 Beitrage zur Systematik der Nyctaginaceen. Jahresberichte der k.k.
Staats-Oberrealschule, Fiinfhaus, Wien 23:1—40.
. 1934. Nyctaginaceae. Jn A. Engler and K. Prantl, Die natiirlichen Pflanzen-
familien. ed. 2. 16c:86-134.
Jackson, B. D. 1928. A glossary of botanical terms; with their derivation and ac-
cent. ed. 4. Gerald Duckworth, London.
McKetvey, S. D. 1955. Botanical exploration of the trans-Mississippi west, 1790-
1850. Arnold Arboretum, Jamaica Plain.
NowickE, J. W. 1970. Pollen morphology in the Nyctaginaceae. I. Nyctagineae
(Mirabileae). Grana Palynol. 10:79-88.
Piz, G. E. 1974. Systematics of Mirabilis subgenus Quamoclidion (Nyctaginaceae) .
Ph.D. thesis, University of California, Berkeley.
Porter, C. L. 1952. Novelties in Hermidium (Nyctaginaceae) and Astragalus (Le-
guminosae) from eastern Utah. Rhodora 54:158-161.
Reep, C. F. 1960. Contributions toward a flora of Nevada. No. 48. Nyctaginaceae.
U.S. Bureau of Plant Industry, Washington, D. C.

132 MADRONO [Vol 25.

SHowaLtTeER, H. M. 1935. A study of the chromosomes and degenerating microspore
tissue in Mirabilis. Amer. J. Bot. 22:594-608.
STANDLEY, P. C. 1909. The Allioniaceae of the United States with notes on Mexican
species. Contr. U. S. Natl. Herb. 12:303-389.
. 1911. The Allioniaceae of Mexico and Central America. Contr. U. S. Natl.
Herb. 13:377-430.
. 1918. Allioniaceae. North American Flora. 21(3) :171-254.
. 1931a. The Nyctaginaceae and Chenopodiaceae of northwestern South
America. Field Mus. Nat. Hist., Bot. Ser. 11(3):71-126.
. 1931b. Studies of American Plants — V. Field Mus. Nat. Hist., Bot. Ser.
8(5) :293-398.
Torrey, J. 1827. Some account of a collection of plants made during a journey to
and from the Rocky Mountains in the summer of 1820, by Edwin P. James,
M. D. Assistant Surgeon U. S. Army. Ann, Lyceum Nat. Hist. New York
2:161-254.

NOTES AND NEWS

RANUNCULUS CALIFORNICUS, A NEW RECORD FOR THE STATE OF WASHINGTON. —
Known as an indicator species for coastal prairies, Ranunculus californicus is typi-
cally found on the islands in southern California and along the coast in northern
California and southern Oregon. Until now, there were apparently only two locality
records north of Lincoln County, Oregon: one disjunction from Sauvie Island, near
Portland, Oregon (ORE), and one from the Trial Islands, near Victoria, British
Columbia (V). Recent exploration has resulted in documentation of seven sites in
the Puget Trough of Washington where Ranunculus californicus Bentham var. cune-
atus Greene is locally common to abundant. All sites are open, south to southwest
facing grassy bluffs or rocky slopes just above the seacoast, at elevations up to 150
feet.

Although R. californicus intergrades with R,. occidentalis (L. Benson, Amer. Mid].
Naturalist 40:1-261, 1948), the features demonstrated by the coastal collections
from the Puget Trough are most similar to R. californicus in these diagnostic fea-
tures: petals 5-14, 2—2.5 times longer than wide; nectary glands 0.5-0.8 mm long;
and beaks of the achenes 0.5-1 mm long, strongly curved or hooked at the apex.
Ranunculus occidentalis does occur in nearby areas but not sympatrically with R.
californicus. The populations show differences in the number of petals per flower
and in the proportion of petal length to width. In the populations from Fidalgo and
Lopez Islands, the petals number (5—) 8-14 per flower and are typically 2—2.5 times
longer than wide. On San Juan Island, the petals usually number 5-8 (—11) and are
about two times longer than wide. The San Juan Island populations show greater
similarity to R. occidentalis than do other populations. Further study of this com-
plex is needed to explain the interpopulational variation.

Specimens of R. californicus var. cuneatus from Washington, all at WTU: San
Juan Co. Lopez Island: Iceberg Point, 8 May 1974, Denton 3420, 17 Apr 1976,
Denton 3975; Point Colville, 17 Apr 1976, Denton 3802. San Juan Island: west
slope of Mt. Dallas, 5 May 1974, Denton 3407, 1 May 1976, Elvander 602; San Juan
County Park, west side of island, 6 May 1976, Lerner 152; English Camp, north-
west side of island, 6 May 1976, Lerner 151. Skacir Co. Fidalgo Island: Deception
Pass, 17 Apr 1976, Denton 3814; Fidalgo Head, 17 Apr 1976, Denton 3813.

I am grateful to R. E. Norris who informed me of an “unusual” buttercup on
Iceberg Point of Lopez Island, to C. L. Hitchcock for verifying my identifications,
and to R. L. Taylor at UBC and the curators at ORE, OSC, UBC, WS, and V for
information about their collections of R. californicus. — Mrtinpa F. Denton, De-
partment of Botany, University of Washington, Seattle 98195.

LEAF ANGLE AND LIGHT ABSORPTANCE OF
ARCTOSTAPHYLOS SPECIES (ERICACEAE)
ALONG ENVIRONMENTAL GRADIENTS

Gatus R. SHAVER

Department of Biological Sciences, Stanford University,
Stanford, CA 943051

The purpose of this research was to test the hypothesis that changes in
leaf energy balance characteristics are correlated with the distribution of
Arctostaphylos Adans. species along environmental gradients. It was ex-
pected that species from low light, low temperature environments should
hold their leaves more horizontally and absorb more light than leaves of
species from warmer, high insolation areas. Similar results have been re-
ported by Billings and Morris (1951) for leaf reflectance and by Moo-
ney et al. (1971) for leaf angle distributions of groups of shrubs from
different habitats. The present research differed from these studies in
_ that it involved a more detailed comparison within a single genus.

There are over fifty recognized taxa of Arctostaphylos, nearly all of
them confined to the west coast of North America. All are woody ever-
green shrubs with flat leaves (Munz 1959, Adams 1949). About half of
the species are highly localized endemics (Stebbins and Major 1965).
Seven species of Arctostaphylos were studied in the field at 54 sites in
California, Oregon, and Washington. The sites and species were chosen
for their distribution along two macroclimatic gradients of air tempera-
ture and insolation. These climatic gradients were associaited with eleva-
tional changes in the Sierra Nevada of California, and with latitude and
distance from the seacoast in northern California, Oregon, and Washing-
ton. Climatic information was obtained from the Climatic Atlas of the
United States (U.S. Environmental Sciences Administration, 1968), the
U.S. National Atlas, and Climates of the States (U. S. NOAA, Dept. of
Commerce, 1973).

METHODS

Leaf angles were measured using a protractor to which a plumb line
was attached at the origin (Kvet and Marshall 1971). The protractor
was held up to the leaf and the angle between the leaf and the horizontal
was measured to the nearest 5°. For these flat leaves, angle to the hori-
zontal was considered to be the steepest angle that could be measured
and did not take into consideration the azimuth angle, leaf-branch angle,
etc. This method was found to give rapid, highly repeatable results. Angle
measurements were made throughout the canopy of the shrubs examined,

Ipresent address: Systems Ecology Research Group, San Diego State University,
San Diego, CA 92182.

133

134 MADRONO [Vol. 25

usually 10-20 leaves from each of several branches per shrub, 3-10
shrubs per site.

Reflectance and transmission of Arctostaphylos leaves were measured
using a Zeiss PMQII spectrophotometer with reflecting sphere attach-
ment. Absorptance was then calculated as: Absorptance = 1 — (Reflec-
tance + Transmittance). All values were calculated as percentages, ei-
ther of reflectance from a standard MgCO; block (reflectance) or of
transmittance measurements taken with no obstructions between light
source and sensor (transmittance). Measurements between 325 and 800
nm were taken using a photomultiplier tube as a sensor, while a PbS pho-
tocell was used between 700 and 2500 nm, with a 100 nm overlap be-
tween sensors. Variation among leaves at a single wavelength was rarely
greater than 5%, and repeated measurements of the same leaf varied by
less than 1%. In most cases, reflectance and transmittance were meas-
ured for at least six leaves of each species at each wavelength.

RESULTS AND DISCUSSION

Figure la and Table 1 summarize the results of leaf angle observations
at 29 sites between Yosemite National Park and Truckee, California.
The data show that A. nevadensis, the species from the highest elevation
sites, has many fewer vertical or near vertical leaves than A. viscida or
A. mariposa, the two lowest elevation species. Arctostaphylos patula, the
species from middle elevations in the Sierra Nevada, is intermediate in
its leaf angle distribution.

Similar results are recorded in Figure 1b. Arctostaphylos uva-ursi,
which is circumboreal in its distribution, was measured at three sites in

Mason and Kitsap counties, Washington, and Curry County, Oregon. On —
these sites, A. wva-ursi grows between 10 and 100 m elevation, rarely —
more than 1 km from the ocean. Arctostaphylos columbiana, which was |

sampled at the same three sites plus another in Humboldt County, Cali-

fornia, grows above A. uva-ursi to 800 m in the Pacific coast ranges. |

Arctostaphylos manzanita grows mainly in the interior coast ranges and
was sampled in Humboldt and Mendocino Counties, California, at two

sites about 50 km from the coastline. The data of Figure 1b show that |

A. uva-ursi, the species from the cool, foggy, coastal habitat, holds its

leaves most nearly horizontal, and that leaf inclination tends to increase —
with distance from the ocean. Leaf inclination of A. manzanita is not as |
steep as in A. mariposa (Fig. 1a, Table 1), but growing season tempera- |
tures and insolation are much lower in the northern California coast |
ranges than in the Southern Sierran foothills (Climatic Atlas of the

United States, U.S. Environmental Sciences Administration, 1968).

Fig. 1. Relative frequency of leaves in various leaf angle classes in the Sierra Neva-
da and Coast gradients. a = mean of all measurements.

1978] SHAVER: ARCTOSTAPHYLOS 135

a5 \G) SlERRA. NEVADA

4 A.mariposa

52 M A. viscida
8) A. patula
= [J A. nevadensis
<I
ip co
=|
iste
©
15
os

O-IO [5-20 25-30 35 40 45- 90 55-60 65-70 75-80 85-90

pee (py NOR fT HatCOAST

& A.manzanita

35 @ A.colombiana
ep) A. uva-ursi
=
i gy
: 4
z ; Gan Z g
Ah ww
n 1 2 ee
g Z
O 15 a Z . %
j ) fe He we eB eg
: : ye | Z
| We! & E
5 ie z Z g gs : Z Z
a F {1s g gf g J ; g
A a A a g g g

0-10 5.20 25-30 35-40 45-50 55-60 65-70 75-80 85-90
ANGLE CLASS

136 MADRONO [Vol. 25

TABLE 1. ELEVATIONAL RANGES, MEAN LEAF ANGLES FROM THE HoriIzontvat (a),
cos a, AND MEAN LicHT ABSORPTANCE IN THE VISIBLE AND NEAR-INFRARED REGIONS
FOR THE SPECIES USED IN THE PRESENT STUDY. Values in parentheses indicate sample
size.

Mean absorptance, %

Elevation, m a cosa 400-675 nm 800-1100 nm

Sierra Nevada

A. mariposa 150-1500 Vis30 (420) 22 — —

A. viscida 150-1500 76.24 (460) 24 78.96 4.01

A. patula 1000—2 800 69.72 (1260) 35 86.58 4.35

A.nevadensis 1700-3500 50.97 (1030) 63 83.83 4.89
North Coast

A.manzanita 200-1500 62°83; (320) 46 — _

A.columbiana 50~— 800 62.01 (720) Ay 87.97 8.01

A. uva-ursi 10— 100 47.51 (410) .68 91.87 8.67

In addition to the differences in leaf orientation, there are also differ-
ences among species of Arctostaphylos in light absorptance, particularly
in the visible region (Fig. 2, Table 1). Arctostaphylos viscida, the species
from the warmest, highest insolation environment, absorbs much less
visible radiation than any of the other species. Arctostaphylos uva-ursi,
from the foggy coastline, clearly absorbs the most light between 25,000
and 15,000 wave numbers (400-675 nm). The intermediate species, A.
patula and A. columbiana, are intermediate in their visible light absorp-
tance characteristics. Arctostaphylos nevadensis is the only species that
contradicts the predictions of the original hypothesis. This anomaly may
be related to the frequently very high radiation intensities during the
growing season at the high elevations where A. mevadensis is found. In
the near infrared region, all three of the Sierran species absorb about half
as much radiation as the Coastal species.

Differences among species in light absorption are much smaller than
those for leaf orientation (Fig. 1, 2), suggesting that leaf orientation
is much more important than absorptance in regulating the energy bal-
ance of Arctostaphylos leaves. Lack of precise correspondence with the
predictions for absorptance of the original hypothesis therefore may not
be critical. In general, however, the predictions of the hypothesis are
validated for leaf angle and for light absorptance of at least the two ex-
treme species, A. viscida and A. uva-urst.

Several other factors also might tend to counteract the predicted
trends. For example, leaves of both A. uva-ursi and A. nevadensis are the
smallest among all species studied, usually less than 2.5 cm long by 1.5
cm wide. Leaves of A. viscida may be over 4.0 cm long and 4.0 cm wide.
If large leaves tend to have greater leaf-air temperature differences, the
advantages of near vertical leaves and low absorptance may be counter-
acted. Field measurements of leaf temperature and photosynthesis are

1978] SHAVER: ARCTOSTAPHYLOS 137

A.viscida ——
A. patula —:-—

A. nevadensis — ——
A. columbiano — ~~

% ABSORPTANCE, upper surface

40} ,
A.uva-ursi «or

ZO

L ale ! | : Je

30,000 20,000

WAVE NUMBERS
ies (oo ‘ L es C
eye: 400 500 667 Kelele@) 2000

WAY Eee EN Gaieicineimn

Fig. 2. Percent of incident light absorbed by leaves of Arctostaphylos spp. Wave
numbers are used in the abscissa of this plot in order to emphasize the shorter, high
energy wavelengths.

needed. This research also has not considered the importance of canopy
structure on light absorptance and leaf temperature. Arctostaphylos ap-
pears to be a genus of plants very well suited to future studies of adapta-
tion to radiation and temperature in the environment.

ACKNOWLEDGMENTS

This reasearch was originally suggested as an undergraduate thesis
topic by Dr. Harold Mooney, who was very generous with his advice and
criticism. Conversations with Dr. Andreas Gigon were very helpful, and
Dr. Ward Watt kindly allowed me the use of his spectrophotometer.
P. C. Miller made helpful comments on the manuscript.

138 MADRONO [Vol. 25

LITERATURE CITED

ApaMs, J. E. 1940. A systematic study of the genus Arctostaphylos Adans. J. Elisha
Mitchell Soc. 56:1-62.

Biiiincs, W. D. and R. J. Morris. 1951. Reflection of visible and infrared radiation
from leaves of different ecological groups. Amer. J. Bot. 38:327-331.

Kvet, J. and J. K. Marswattr. 1971. Assessment of leaf area and other assimilating
plant surfaces. In: Z. Sestak, J. Catsky, and P. G. Jarvis, eds. Plant photosyn-
thetic production: Manual of methods. W. Junk, The Hague.

Mooney, H. A., S. L. Gutmon, D. J. Parsons, and T. F. Harrison. 1976. Morpho-
logical changes within the chaparral vegetation type as related to elevational
gradients. Madrono 22: 281-285.

Mowz, P. A. 1959. A California flora. University of California Press, Berkeley.

STEBBINS, G. L. and J. Major. 1965. Endemism and speciation in the California flora.
Ecol. Monogr. 35:1-35.

NOTES AND NEWS

Notes ON Two Rare, ENDEMIC SPECIES FROM THE KLAMATH REGION OF NORTH-
ERN CALIFORNIA, PHACELIA DALESIANA (HyDROPHYLLACEAE) AND RAILLARDELLA PRING-
LEI (CompositaE). — Phacelia dalesiana J. T. Howell was originally described from
a population in the Scott Mountains at what is now Scott Mountain Summit on
Highway 3 in Trinity County (Howell, J. T., 1937. Leafl. W. Bot. 2:51-52). The
second known locality for this species was found 31 km to the south in the Trinity
Alps (7 July 1975; J. M. Di Tomaso 109, DAV). The population consists of 700-
1,000 individuals and is scattered in open Red Fir Forest at 2,011 m along the trail
from Deer Flat to Shimmy Lake, 9.6 km WNW of Trinity Center; T36N, R8W,
Section 5, NW-% (41°0’50”N, 122°48’40”W). Additional specimens have been de-
posited at CAS, HSC, and JEPS (8 July 1976; Ferlatte and Di Tomaso 1776).

Raillardella pringlei Greene was first collected by Pringle in 1881 from the “moun-
tains about the head waters of the Sacramento River” in Siskiyou County and de-
scribed by E. L. Greene (Bull. Torrey Bot. Club 9:15-17, 1882; isotype: CAS!). It
has also been collected in the same general area near Gumboot Lake 12 km south of
Mt. Eddy (D. Barbe 538, CAS, JEPS). Ferlatte (A Flora of the Trinity Alps of
Northern California, Univ. Calif. Press, Berkeley, pp. 50-51, 1974) reported this
species from Union Creek and Landers Creek in Trinity County at elevations from
6,500-7,200 ft. (1980-2195 m). Further field work in the Trinity Alps has shown
Raillardella pringlei to be relatively common in the Swift Creek drainage and to
occur as low as 1,295 m (J. Di Tomaso 640, CAS, DAV, and HSC). J. L. Strother
obtained chromosome counts of 2m = 17 II from populations on Union Creek
(Ferlatte 1805, 1806) and Landers Creek (Ferlatte 1812). Vouchers have been de-
posited at HSC, JEPS, RSA, and UC. In all cases where Raillardella pringlei has been
observed in the field it occurs in wet places such as stream banks or boggy areas,
usually among serpentine rocks or in soils derived from serpentine or related ultra-
mafics. Associated genera include the following: Darlingtonia, Caltha, Schoenolirion,
Carex, Adiantum, and Dodecatheon. I thank Joseph M. Di Tomaso and John L.
Strother for their contributions to the data presented here. — WILLIAM J. FERLATTE,
California Dept. of Agriculture, 3288 Meadowview Rd., Sacramento, CA 95832.

FLORA AND CHOROLOGY OF THE
PINUS ALBICAULIS—VACCINIUM SCOPARIUM
ASSOCIATION

FRANK FORCELLA
Department of Botany — Microbiology
University of Oklahoma, Norman 73019

Floristic variation within a plant association may indicate that the
habitat of the association is not uniform throughout, and that two or
more plant associations are being considered as one, the extreme case
being that each community is an individual, an association unto itself.
Other interpretations exist; summaries can be found in several texts
dealing with vegetation. In this report I shall describe the compositional
variation of the Pinus albicaulis—Vaccinium scoparium association, and
relate some of the variation to one factor of vegetation formation (Major,
1951), i.e., the flora from which the vegetation may have originated.

The suggestion that a single plant association varies according to the
flora available to it suggests that this association may exist in more than
one floristic region. If regional climate and events during evolutionary
time determine floristic regions, one might conclude that these factors
could act differentially within the association and thereby affect its vari-
ation. Alternatively, if one assumes that a recurring mixture of plant
species indicates a particular set of environmental conditions, and that
the probability of two or more of these species concurrently evolving the
same degree of ecotypic variation is low (at least lower than with a
single taxon), then it follows that the habitat within which this associa-
tion of plant species exists is more or less equivalent throughout (if it is
integrated over ecologic time). Thus, if this be the case, floristic differ-
ences of communities with “identical” habitats must be a result of either
the availability of their flora at the time of their establishment (Egler,
1953) and/or through the remainder of their existence. Major and Pyott
(1965) give an interesting review and discussion of this topic. More
recently, Westhoff and van der Maarel (1973) and Mueller-Dombois
and Ellenberg (1974) have included this aspect of vegetation in their
writings.

METHOpDS

Stands in Wyoming, Idaho and Montana with overstories dominated
by Pinus albicaulis, understories dominated by Vaccinium scoparium,
and lacking conspicuous populations of Abies lasiocarpa seedlings and/or
layered shoots (i.e., Abies reproduction less than that of P. albicaulis),
and soils not stony or rocky (Soil Survey Staff, 1976) enough to obvi-
ously affect the growth and distribution of plants, were sampled. Within
a 600 m’ area the coverage of each vascular plant taxon was estimated.
Foliose aboreal lichens were also collected, but not systematically. No-

139

140 MADRONO [Vol. 25

menclature follows that of Hitchcock and Cronquist (1973) for the
Pacific Northwest vascular plants, Munz and Keck (1968) for other
vascular plants, and Hale (1969) for lichens. Taxonomic authorities not
in the text are listed in Table 1.

RESULTS

Four species other than Pinus albicaulis and Vaccinium scoparium
were nearly ubiquitous in the sampled stands: the widespread Carex
rossu, Abies lastocarpa and Poa nervosa with constancies of 80, 90 and
70% respectively, and Arnica latifolia (80% constancy, though absent
from most Wyoming stands). The presence of these taxa lends some sup-
port (or degrees of freedom in a statistical sense) to the initial assump-
tion of the improbability of two or more species concurrently evolving
associated ecotypes.

In Table 1, the flora and some other stand characteristics are pro-
vided in relevé form. This table lists the stands in a latitudinal sequence,
with adjustments to accommodate latitudinally-similar stands with widely
separated longitudinal ordinates (cf. Fig. 1). Stands have not been
sorted according to their floristic similarities as is usually done in
relevé analyses (Mueller-Dombois and Ellenberg, 1974). However, taxa
have been arranged to give the maximum impression of latitudinal change
to demonstrate that the flora of the association changes clinally with
latitude, and that the flora of any stand is at least a partial consequence
of the floristic region in with the stand exists (certainly other factors are
also involved ).

Table 1a consists of those taxa with constancies > 15%, and whose
presence appears to be nonrandomly distributed within the association.
When the within-association distributions of these taxa are compared to
their general distributions listed in standard Floras for the Pacific North-
west (Hitchcock and Cronquist, 1973; Davis, 1952; Booth and Wright,
1966; Shaw, 1976; Despain, 1975), approximately 15% are found to
be distributionally restricted from attaining 100% constancy in the P.
albicaulis-V. scoparium association. Similarly, of those taxa with be-
tween 5 and 15% constancies, and those with < 5% constancy (Tables
1b and 1c), about % of the former and % of the latter are distribu-
tionally restricted from ubiquity in the association. Taxa which exhibit
no latitudinal affinities (Table 1c) are characteristically widespread in
their general distributions. In total, about 25% of the association flora
shows limited general distributions within the area encompassed by the
association.

A few taxa in Table 1 deserve special mention. The low glandular
shrub, Leptodactylon pungens, and the similar but more cushion-like
Arenaria aculeata both have stiff spinulose leaves often found in desert-
region plants, as indeed both are. In P. albicaulis forests these species are
found in the southern Bitterroot Mountains and the Salmon River Moun-

141

PINUS—VACCINIUM

.
.

FORCELLA

1978]

h 2 2 I I . . . . . . . . . . . . . . . . . . . . . ° . . TIUW-JOWRH 224D00YOR IY DUNZNT,
Ne T . . I e . . . a ey are . . . . . . . . . . . . . . . . . BUseAID (*YOOH) DZDUDZ DIDUUaZUY
fe pes ee te ee asleep aie ah at ce A ale“ eniles Vea oes Mae Sg ANS: Vg Pe ‘uyebugd zhiuavd snouns
BR ace ne degen rst res anise Wan gg Ns aie ssc Varta” wie er Morin Gee neal 'e eeae'h, Uys barn ng ‘Z3ay DoLpww7ZUeeUh siiv]no1pad
th . . . I . . . . . . . su . . . . ° I I . . : . . . . . . easoy 8 *FINOD (*S}eM) 22yO7TSNO UuM17zDW0T,
Steg oe Cae reel] es CONC ee ok oN NET'S CE Cy Ci a See rg is Bp te ra Wish pane ae : 3300g wafhab rain)
oe a ; . ; é 5 dG . ° i ' yo. tot . 1 eZ ops . . . . YUSINd WUNLOTTUDAB unwMody7z Nag
ic aes SAN A Os Oar 9 = Aa et oge Aeaipe cee Avs a Sey oo AS ee Buog 027 0F7302 "V
0 4 . A . 5 eG . I ° . fj CE saa . ie ee Glee Wi AP ek ee eae *yooH D170f[1pdoo vo1udy
Se a NG eon We OR Sys ies see SATs She we a eter a ms Gt tal ae “uyel 0270f1tsdaa1p D11242Ua2 Od
Sa ep Ee ee Ge TO DO Re orem, SA Oe eo Pare ee ysang snajzuabup snurdny
. zZ . . . e . « e . . . ae + . . . . . + . E T T . 4: T . . ZOAYOTY (°'1) ung voids uNG ASIA TL
a nC SMS ceased fs ey Wein. MNS T cate en ee ROT gM Pc ck Ms We ho ML Re es Aeig snowidy sneovi10f 4a7SVyy
ge chez =a sso sc Go a rt a © Senator A ar Eien ct aaa : ‘wTebugd 0170f14D] bzuoZUOD SsMu1d_
Ee eg an ea ee Ret “zea (YSang) vonn1{b s1uesoby
Ce ee ns Sai <a kc Se Oe Ga “qpky v270freeya vla77 13809
i . 1G 8 y ¢ a + * : : i . : c i. 1S Pe Oe ere Sone a es “ATW D2Dippdz3zyNu Obvp217105
eet 2 anna oN NR) No Scr 2 a using 03D1]0e0Uv] D1U0Zh01
eet es Bee Ge ae ee ew oe ee "1 unz10fe7] WA vaz712YyoY
SP al RESP ELE GUS URS TS eS tC EOI CO-OP Cag Ce Ge SR eae eae *AIOL ungvp1oeoUun] uNvaS
ec cme sc 90 Ot) c0 GOMOneO A090 el ti GT Hl tloe Ze Ge eo 100160 Oc OF GU Ll OF uo = = TequaN pues
cl Of EZ Ee OT LE 9E 65 Hd SL SL 49 EH SS HS OW 62 226 OL SH HT 65 S26 9tiIz2d beso he Eid Po maus sed
OS BT £1 zz 2€ -- 99 O¢ LE HS 19 15 E24 On ES 9S dh 69 O€ TE 26 EL 2S 4S 6H £9 Id £9 £9 Ore) Bae glee cera
= r a rm \ m r a

Deine cae en ee eG ees es a ene pase

“(ANTeA paprosar Asaysty 94d STE ,7,) YBFY OF MOT WOIZ SasseTO ADUaNbery Juasaidai ,¢-[, seateym ‘,4, eB Aq pajuasal

-dai st eaie 3ur[dues = 009 943 UF aoUasaid aTdwtTs *(,) sysfaeqjse Aq pajoUuSp vie sUOTINqTAISTp [BUOTZaA ,MOTIeU, TeNIIe AfeyI OF eNp uo AePoosse
242 jo eduer ay} UTYIFM SUOCFINGFAISFP peITWTET YIIM exe] ‘asTm1syO pazeIs ssetuN Joy tds oFZPOeds 03 JuaTeAtTNba 10 AQoTaeA “IW AYIOY uoUMOD aq

OJ Uoxe} YORI JO Snje4S [eJaFAeA suUNSSsy *(3xX9} puke [ BTy °F) YyI1ou 02 YANOS wWo1z soUSNbaes [eUTpNITIeT[ e UT pedueaze aaevy spueys sy] *(UOFIeTIOS
~SB 942 UFYITM SUOFINAFAISTP PaIFWET IFQTYXoO YOFYM puke) YCT ueYy Jaee13 sapoueJsUOD YITM UOFAeTOOSSe |9yQ JO exe. VY YIM BuotTe umoys aie a3eI—a

-A02 umzrpdoos *j; % pue a8e1eao0d Adoued Teaioqie % ‘(sieak) a8e pueIs “NOILVIDOSSV WIYVdOIS WAINIQIVA - SITNVOIGTY SANId AHL AO V4OIA ‘PIL FTAVL

MADRONO

142

ao

. a. . . . . . . . i . . . . . . . . . . . . . . . . .
4 4 . . . e . . . . . . . . . . . . . e . . e e . . .
in rs . . . ° . . . . . . . . ° . . . . . . . . . ° . .
ic iE . + . . . . . . . . . . . . . . . . . . . . . .
+ . . a . . . . . . . . . . . . . . . . ° . e . . . ae
. T + ry Bt . . . . . . . . . . . . . . . . . . . . . .
+ . . . 4 . T . . . . . . . a . . . . . . . . e . . °
. . . ry e . T ue . . . . . . . . . . . U . . . . . . .
. . . . ° . . . ee . a . . . ‘ . . . e . . . ’ . . .
. . . . . . . . . . . . . . . . I it I . . . . . . . .
. . . rs . « . . . . . . . . . . . ee + . . . . . . . .
. . . T . . . . . . . . . . . . . e ag ae . . . . ‘ . .
. . . . ° . . . . . . ° e . . . . . + Be at T . . . . .
. . . . . . . . re . e . . . . . . . . T . ae . . . . .
° . e . . . . . . . « . . . . . . . . pts . no . . . . .
. at . . . . . . . . . . . . . . . . . . ae Be . . . ie .
. . . . . . . . . . ° e ° . . . T « . . . . . . T at .
. . . . . . . . . . . . . . e . . . . a e . . . at BS .
. . . . . . . . ° . . . . . . . . - . ae . . e . 4 . +
e . e . . . . . e . . . . . . . e . . e e at . . T . +
. e . . . . . . . . . . . . . . . ° . . . . ° . y . +

€z nz $2 92 O€ €O SO 40 BO ZO 90 ZI IT ST HT €T 62 Ze Ee2 ZO TO OT 60 O02 OT

io @)

BCL 9t

"uod *d (‘MS) squtofiuzgadua a00p01 7 fyd
*buorzD (4x9ey) snqvdudas snsoo1gnsif uUowazsuag,
"TAed 2277DA] r1aIvT,

“qpdéy 077 hydoxo1w pruDuUue2Uy

*doos (°*T) DyNsdtyu siqddyy

"34MN (USINg) xDUa, unz7AYydouey,

"T1ed D270f17N,zaq DavArds

*Tbnod voL1upurzAo Daayonay

"1 suaqumooud vip] 0qq1s

"SQEM DGDE7NOD V1IADUadyYy

‘pusy (esoy 9 Aqued) tHpeeqy v7 ~WdoUuc1Yy),
UaTWS (*F4NN) 22WQSAYy Uuo1UnI19

"I DO101p DUDWaID/

Keay snjnssvdo 02108uaS,

*psyns (eSoy 3 *3TNOD) vaeunduind vz1yYuOWsSOy
*YOOH DSOWBODA D1IADUUaZUY

"Zsu0D 3 "UFEW (°34NN) UNgoass1p um2q.DU0T

24O3TH (*qpAu) suaosauundq $1710DuB D7714uUa40g

YsINg 1277D2NU 07014
*szem suaosaanzs 016a7,1nby

*sutqqoy (Ae19) vapwbAd p1S81m27

“DT 8TqeL UF BSOY MOTTOF suoFIeoTzPoads aTqeL *(NOILVIOOSSV AHL NIHLIM SNOILAUIULSIG GALIWII LIGIHXY
HOIHM GNV) %ST GNV S NAAMLAG SAIONVLSNOD HLIM NOILVIOOSSV JHL dO VXVL ‘NOILVIOOSSV WIIYVdOIS WNINIOOVA - SITAVIIGTY SANId AhL dO VHOTA “QT ATAVL

143

PINUS—VACCINIUM

FORCELLA

1978]

"Ae1Q tAuupd vynupduog 72 $*S3eM ("Ad *Y) 02720f2aedq vULAO vONZSay ‘KhaseA 11bueqpuDs DLg ‘keyx0TD Ssiuoshpd xaun]

€Z $°H 9 °d (1) 0a0D¢1uDbuDW s1zDYydDuy ‘*TTeg 0270f1suap vapards, ‘aueeig (*400}H) popiqubunb bono] b sikasoby ‘+ yo3T}, (sauor) an2ucguou umsow

-aof unordedhy ‘Keasea Dipautezu2r DruoyzUuDng ‘*yxOoH unto1 f1gq]0 untovder}; ‘TTauueg sugosaan7 fs uowajzsuag, ‘*uemg (Ab19) 1277044NU UNdgsDoyzuDoU1q, ‘AeAD

ZLYOLSND VoOUoUeA, ‘*°1 DIZOf20277AYyd x270g **34NN 4 °270f2oOMDd cauny 47 £°379NN (*410]) suabund uo7zAqzovpozdeT,sz $*¥ 9 *H SnqDurzEDvO snuoug ‘ke19

SadQUOUSZS LAZSYy JZ Sz ueIrg vdorouow shy2dodhy gE t3eemg (°1) vUu2GOues DIpAo7ZT ‘*34NN s2zNpe DUDA] DA € S*T8IN0G wWP.aoDU_AqUaWU UNLULDOVAy

fy] vuzdj]D vog { t*3eu 9270f[26u0z qo Duahpoop GI ‘:*saaq suadau auazzg LZ tkeay (*339NN) snsoo1niffns snddvdoq ddpjj,**qanyy £27muapi000 vd145 gz $10

=1oS) “suraq_(*qpdAu) DG2128N4 qo DUdibUady 0% *°1 071 adbuw Yuyrquay **uog *y (*aa0L) OD Z2e

-310qg Z2uUapADYy un2j0f24J, 7 taueeag (ysang) snsoo1HNss snsoo1jnds uoweq sued OT *°1 s27v2YyoUoUG vbvafixvg ‘Suokal (UOXEW) UNuo{NdoDs vsuey D2 ]eU2bD
2]

DLSUALAAWly OT $°3I0N SN7ZwMd Uodabidy [[ :a41e exe. BAT

-oedsa1 ateya YITM spueys ayzZ faaoge uMOYS soUaNbas TeUTpNITIeT aya ul *({G AoUeISUOD) Spue Is ay UT SaoUazaNdD0 aTBZuUTS ATUO aaey BX] OMj-AZ ATUL /T

se pe 2 Hd oo Bo eof om tc ; 4 se ge be es : . ‘ATOd (*SA8q) Bigsnov] saqiy
. . . . . . lo . . ° Reba aen é : . 5 0 3 . . . . l . . . TTRH ("UpAd) D7D2adA3D]d DuDLULb41Aa D1IAvHDAY
. . . . . . . + . I . . . . . . . BY . . . . . ny . . a . using 80P104 407 S2q unuobR7 og
Do Ses Ge Rc BI CA Se Oe Se OP oe Oa ON Ot aueery (Yysaind) snuidbaued uouab1ay

a . in a . . a ° . . + ° . . . . + I 4 . + . e . ’ 4 if . aztnbew (*qpAu) vap2107 Dydaa 03 sabu00 DUIDuady

. . . . . . . . . . . Ae . ‘ ‘ : ie © . . . ‘ . . ‘ a “qanuL ondica sisdozhug
Sa he °C =" tO eS) G72 so es" a SG SLs @ 2 SO ae on “STON ‘VW 22bueqpAa uowazsUuag
ae tS 8 St ee ies : y 0 ay a : : : : ec ees Go 5 i ‘ s + "ZOANL DZDSNAGO Dpunvas D1 Oang
1D, tae ve) ag 8 eG eR ee EO Sth G4) es U0 ac Aqieg 21uuduzabue va01g
ct ea mgm cm nc et +S cco (rt RD oO to Sect *qpky D7] aU1dqUN DIdDUUaqUy
mg ceils NOS Zl Wey ea ee Gus ae Wil XA cc SOC EN cg Coa Prag acy i cae a “QIW DUDZUOU stUnuUCD snuadiune
oe Gen 7 eas ag) Sige OE Ge oe OM RSET © STO Sr Ea SoCo Sr > > "I wn170f23snBup un2qo7144
1G CS Bore i) i OEE NBs te Ge [hut Ug REE ARN ey eee a *HOOH 37 20DaD wNtODsaLY
etre Geer Ge ee | te ey meme Lye tne : . : ts ee eee oe ies ES I ANE ran oil Aasea (*XOOH) DSouteu Dog
(Ales ap a wlG. Fill ear w) ° + + > Pont se 4 oe eee . I tT 3 [hs a I + 33004 226804 Ladd)
1g a I tT +4 I on ee + I + 4 eae YY l I 4 rf + ie l ' rf t ° l . : “qann (*400H) vdupoeoisy] selqy

td td. Sid 9d OF €0-S0 70 80-20-90 ST Tl ST HT €1 6c..d6 82 ZO 10 OF 60 Oc 6T Bt at OT

‘DL PTGeL UF eSOy MOTTOJ SuOoFAeoTjJtoeds atqey
SER SAILINIGAV IVNIGALILVI ON ONILIGIHX NOILVIOOSSV AHL dO VXVL YHL ‘NOILVIOOSSV WIINYdOOS WIINIQOVA - SITAVOIAIY SAMI GC FHL dO VNO'ld "el FIEVEL

144 MADRONO [Vol. 25

tains of east and central Idaho. The dry finger-like intermountain valleys
(Lemhi, Pahsimaroi and Lost River Valleys) which extend from the
northern edge of the Great Basin and abut these mountain ranges prob-
ably supplied the migratory path for these species from the deserts to
P. albicaulis forests. To find either taxon in a mesic subalpine forest is
surprising, but would have been much more so if that forest had been
in central Montana, rather than east-central Idaho with its direct con-
nection to the Great Basin.

16° 4° Wee 110° 108°

A x 4ge

MONTANA

47°

45°

43°

Fic.1. Distribution of the Pinus albicaulis-Vaccinium scoparium association. The
numbers and their associated characters (e.g., n-, -n-) represent stand numbers and
the three geographic/floristic regions referred to in the text. The geographic extent
of the stands is thought to depict the range of the association. The unhatched area
inside the dotted line represents the gap in the distribution of Pinus ponderosa
(from Little, 1971). (There is no Stand 21).

1978] FORCELLA: PINUS—VACCINIUM 145

Three species of the P. albicaulis-V. scoparium association are rela-
tively narrow endemics: Penstemon flavescens (Idaho Co., Idaho and
Ravalli Co., Montana), Chionophila tweedyi (central Idaho and adjacent
Montana), and Aster stenomeres (central Idaho and adjacent Montana
to northeastern Washington and southeastern British Columbia). These
taxa would not be expected to occur in whitebark pine stands outside
their restricted ranges. Similarly, the typically alpine Trifolium haydenu
extends only as far north as southern Montana. It occurs in a whitebark
pine stand immediately adjacent to alpine meadows and scree in the
Madison Range (Gallatin Co., Montana) and would not be expected in
P. albicaulis forests farther north.

Though the general regional occurrences of the above taxa are easily
obtained from standard Floras, their equally important intraregional
distributions are not so readily available. For example, Arnica latifolia
and Carex geyeri, which are prominent in many Montana whitebark
pine stands, are absent from the stands in the Wind River Mts. This
mountain range does support both taxa but their populations are not as
extensive as elsewhere. In such cases the chance of limited taxa reaching
Pinus albicaulis forests is low. King (1977) has noted-this same phe-
nomenon but on a much smaller scale; the ability of a plant to colonize
ant mounds in British pastures is determined by its relative abundance
and distance from the mounds.

Lichens. Both Letharia vulpina (L.) Hue and Hypogymnia vittata ( Ach.)
Gas. were widespread throughout the association, the former being much
more prominent. Alectoria oregana Tuck. and A. americana Mot. were
confined to the northern-most stands. Both alectorias have limited distri-
butions in the Rocky Mts. which correlate with their presence in the
whitebark pine forests.

Ecological Factors. Ecological factors (s. strictu) do, of course, differ
somewhat between stands, and plants respond accordingly. For instance,
whitebark pine forests in the Wind River Mts. of Wyoming are more
stoney than most. A very sandy soil develops about the stones that lie on
the general soil surface; these small patches of ‘‘open” sandy soils are
preferred sites for Sedum lanceolatum (cf. Table 1a).

Superficially at least, in some regions of P. albicaulis forests there
appears to be taxonomic replacement within ‘“‘life forms”. Arnica cordi-
folia is generally prominent in those stands in which A. latifolia is not,
and Luzula hiichcockii is relatively important in the Bitterroot Mts.
where Carex geveri is not.

Disjunctions. Daubenmire (1975) has applied the term ‘“‘oceanic ele-
ment” to taxa with distributions largely restricted to maritime-influenced
climates of the Pacific Northwest (NW Montana, N Idaho, W Oregon,
W and NE Washington and the adjacent parts of Alberta and British

146 MADRONO [Vol. 25

Columbia). I had considered both Xerophyllum tenax and Luzula hitch-
cocku to be strict oceanic elements, but their actual distributions are, in
fact, more extensive. Widely disjunct populations of both species occur
as far south as Teton Co., Wyoming (Shaw, 1976; Pfister et al., 1974;
Maule, 1959). Menziesia ferruginea Smith and Pinus monticola Doug].
(not in whitebark pine forests) are other oceanic elements often found
close to or associated with X. tenax and L. hitchcockti. They also have
disjunct distributions nearly identical to the others (Hickman and John-
son, 1969; and personal observations). Perhaps in past times, the paleo-
climate was sufficiently different to support an “oceanic” vegetation
throughout the northern Rocky Mountains, as presently exists in NW
Montana and N Idaho. Additional evidence for such a maritime paleo-
climate is the discovery of Taxus brevifolia Nutt. (an unquestionably
oceanic species) wood remnants during archaeological excavations in the
Yellowstone Valley, SW Montana (Arthur, 1966; the same valley pres-
ently supports very localized populations of X. tenax and M. ferruginea).
Radiocarbon dates for the Taxus materials were 5000 years BP (Such an
age predates the well-known use of Taxus wood for archer’s bows, thus
long distance transport of the wood to this site is not likely.) The early
Holocene epoch in the Rocky Mountains is thought to have been cool
and wet (Hansen, 1947); the Xero- or Altithermal interval began about
7500 BP and lasted until the onset of Neoglaciation, ca. 4000 BP (Rich-
mond, 1970). Wells (1970) has suggested that the ‘““Xerothermal” inter-
val in the Laramie Basin of Wyoming was wetter, not dryer, than present.
Unless these plant disjunctions and excavations represent relict vegeta-
tion from pre-Pinedale Glacial times, with the recession of Cordilleran
ice (12,000 BP; Richmond, 1970), a Pacific maritime climate and vege-
tation may have pervaded the entire northern Rocky Mountains. A sub-
sequent cooling and drying trend in W Wyoming and SW Montana could
not support a maritime vegetation, and extinctions and disjunctions re-
sulted. High elevation bog pollen profiles in Yellowstone National Park
(Waddington and Wright, 1974) are dominated by Pinus contoria from
ca. 11,600 BP to present; an increase of Picea engelmanni pollen at
5O00BP implies climatic cooling. That the W Wyoming-SW Montana
area is still subjected to a relatively cold climate can be seen by the
present gap in the distribution of Pinus ponderosa Laws. (Fig. 1), a
typically “warm” pine (Mirov, 1967). Curiously, the absence of P.
ponderosa from this area correlates generally with the occurrence of the
P. albicaulis-V. scoparium association. If those whitebark pine stands
with oceanic elements are omitted the correlation is nearly perfect.

Species Number. The number of species in the whitebark pine stands
ranged from 6 in the oldest (640 years) to 33 in one of the youngest (33
years). There was a general trend in decreasing species number with
stand age but stands that were proximal tended to have similar species
numbers despite age differences.

1978] FORCELLA: PINUS—VACCINIUM 147

Management Implications. Although whitebark pine forests receive rela-
tively little resource management attention at present, this can be ex-
pected to increase rapidly. P. albicaulis produces exceptionally large mast
crops (Forcella, 1977) and such production may significantly affect the
habits of wildlife (Craighead, 1976; Forcella, 1977). Total net primary
productivity in these forests may exceed 900 g/m?/yr. and standing
crops may approach 60 kg/m? (Forcella and Weaver, 1977); economi-
cally these figures are substantial.

There are also some practical aspects involved with the floristic distri-
butional anomalies of whitebark pine forests. The three dominant herba-
ceous species of the association are Carex geyeri, Arnica latifolia and
A. cordifolia. These taxa all have known forage value for both domestic
and wild ungulates. The biomass and energy (kcal) per unit area of each
species can be readily predicted from their canopy coverages (measured
separately; Forcella, 1977). Further, as can be seen in Table 1, the
species have distributional limits within the association. If the 29 stands
are split into three geographic/floristic regions (Fig. 1; separations based
on plant distributions and agglomerative cluster analysis), the mean
energy value per m? for each species differs significantly between at least
two regions (t-test, p = 0.01). In vegetation mapping, the P. albicaulis—
V. scoparium association as a whole would probably comprise a single
cartographic unit. Knowledge of regional differences in forage avail-
ability within associations might prove valuable to resource managers.

Chorology. The stands shown in Fig. 1 essentially outline the distribution
of the P. albicaulis-V. scoparium association. To the north and north-
west, Abies lasiocarpa, Larix lyallii and Vaccinium membranaceum gain
importance in whitebark pine forests. In Alberta, Canada (on acidic sub-
strates), P. albicaulis occurs with equal amounts of Picea engelmani and
A. lasiocarpa in the overstory. Understory components always contain
V. scoparium, but it may be accompanied or dominated by Vaccinium
caespitosum Michx., Empetrum nigrum L., Dryas octopetala L., Salix
arctica Pall., or Spiraea ssp. In Banff National Park, I found one stand
on dolostone totally dominated by P. albicaulis ; its understory, in order
of importance, consisted of Betula glandulosa Michx., Potentilla fruticosa
L., Linnaea borealis L., Shepherdia canadensis (L.) Nutt., Juniperus
communis and Dryas octopetala. There were no vacciniums in this stand,
probably due to its basic substrate.

The eastern limit of the P. albicaulis—V. scoparium association is corre-
lated with the eastern extent of acid-rock mountain ranges in Alberta
and Montana. Limestone ranges such as the Big Snowy Mountains
(Montana) do not contain this association. The eastern limit in Wyo-
ming is the Absaroka and Wind River Mountains; the granitic Big Horn
Mountains, 170 km eastward, have only scattered populations of P. albi-
caulis (Hoffman and Alexander, 1976; D. Despain pers. comm.)

To the south, the Medicine Bow Mountains (Wyoming), the Colorado

148 MADRONO [Vol. 25

Rockies, and the Uinta Mountains of Utah all lack whitebark pine. That
the southern limit of P. albicaulis coincides with the northern boundary
of other edible large-seeded, grove-forming pines (P. edulis Engelm., S
Wyoming; P. monophylla Torr. & Frem., S Idaho to California) may be
more than coincidental. Forcella and Rumley (in prep.) hypothesize
that prehistoric man carried seed of P. sibirica L. ( = P. albicaulis)
across Beringia. His dispersal of the energy-rich seed ceased when con-
tact was made with native large-seeded pines.

In far western Wyoming (the Wyoming Range), P. albicaulis forests
contain an understory of Ribes montigenum McClatchie (which forms
conspicuous Closed circles under the canopies of the rather widely spaced
trees) and Bromus carinatus. To the northwest, in the White Cloud
Peaks and Sawtooth Mountains of central Idaho, P. albicaulis stands
often support an understory of Artemisia tridentata Nutt. and/or a
carpet of forbs, Lupinus argenteus being the most prominent.

Within the distributional limits of the P. albicaulis —V. scoparium
association, there may be other associations which contain P. albicaulis.
On limestone outcrops, Weaver and Dale (1974) mention a stand in
which P. flexilis James and various forbs associate with whitebark pine.
I have seen such stands and others similar, but always including Arcto-
staphylos uva-ursi (L.) Spreng. This type of community, with a dis-
tinctly different habitat (limestone), appears to have been lumped with
the P. albicaulis-V. scoparium association in the “habitat-type” classifi-
cation of Pfister et al. (1974) and Reed (1976). Also, on what may be
more mesic sites, Abies lasiocarpa shares the overstory with whitebark
pine, and V. membranaceum is often present in the understory. It is pos-
sible that alternate plant associations (Abies vs. Pinus) may exist on
the same site at different times, the occurrence of either possibly being
a function of its seed crop size at the time of stand establishment. Seed
production of P. albicaulis varies significantly from year to year (For-
cella, 1977). A treeline form(s) of whitebark pine community occurs
also; its distinguishing feature is, of course, the stunted growth and
flagged structure of the trees (Daubenmire and Daubenmire, 1968).
Clausen (1965) speculates a genetic basis for the stunted P. albicaulis
of the Sierra Nevadan krummholz.

CONCLUSIONS

The Pinus albicaulis-Vaccinium scoparium association is limited to
subalpine sites on non-calcareous substrates in western Wyoming, south-
western Montana and east-central Idaho. Its floristic composition changes
clinally with latitude, but this does not necessarily imply a change in hab-
itat. Nearly 25% of the taxa which comprise the association are distri-
butionally restricted from occurring in all stands of the association. This
suggests that to some degree the floristic composition of a stand is a
function of the local flora available to it.

1978] FORCELLA: PINUS—VACCINIUM 149

ACKNOWLEDGMENTS
In addition to J. H. Rumely and T. Weaver, I thank W. Ferlatte and J. Major,
the reviewers for Madrono, for making valuable comments on this paper. The
summer field work for this study was supported by the U.S. Forest Service under
Contract No. 12-11—204-12, Suppl. No. 33.

LITERATURE CITED

ArtHurR, G. W. 1966. An archaeological survey of the upper Yellowstone River
Drainage, Montana. Mont. Agr. Exp. Stn. Econ. Res. Rept. 26, Bozeman. 199 pp.

Bootn, W. E. and J. C. Wricut. 1966. Flora of Montana, II. Montana State U.,
Bczeman. 305 pp.

CLAUSEN, J. 1965. Population studies of alpine and subalpine races of conifers and
willows in the California High Sierra Nevada. Evolution 19(1) :56-68.

CRAIGHEAD, J. 1976. Mapping grizzlies by satellite. Nat. Geographic 105(1) :148-158.

DAUBENMIRE, R. 1975. Floristic plant geography of eastern Washington and north-
ern Idaho. J. Biogeography 2:1-18.

and J. B. DAUBENMIRE. 1968. Forest vegetation of eastern Washington and
northern Idaho. Wash. Agr. Exp. Stn. Bull. 60. 104 pp.

Davis, R. C. 1952. Flora of Idaho. Wm. C. Brown Co., Dubuque, Iowa. 828 pp.

Despain, D. G. 1975. Flora of Yellowstone National Park. Yellowstone Library and
Museum Assoc. 155 pp.

Ecier, F. E. 1954. Vegetation science concepts I. Initial floristic composition, a
factor in old field vegetation development. Vegetatio 14(6) :412-471.

ForceLLa, F. 1977. Flora, biomass and productivity of the Pinus albicaulis-Vac-
cinium scoparium association. Thesis, Montana State U., Bozeman.

and T. WEAVER. 1977. Biomass and productivity of the subalpine Pinus
albicaulis-Vaccinium scoparium association in Montana, U.S.A. Vegetatio
35(2) :95-106.

Hate, M. 1969. The Lichens. Wm. C. Brown Co., Dubuque, Iowa. 226 pp.

Hansen, H. P. 1947. Postglacial forest succession, climate and chronology in the
Pacific Northwest. Trans. Am. Phil. Soc. 37:1—130.

Hickman, J. C. and M. P. Jounson. 1969. An analysis of geographic variation in
Western North American Menziesia (Ericaceae). Madrono 20(1):1—11.

Hitrcucock, C. L. and A. Cronoutst. 1973. Flora of the Pacific Northwest. U. of
Washington Press, Seattle. 730 pp.

HorrMan, G. R. and R. R. ALEXANDER. 1976. Forest vegetation of the Big Horn
Mountains, Wyoming: A habitat type classification. USDA Forest Service Res.
Paper RM-170, Rocky Mt. Forest & Range Exp. Stn., Fort Collins, Colorado.

Kino, T. T. 1977. The plant ecology of anthills in calcareous grassland II. Succes-
sion on the mounds. J. Ecol. 65:257-278.

Littre, E. L. 1971. Atlas of United States Trees I, Conifers and Important Hard-
woods. USDA Forest Service, Misc. Publ. 1146.

Major, J. 1951. A functional, factorial approach to plant ecology. Ecology 32:
392-412.

and W. T. Pycrr. 1965. Buried, viable seeds in two California bunchgrass
sites and their bearing on the definition of a flora. Vegetatio 13:253-282.

Mauvte, S. M. 1959. Xerophyllum tenax, squawgrass, its geographic distribution
and its behavior on Mount Rainier, Washington. Madrofio 15(2) :39-48.

Mirov, N. T. 1967. The Genus Pinus. Ronald Press Co. New York. 602 pp.

MveEtter-Domepors, D. and H. ELLenperc. 1974. Aims and Methods of Vegetation
Ecology. John Wiley & Sons, New York. 547 pp.

Mouwz, P. A. and D. D. Keck. 1968. A California Flora and Supplement. U. of Cali-
fornia, Berkeley. 1905 pp.

Prister, R. D., B. L. Kovatcuix, S. F. Arno and R. C. Presspy. 1974. Forest
Habitat Types of Montana. USDA Forest Service, Intermountain Forest and
Range Exp. Stn. Missoula, Mont. 213 pp.

150 MADRONO [Vol. 25

RicHMonpD, G. M. 1972. Appraisal of the future climate of the Holocene in the
Rocky Mountains. Quat. Res. 2:315-322.

SHAw, R. J. 1976. Field Guide to the Vascular Plants of Grand Teton National
Park and Teton County, Wyoming. Utah State U. Press, Logan. 301 pp.

Sort Survey Starr. 1975. Soil Taxonomy. USDA Soil Conservation Service, Agr.
Handbook 436. 754 pp.

WappincTon, J. B. and H. E. Wricut. 1974. Late Quaternary vegetation on the east
side of Yellowstone National Park, Wyoming. Quat. Res. 4:175—-184.

WE Its, P. V. 1970. Postglacial vegetational history of the Great Plains. Science
167:1574-1582.

WestuHorr, V. and E. vAN DER MAareEL. 1973. The Braun-Blaunquet approach. In
Handbook of Vegetation Science V: Ordination and Classification of Communi-
ties. R. H. Whittaker (ed.), pp. 617-726. W. Junk Publ., The Hague.

SPECIAL OFFER

In July 1969 the California Botanical Society published a large, 128-page double
issue of Madrono (vol. 20, no. 3) to commemorate the meeting of the XI Interna-
tional Botanical Congress in Seattle. This issue contains four articles of general interest:
Jack A. Wolfe: Neogene floristic and vegetational history of the Pacific Northwest.
R,. Daubenmire: Ecological plant geography of the Pacific Northwest.

A. R. Kruckeberg: Soil diversity and the distribution of plants, with examples from
western North America.

W. B. Schofield: Phytogeography of northwestern North America: bryophytes and
vascular plants.

We have an unusually large stock of this special issue. Consequently, to reduce our

inventory, we are offering (until 1 Mar. 1979) this commemoration issue at the spe-

cial price of $2.00 postpaid instead of $3.00.

In addition, we offer the following at the special price of $1.00 postpaid (usual
price is $3.00): Howard E. McMinn, Studies in the genus Diplacus [= Mimulus]
(Scrophulariaceae), which was the entire number 2 issue (pp. 33-128) of volume 11.

For either or both issues please send remittance to the following address (a self-
addressed label would be appreciated) : Corresponding Secretary, California Botani-
cal Society, Department of Botany, University of California, Berkeley, California
94720.

MOVING? MOVED?

The California Botanical Society must pay 25 cents each time change of address
information is supplied by the post office. Copies of Madrofo returned to us cost 50
cents or more, depending on the weight of an issue. In addition, members pay a
forwarding charge if they instruct the post office to reroute their issues of Madrono.
Please save yourself and us money by promptly informing us of any change of ad-
dress. — Corresponding Secretary, California Botanical Society, Dept. of Botany,
University of California, Berkeley CA 94720.

Errata, Madrono, volume 24

. 140. Seventh line from bottom should read “Cactaceas” for “Cactacea”.

. 141. Second line should read “segmentisque” for “sementisque’’, “enationibus” for
“enatioaibus”, and “filamentis” for “fiilamentis’’.

. 145. Last line should read “synonymy” for “snyonymy”’.

. 150. Sixth line from bottom should read “Mason” for “Masin”.

. 155. Thirteenth line should read “filamentisque” for “fiilimentisque”’.

.156. Third paragraph, third line, should read “collections” for “collection”.

30 'U

soy seek is)

THE GENUS TRICHOSTEMA (LABIATAE) IN MEXICO

HARLAN LEwIs
Department of Biology
University of California, Los Angeles 90024
JERZY RZEDOWSKI
Escuela Nacional de Ciencias Bioldgicas,
Instituto Politécnico Nacional,
Mexico 17, D.F.

Trichostema L. is a North American genus consisting of 16 species. At
the time of the most recent taxonomic revision (Lewis, 1945) only 19
collections representing six species were known from Mexico. Least
known of these was 7. purpusii, the one species of this genus endemic to
Mexico, which was represented only by the type specimen collected in
1907 and a second collection made in 1908.

Our present interest in the genus 77ichostema in Mexico was gener-
ated in 1967 when one of us, at the instigation of Dr. Peter H. Raven,
collected seeds of T. purpusii (Rzedowski 24915). Plants from these
seeds were grown at UCLA in 1968 and flowered about six months after
germination. To the great surprise of the senior author the flowers were
bright rose-pink in color and not red or scarlet as he had thought prob-
able from small traces of pigment in the flowers of the herbarium speci-
mens previously known to him. Under cultivation the plants proved to
be shrubs 1 m or more tall with erect herbaceous tips as much as 50 cm
long rather than a suffrutescent perennial 4 to 5 dm tall as described
earlier (Lewis, 1945).

Chromosomes of plants from this collection were examined using stand-
ard squash preparations of microsporocytes. Ten pairs of chromosomes
were found consistently in all cells examined and no meiotic irregularities
were observed. A haploid number of 10 also characterizes other species
that occur in Mexico, namely T. arizonicum and the two shrubby species
in sect. Chromocephalum, T. lanatum and T. parishiit (Lewis, 1960).
With the chromosome number of every species in the genus now known
it is apparent that all of the perennial species have m = 10 except for
T. suffrutescens in southern Florida, which like other species of sect.
Trichostema has n = 19. The strictly annual sect. Orthopodium, repre-
sented in Mexico by T. lanceolatum, T. micranthum, and T. austromon-
tanum, has a constant haploid number of seven except for the tetraploid
T. austromontanum with n = 14 (Lewis, 1960).

With living plants available and with the chromosome number known,
it became apparent that T. purpusii is closely related to T. arizonicum.
The conspicuous differences in the color, size, and conformation of the
flowers that earlier led to placing the two species in different monotypic
sections do not, in living plants, obscure the close overall morphological

151

152 MADRONO [Vol. 25

similarity of these two species. We are now convinced that they are much
more closely related to each other than to other species in the genus. We
have, therefore, as indicated in the listing below, included them in the
same section.

Trichostema purpusii was known to us only from the State of Puebla
when our material was grown. We have since learned from a paper by
Williams (1973) that Dr. Robert Cruden collected this species in the
State of Oaxaca in 1966 and in the same general area in 1971. The latter
collection is the type of Eplingia saxicola Williams, the type species of
the genus Eplingia Williams. We are truly sorry that a genus named in
honor of a distinguished student of American Labiatae must be relegated

‘AutAUOUAS 0}

In the course of examining material of Tvichostema from Mexico it
became apparent that not only has the number of available collections
increased significantly since the genus was revised in 1945, but among
these collections are some that represent significant extensions of known
range of the species concerned and the first collection of T. austromon-
tanum from Mexico. We have summarized below our knowledge of the
genus in Mexico with a key to the seven species known to occur there and
a list of all collections from Mexico seen by us. We have not given full
synonymy, descriptions, ranges of distribution in the United States, and
other information available in earlier publications (Lewis, 1945; 1960).

KEY TO THE SPECIES OF TRICHOSTEMA IN MExIco
A. Plants annual; Baja California.
B. Stamens 10-20 mmlong.. . . . . . . . JT. lanceolatum
BB. Stamens 2—6 mm long.
C. Stamens 2—3 mm long, straight or slightly

curved, barely exserted. . . . . TT. micranthum
CC: Namie 3—6 mm pe strongly preled
and exserted. . . 2. 2... DL. austromontanum

AA. Plants perennial.
B. Leaves linear; Baja California.
C. Corolla tube 4-7 mmlong.’. 2.0 2) 2. we parish
CC. Corolla tube 9-14mmlong. . . . . . . YT. lanatum
BB. Leaves oblong to ovate.
C. Flowers blue; stamens strongly arched;
corolla tube 3-4 mm long. . . . . . . JT. arizonicum
CC. Flowers rose-pink; stamens nearly
straight or slightly curved; corolla
tube 9-10 mm long.) we ie a ed pur busu

TRICHOSTEMA COLLECTIONS FROM MEXICO
Section PANICULATUM Lewis (Section Rhodanthum Lewis)
T. purpusii Brandegee (Eplingia saxicola Williams). Oaxaca: Cerro del Camello,
2 km al S de Tepelmeme de Morelos, 2500 m, Cruz Cisneros 2189; Route 15 at K

1978] LEWIS & RZEDOWSKI: TRICHOSTEMA 153

376, ca 13.5 km NW of Tamazulapan, 2260 m, Cruden 1094; Route 190 between
K 72 and 73, ca 5.5 km NW of Yanhuitlan, Cruden 1950. Puebla: Cerro de la Yerba,
Purpus 2559; vicinity of Puebla, Cerro de Santa Maria de Zacatepec, Bro. Arséne in
1908; Ladera E del Cerro Tecajete, cerca de San Miguel Papaxtla, Municipio de
Cholula, Rzedowski 24915; San Luis de los Pinos, municipio de Ajalpan, Robert &
Moreno 364.

T. arizonicum Gray. Chihuahua: Colonia Juarez, Sierra Madre, Jones in 1903;
Carretas, Municipio de Janos, White 962. Coahuila: Puerta de San Lazaro, Sierra de
San Lazaro, Municipio de Castanos, Muller 3055. San Luis Potosi: 8 km al NE de
Laguna seca, km 20 carretera San Luis Potosi-Antiguo Morelos, 2300 m, Rzedowski
6352. Sonora: San Bernardino, Thurber in 1852: Pinal, Sierra Charuco, Gentry 1693;
Las Tierritas, Phillips 656; N of Horconcitas, Phillips 866; Puerto de los Aserraderos,
White 3221; Canon de E] Temblor, White 3373; Valle de Teras, near La Angostura,
W hite 3549; El Rancho del Roble, NE of El] Tigre, White 4212.

Section CHROMOCEPHALUM Lewis

T. lanatum Bentham. Baja California: Salada, Orcutt 1345; Aliso, Brandegee in
1893; 5 mi NE of Cerro Coronel (32°20’N 116°51’W) 560 m, Moran 21743; 1 mi W
of Buena Vista (31°02’N 115°48’W) 750 m, Moran 15109; Los Alisos, 9 mi W of
Valladares (30°52’N 115°51’W) 525 m, Moran 16245; Moran 16253; 4 mi W of
Valladares, 700 m (growing with 7. parishii), Moran 16427; Rio Santo Domingo,
Hamilton Ranch, Moran 22387 ; 20 mi E of Socorro, 30 mi N of Rosario, Humphrey
6833e; near Rancho E] Ciprés (30°23’N 115°38’W) 475 m, Thorne 31951; 1.5 mi E
of Rancho El Ciprés, 550 m, Moran 11046.

T. parishit Vasey. Baja California: Guadalupe Mountain, Orcutt in 1883; San
Rafael Hills, Orcutt in 1889; Aliso, Brandegee in 1893; Nachoguero Valley, Mearns
in 1894; Tecate, Orcutt in 1884; Fosberg 8396; 8 mi SE of Tecate, Munz 9487; 13
mi E of Tecate, Moran 14887; 15 mi E of Tecate near road to Mexicali, 3000 ft,
Wiggins & Thomas 431; near E] Compadre, Hohenthal 33; Sierra Pinal, 3.7 mi N
of El Compadre along road from Ojos Negros to Tecate, Wiggins 21650; 12 mi W
of La Rumorosa, 4000 ft, Hevly 2048 & Pitman 216; Route 2, 59.3 mi W of main
route to Mexicali, McGill & Pinkava 8699; Alaska, on road from Mexicali to Ti-
juana, Cota in 1932; 8 mi S of Machaguera on road to Hanson Lagoon, 3200 ft,
Wiggins 11254; summit of Cerro Bola, 1275 m (32°19’N 116°40’W), Moran 17821;
north slope of Sierra San Antonio Jenequa (Cerro Blanco) 800 m (32°05’N
116°30’W), Moran 8415; upper N slope of Cerro Blanco, Moran 17603; 1 mi S of
Rosa de Castilla (32°02’N 116°08’W) 1200 m, Moran 14945; 15 mi E of Ensenada,
Kappler 819; Kappler 8820; 20 mi E of Ensenada, Wiggins 11870; 20 mi S of Ense-
nada, Flemming in 1951; 1 mi E of San Antonio (31°59’N 116°36’W) 300 m, Moran
13966; Sierra Juarez, 7.5 mi SW of El] Rayo (31°56’N 116°04’W) 1330 m, Moran
16504; Cafion Donia Petra (31°56’N 116°36’W) 250 m, Moran 22803; 1 mi N of
Rancho Escondido (31°47’N 116°14’W) 800 m, Moran 13927; 2 mi NE of El Florido
(31°32’N 116°02’W), Moran 17691; Pine Canyon near San Antonio Mesa, Epling
& Stewart in 1936; Epling & Robison in 1940; 1.5-2.5 mi upstream from Rincon,
3 mi NE of Santa Catarina, 4250 ft, Broder 504; road from Valle Trinidad to Arroyo
Calentura, 3400 ft, Hohenthal 13; ridge 10 mi SW of Valle Trinidad (31°20’N
115°53’W), Moran 8221; Sierra San Pedro Martir near Las Encinas, 6000 ft, Powell
in 1958; 2 miS of Tepi (31°08’N 115°45’W) 1000 m, Moran 10971; above Rancho
San Pedro Martir on road to Sam’s Corral (31°04’N 115°35’W) 1900 m, Moran
14557 ; 10 km W of Rancho San Jose (Meling Ranch), Wiggins 20971; vicinity of
Rancho San Jose, 25 mi E of San Telmo, Meling 15; Campo Sotol (30°50’N
115°30’W) 1700 m, Wiggins 16560; 5 mi E of Rancho San Jose along old wagon
road to old mining camp of Socorro, 3500 ft, Wiggins 9793 ; old Socorro mining camp
E of Meling Ranch, 4200 ft, Blakley 7153; 4 mi W of Valladares (30°53’N 115°45’W)

154 MADRONO [Vol. 25

725 m (growing with T. lanatum), Moran 16426; 2 mi W of ex-Mision San Pedro
Martir (30°47.5’N 115°29’W) 1450 m, Moran 22172; Arroyo 3 mi S of Santa Eula-
lia (30°40’N 115°19’W) 1800 m, Moran 11159.

Section OrTHOPODIUM Bentham

T. austromontanum Lewis. Baja California: Near 30°55’N 115°38’W, 875 m,
Moran 16320.

T. lanceolatum Bentham. Baja California: Near large pond 1 mi E of Tijuana
Airport tower, Moran 16106; 24 mi N of Ensenada, Wiggins & Gillespie 3986.

T. micranthum Gray. Baja California: Hansen’s Ranch, Orcutt 1247; Rancho La
Botella (31°57’N 115°50’W) 1680 m, Moran 16474; Sierra Juarez, 1.5 mi S of
Rancho Marcos (31°58’N 115°52’W) 1700 m, Moran 13503; 3 mi S of Rancho
Marcos, Moran 13598.

LITERATURE CITED

Lewis, H. 1945. A revision of the genus Trichostema. Brittonia 5:276-303.
. 1960. Chromosome numbers and phylogeny in Trichostema. Brittonia
12:93-97.
WituiAMs, L. O. 1973. Eplingia, a new genus of the Labiatae from Mexico. Fieldi-
ana, Botany 36: 17-20.

Books RECEIVED AND LITERATURE OF INTEREST

Intermountain flora: vascular plants of the Intermountain West, U. S. A. By
ARTHUR CRONQUIST, ARTHUR H. HOLMGREN, NoeL H. HoLMcrREN, JAMES L. REVEAL,
and Patricia K. Hotmcren. Vol. 6: the monocotyledons. 584 pp. 1977. Columbia
University Press, New York. $54.00.

Flora of the Black Hills. By Ropert D. Dorw. Illustrations by JANE L. Dorn. x +
377 pp. 1977. Robert D. Dorn, P. O. Box 1471, Cheyenne, Wyoming 82001. $7.50
paperbound. Shipping extra on foreign and credit orders.

Atlas of United States’ Trees, vol. 4. Minor Eastern Hardwoods. By Elbert L.
Little, Jr. U.S. D. A. Misc. Publ. 1342: 17 pp., 230 maps. 1977. $8.75.

Rare and local trees in the national forests. By Elbert L. Little, Jr. U. S. D. A.
For. Serv. Conserv. Rep. 21: 14 pp. 1977.

A provisional checklist of species for Flora North America (revised). Ed. by Stan-
wyn G. Shelter and Laurence E. Skog. Monographs in Systematic Botany from the
Missouri Botanical Garden, vol. 1; Flora North America Report 84: xix + 199 pp.
1978. $6.50.

Intermountain biogeography: a symposium. Ed. by Kimball T. Harper and James
L. Reveal. Great Basin Naturalist Memoirs 2:1-268. 1978. $15.00. Brigham Young
University, Provo, UT 84602.

A revision of the Mexican Central American species of Cavendishia (Vacciniaceae).
By James L. Luteyn. Memoirs of the New York Botanical Garden 28(3) :1-138.
1976. $18.95.

A revision of the genus Declieuxia (Rubiaceae). By Joseph H. Kirkbride, Jr.
Memoirs of the New York Botanical Garden 28(4) :1-87. 1976. $10.95.

GOSSYPIUM TURNERI (MALVACEAE),
A NEW SPECIES FROM SONORA, MEXICO

PAut A. FRYXELL, Research Geneticist
Federal Research, Science, and Education Administration
U.S. Department of Agriculture,
Texas A & M University, College Station 77843

In spite of extensive study of the genus Gossypium for more than 100
years, new species continue to be discovered and described, enlarging
our understanding of this economically important and evolutionary inter-
esting genus. Fryxell (1966) estimated that new species of Gossypium
have been discovered and described in the first half of this century at an
average rate of about one species every five years, excluding the excessive
number of names that have been based on variants among the cultivated
species (cf. Fryxell, 1976). The number of species described in recent
decades has exceeded that rate (G. lobatum Gentry 1956, G. longicalyx
Hutch & Lee 1958, G. barbosanum Phillips & Clement 1963, G. nande-
warense Derera 1964, G. laxum Phillips 1972, G. pilosum Fryxell 1974,
G. nelsoni Fryxell 1974). The discovery of yet another species, de-
scribed herein, suggests that the ‘age of discovery” even in this much-
studied genus is not yet completed.

Gossypium turneri Fryxell, sp. nov. subsectionis Caducibracteolatae.
Frutex circa 1 m altus, nigropunctatus; ramulis tomentulosis, foliis trilo-
bis parvis 1.5—2.0 cm longis, petiolis laminis subaequantibus; pedunculis
petiolos consociatos aequantibus vel excedentibus, articulatis bractea in
nodo; bracteis involucellorum late rotundatis 3—7-laciniatis, per vel post
anthesin caducis; calycibus subtruncatis, petalis luteis ad basem rubris,
columna staminalis pallidis 16-17 mm longis, filamentis 5-6 mm longis;
fructibus 3-cellularis manifeste glandularis; seminibus ignotis.

Shrub ca. 1 m tall. Twigs densely stellate-tomentulose, the radii ca. 0.1
mm long, only partially concealing the underlying gossypol glands;
glands blackish, often somewhat raised, 0.1-0.2 mm diameter. Stems be-
coming woody, glabrate; bark red-brown with prominent lenticels 0.3-1
mm diameter. Leaf lamina cordate, shallowly trilobed, 1.5—2 cm long,
slightly broader than long, the margin entire, at apex acute to obtuse,
nearly glabrous above and below, palmately 5—7-veined, with prominent
black gossypol glands scattered throughout lamina, these somewhat more
abundant on margin. Foliar nectary single on the midrib beneath, 1-2
mm from base of lamina, ca. 1 mm long. Petiole stellate-tomentulose and
glandular, 1.2—2 cm long. Stipules 1-4 mm long, linear, black-glandular,
caducous. Peduncle axillary, solitary, ca. 2 cm long, equaling or exceed-
ing subtending petiole, with pubescence like petioles, articulate at or
above middle, bracteate at articulation, the bracts stipuliform, 1.5—4 mm

155

156 MADRONO [Vol. 25

long, early caducous. Nectaries of involucel 3, prominent, more or less
triangular, whorled at apex of pedicel. Bracts of involucel 3, inserted im-
mediately above nectaries, prominently punctate, subglabrous, basally
cuneate to truncate, broadly rounded proximally, 3—7-laciniate distally
(rarely entire), in outline ovate, 9-16 mm long, 7-9 mm wide, the long-
est (central) tooth up to 7 mm long, the bracts caducous at or shortly
after anthesis. Calyx 5-8 mm long, subtruncate with 5 inconspicuous
teeth, prominently and uniformly glandular, subglabrous or with a few
minute stellate hairs on margin. Petals 4—4.5 cm long, bright yellow with
small red spot at base, black-glandular except toward base, ciliate on claw
with hairs 1 mm long, densely and minutely stellate-pubescent externally
where exposed in bud, otherwise glabrous. Staminal column glabrous,
with a few black glands at base, pallid, 16-17 mm long, surmounted by
5 teeth, staminiferous in upper half; filaments pallid, 5-6 mm long; an-
thers one-celled, purplish, 1-2 mm long; pollen yellow-orange, spherical,
echinate; anther mass ellipsoid. Style exceeding staminal column by 5-15
mm, prominently glandular, sparsely pubescent below; stigmatic surfaces
densely pubescent, decurrent on style. Fruit a 3-celled capsule, 1-1.5 cm
long, globose to ovoid, beaked, with long (2 mm) white hairs along su-
ture margin after dehiscence but otherwise glabrous, with abundant and
prominent black glands externally; carpel wall 0.5 mm thick. Seeds
unknown.

Type: Mexico, Sonora, growing on windswept rocky outcrop near
beach, western base of Tetas de Cabra (near San Carlos Bay), lat.
27.9°N, long. 111.1°W; elev. 5 m, 24 July 1977, R. M. Turner & D. E.
Goldberg 77-49 (holotype: ARIZ; isotypes: CHAPA, MEXU, UC, pf).

Gossypium turneri finds a natural position in Gossypium subsection
Caducibracteolata Mauer (Fryxell, 1969), which also includes G. hark-
nessu Kellogg and G. armourianum Kearney. Phytogeographically the
subsection maintains its integrity: G. harknesst and G. armourianum
occur in the Baja California peninsula and on adjacent islands; the find-
ing of G. turneri across the Gulf of California in Sonora constitutes only
a minor extension of the range of the subsection.

Gossypium turneri shows its alliance with the other two species of sub-
section Caducibracteolata in the following characters: its caducous in-
volucellar bracts; its growth habit as a low shrub; its small, thick (xero-
morphic) leaves; its large yellow corolla with a red center; and its small,
3-locular, prominently glanded capsule that flares widely at maturity.

It may be distinguished from the other two species of the subsection
most easily by the laciniate involucellar bracts (fig. 1), but differs also in
several other characters, which are compared in Table 1. The color of the
seed hairs distinguishes G. armourianum (brownish) from G. harknessi
(whitish), but this character is as yet unknown for the new species.

The new species is named in honor of Raymond M. Turner, who first
collected it.

FRYXELL: GOSSYPIUM £57

1978]

O55
L1-91
sIsoyyue
6-2

9}[OIUL]-I} BAO

syeIq WAofI[NdIs [ews YIM

374
StsoyyUe
L-v

9} e[09DUP]

(¢) JUasge jDRIq

Ievoul|

Jeal o}e[otjed paonpat YM

(UIUL) SSoUyITY} [[eM jadie>)
(WU) Y4SUI] JUDIE
(WW) YjSUaT UWWNIOD [eUTUIe}S
SSO[ JO a3R4sS-
(uuu) YIPIM-
odeys-
[pRiqg LeppsonpoAuT

uole[NoyAe spunpag

(dan) SI-S'0 [=e (WD) YIsUa, BpUNpsg
JUIPIAI d1nosqo dInosqo (aoejins toddn) spurs jeoy
peqo] peqo] poqoyun WIAIO0} JeayT
1AQUuAN] “4D ussaugany 73) WNUDIANOWAD * +) ; Jayoeieyy

‘WIVTOFULOVAEIONGV,) NOILOASHANS WAIdASSOr) AO Sd10dd§ AMUN] AHL ONILVILNAYGAAIC SYALOVAVH,) 'T ATIVE

158 MADRONO [Vol]. 25

Fig. 1. Bud of Gossypium turneri with subtending leaf. A, involucellar nectary ;
B, articulation; C, foliar nectary; D, stipule. X 4.

———— nr—————— aes

1978] NOTES AND NEWS 159

LITERATURE CITED
FRYXELL, P. A. 1966. The need for taxonomic research in Gossypium L. Proc. 18th
Annual Cotton Improv. Conf. pp. 304-308.
. 1969. A classification of Gossypium L. (Malvaceae). Taxon 18:585-591.
1976. A nomenclator of Gossypium — the botanical names of cotton. U. S.
Dept. Agric. Tech. Bull. 1491. 114 pp.

NOTES AND NEWS

HEMIZONIA CONJUGENS (COMPOSITAE): DISTRIBUTION, CHROMOSOME NUMBER,
AND RELATIONSHIPS — Hemizonia conjugens Keck (Aliso 4:101-109, 1958; type:
River bottom land near Otay, San Diego County, California, Abrams 3521, UC!;
isotypes, CAS!, POM!, NY) is a rare and poorly understood annual previously
known only from southwestern San Diego County. Prior to this report this species
was represented by only five other collections (Abrams 3902, Stokes s.n., Hall 3891,
Hall 3916, Wolf 7969) from near Otay Mesa, south of Chula Vista. Owing to the
absence of this species from recent collections of the area, plus the extensive subur-
ban and agricultural development of the area, thereby eliminating many natural
habitats, it was considered possibly to be extinct. This area is fascinating as a result
of its high degree of endemicity.

In May 1977, I was informed by Dr. Reid Moran of the San Diego Museum of
Natural History that he had discovered a population that was apparently referable
to the original description of Hemizonia conjugens (Moran 24152: Tanowitz 16606).
The population was growing in mildly disturbed sandy loam, approximately 3 km
SW of La Presa, Baja California (32°26’N 116°56’W). These plants were densely
distributed over several acres. This represents a definite range extension for the spe-
cies. Collections were made for morpholoical, cytological, and phytochemical studies;
specimens will be distributed to various herbaria. A chromosome number, = 12, is
reported for this species for the first time.

Keck (Munz, A California flora, 1959; op. cit.) assigned H. conjugemns to sect.
Deinandra based upon its annual habit, sterile disk achenes, keeled phyllaries, and
receptacular bracts of a single series surrounding the outermost series of disk florets.
He considered H. conjugens to be most closely related to H. fasciculata T. & G.
(1 = 12; Johansen, 1933; Tanowitz, ined.) and H. paniculata ssp. paniculata
A. Gray (1 = 12; Johansen, 1933; Tanowitz, ined.). Based on morphological traits
and geographical distribution, H. conjugens appeared to be intermediate to these
species. Among the most significant of these of the former are the sterile and more
or less glabrate disk achenes and among the most significant of the latter the number
of ray and disk florets. They also show intermediacy in pubescence and distribution
of pubescence and growth habit. Hence, Keck (op. cit.) postulated it to be an
amphidiploid derivative of a hybrid between the two. This does not appear to be
the case since the gametic chromosome number of H,, conjugens is the same as the
other two species: however, this evidence does suggest strongly that they are indeed
related. Furthermore, preliminary flavonoid analyses of exudate aglycones display
interesting patterns as well. Patuletin, quercetin, and two highly methylated flavo-
nols are common to all three species. There is at least one flavonoid (methylated
flavone) and one phenolic that is unique when compared to the other two.

The morphological, cytological, chemical, and geographical data strongly indicate
that these three species are closely related and may have arisen either from the same
ancestral stock or as a stabilized, diploid hybrid derivative of Hemizonia paniculata
and H. fasciculata. Cytogenetical and more extensive chemical investigations are in
progress to elucidate these speculations further.

I thank Dr. Reid Moran for his aid in this study and Dr. Robert Patterson for
comments on the manuscript. This study was supported by a grant from the Regents
of the University of California, PFG 5347. — Barry D. Tanowirtz, Department of
Biological Sciences, University of California, Santa Barbara 93106.

CHROMOSOME NUMBERS IN ASTERACEAE

A. MICHAEL POWELL and SHIRLEY A. POWELL
Department of Biology, Sul Ross State University
Chihuahuan Desert Research Institute, Alpine, Texas 79830

Reports of 135 new chromosome number counts are recorded for 100
species and 54 genera of Asteraceae mostly from North America. Three
genera, Tessaria (n = 10), Carramboa (n = 19), and Pelucha (n = 10),
and 15 species were previously unreported. Discussion is limited to those
taxa for which chromosome numbers are previously unreported or those
published counts that suggest a need for some elaboration. Many counts
presented here confirm earlier reports listed in the chromosome number
indexes edited by Federov (1969) and Moore (1973; 1974), or in other
recent literature.

The chromosome numbers reported here were obtained through stand-
ard acetocarmine squashes of microsporocytes. Voucher specimens for
most of the collections are deposited in TEX or SRSC. Vouchers for col-
lections by Breedlove, Roe, and Raven are deposited in DS. Asterisks in
Table 1 denote taxa for which chromosome numbers are previously unre-
ported. Abbreviations of collectors preceding collection numbers in Table
1 include: B, Babcock; Br, Breedlove; I, Ittner; Pa, Patterson; P,
Powell; R, Raven; S, Sikes;T, Tomb; Tu, Turner; W, Weedin.

Astereae — Erigeron mimegledes (2n = 271) was reported earlier as
2n = 361 by Turner and Flyr (1966) who discussed the species as pos-
sibly apomictic on a base of x = 9 and probably related to E. geiseri
(2n = 271), conclusions substantiated by our count in Table 1. Evigeron
sp. (n = 18; Table 1) is an undescribed species. Determinations for
Machaeranthera pinnatifida (n = 4, 8; Table 1) follow Hartman (1976)
and Turner and Hartman (1976).

Inuleae — The chromosome number listed in Table 1 for Tessaria
sericea (n = 10) isa first report for the genus and supports what appears
to be a close relationship of this genus with Pluchea (x = 10).

Heliantheae — Although the count for Aspilia purpurea (n = 12 or
13; Table 1) is tentative and differs from previous reports of x = 14
(Solbrig et al., 1972) and m = ca. 17 (Powell and Cuatrecasas, 1970)
for four other species of the genus, there is reason to suspect that this
large taxon of North America, South America, and Africa might be chro-
mosomally variable and thus amenable to cytotaxonomic investigation.
Bidens ostruthioides (n = 12) has been reported previously as a tetra-
ploid, but a diploid number for the taxon is recorded in Table 1. Calea
scabra is reported to be m = ca. 27 (Table 1) evidently on a base of
x = 9 (Solbrig et al., 1972) although this species has been recorded pre-
viously as m = 16, 32 (De Jong and Longpre, 1963; Powell and Turner,
1963). The listing in Table 1 for Coreopsis integrifolia (n = 13), appar-

160

microce phala

Haplopappus divaricatus (Nutt.) Gray 4

Heterotheca villosa (Pursh) Shinners
Isocoma corono pifolia Greene

Machaeranthera brevilingulata

18

6

ca. 9

(Sch.-Bip. & Hemsl.) Turner & Horne

Machaerantha gracilis (Nutt.)
Shinners

Machaerantha gracilis (Nutt.)
Shinners

Machaeranthera gymnocephala (DC.
Shinners

2

gb

) 4

1978] POWELL & POWELL: CHROMOSOME COUNTS 161
TABLE 1. CHROMOSOME NUMBERS IN ASTERACEAE.
Gametic Locality
Species chromosome and
number collection number
EUPATORIEAE
Ageratina wrightii (Gray) i, TEX. Hudspeth Co. Sierra Blanca
King & Robins. Mts., Wilson s.n.
Ageratum nelsonii (Rob.) 10 MEX. Chiapas: 11 mi S of La
M. F. Johnson Trinitaria, Br 13269.
Barrotea laxiflora Brandeg. 9 MEX. Sinaloa: 45 mi S of Mazatlan,
S and B 204.
Carphochaete bigelovii Gray ier TEX. Brewster Co. Chisos Mts.,
Averett 240.
Piquieria trinervia Cav. 7 MEX. San Luis Potosi: ca. 13 mi W
of Ahualulco, P and T 2589.
Stevia lucida Lag. var. lucida 12 MEX. Durango: 3 mi NW of
Morcillo, S and B 374.
Stevia ovata Willd. ca.33 I MEX. Oaxaca: 3 km E of Ixtlan de
Juarez, Br 12244.
ASTEREAE
Aphanostephus ramosissimus DC. 4 MEX. Chihuahua: 25 mi § of
var. humilis (Bentham) Birdsong Jiménez, S and Pa 404.
Aphanostephus riddelii T. & G. 5 MEX. Coahuila: 43 mi NW of
Muzquiz, P, Pa, and I 1576.
Aster subulatus Michx. var. ligulatus 5 TEX. Maverick Co. Eagle Pass,
Shinners Hudson 16.
Astranthium purpurascens (Rob.) 8 MEX. Chiapas: Cerro San
Larsen Cristobal, Br 13298.
Erigeron mimegledes Shinners 271 MEX. Coahuila: 43 mi NW of
Muzquiz, P, Pa, and I 1585.
* Erigeron sp. 18 MEX. Coahuila: high pass E of
Rancho E] Jardin, P, Pa, and I 1602.
Grindelia havardi Steyerm. 12° TEX. Brewster Co. 13 mi S$ of
Alpine, P 2453.
Grindelia microcephala DC. var. 6 TEX. Dimmit Co. 10 mi S of

Catarina, P 2348 ; Maverick Co. 13
mi E of Eagle Pass, P 2349.

TEX. Midland Co. 4 mi E of
Midland, R 19219,

TEX. Hudspeth Co. 12 mi E of

Van Horn, P 2782.

MEX. San Luis Potosi: ca. 13 mi W
of Ahualulco, P and T 2587.

MEX. San Luis Potosi: ca. 13 mi W
of Ahualulco, P and T 2590.

TEX. Brewster Co. Alpine, Hudson
1; 20 miS of Alpine, Sloan 2.

TEX. Brewster Co. Alpine, Young 2.

MEX. Durango: 18 mi N of
La Zarca, S and B 365.

162 MADRONO

Machaeranthera pinnatifida (Hook.) 4
Shinners var. chihuahuana Turner &
Hartman

Machaeranthera pinnatifida (Hook.) 7
Shinners var. chihuahuana Turner &
Hartman

Machaeranthera pinnatifida (Hook.) 8”
Shinners var. chihuahuana Turner &
Hartman

Machaeranthera pinnatifida (Hook.) 4
Shinners var. pinnatifida

Machaeranthera pinnatifida (Hook.) 8
Shinners var. pinnatifida

INULEAE
Pluchea odorata (L.) Cass. 10
*Tessaria sericea (Nutt.) Shinners 10
PALAFOXIINAE
Palafoxia callosa (Nutt.) T.& G. 10
Palafoxia texana DC. var. texana 11

Palafoxia sphacelata (Nutt. & Torr.) 12

Cory
HELIANTHEAE
Aspilia purpurea Greenm. 12 or 13
Bahia pedata Gray 12
Baltimora recta L. 15

Bidens ostruthioides (DC.) Sch.-Bip. 12
Calea scabra (Lag.) Rob. ca. 27
*Carramboa littlei (Cuatr.) Cuatr. 19

Coreopsis integrifolia Poir. 13

[Vol. 25

MEX. Chihuahua: Sierra de
Arbolitos, P, Tu and S 2505; TEX.
Brewster Co. Big Bend Park
vicinity, Tw s.n.

TEX. Presidio Co. Charro Canyon,
P 2774.

MEX. Chihuahua: 25 mi S of
Jiménez, S and Pa 405.

MEX. Coahuila: 43 mi NW of
Muzquiz, P, Pa, and J 1577; TEX.
Brewster Co. Alpine, Sloan 4;
Culberson Co. 8 mi S of Van Horn,
S and B 324; El Paso Co. 10 mi E
of El Paso, S and B 321; Hudspeth
Co. 32 mi E of El Paso, S and B
322; 37 mi E of El Paso, S and B
323; Pecos Co. 14 mi FE, of Ft.
Stockton, S 438; S 439; Presidio
Co. Marfa, Foster 18.

TEX. El Paso Co. 10 mi W of

El Paso, S and B 318; Hudspeth Co.
ca. 12 mi E of Dell City, P, P, and
W 2837.

MEX. Sinaloa: 2 mi N of Mazatlan,
S and B 198.

TEX. Hudspeth Co. Quitman Mts.,
P, P, and W 2941.

TEX. Val Verde Co. 1 mi S of
Pandale, P, P, and W 2965.

TEX. Maverick Co. Eagle Pass,
Hudson 17.

TEX. Winkler Co. 2 mi N of Wink,
Foster 23.

MEX. Chiapas: 19 km N of
Arriaga, Roe 853.

NEW MEX. DeBaca Co. Lake
Sumner, W and W 334.

MEX. Chiapas: 5 km N of Huixtla,
Roe 830; Yucatan: 2 km § of
Tekax, Roe 1333.

MEX. Chiapas: 9 mi SE of San
Cristobal Las Casas, Br 13426.
MEX. Chiapas: Lagunas de
Montebello, Roe 965.
VENEZUELA. Merida: La
Carbonera, Lopez and Ruiz 14013.
S. CAR. Colleton Co. R 20466.

1978]

Eclipta alba (L.) Hassk.

Encelia scaposa (Gray) Gray

Engelmannia pinnatifida T. & G.

Helianthella quinquenervis (Hook.)
Gray

Helianthus annuus L.

*Helianthus paradoxus Heiser

Hymenopappus filifolius Hook. var.
cinereus (Rydb.) I. M. Johnston

Hymenothrix wislizenii Gray
Hymenoxys odorata DC.

Tostephane trilobata Hemsl.

Melampodium divaricatum
(Rich.) DC.

Melam podium gracile Less.
Melam podium leucanthum T.& G.
Perymenium cf. asperifolium
Sch.-Bip. & Klatt
Perymenium grande var. nelsonii
(Robins. & Greenm.) Fay
*Perymenium klattianum Fay
Perymenium mendezii DC. var.
verbesinioides (DC.) Fay
Perymenium cf. purpusi Brandeg.
Psilostrophe tagetina (Nutt.) Greene
Psilostrophe tagetina (Nutt.) Greene

Ruizlopezia bromeiicides (Cuatr.)
Cuatr.

Sanvitalia ocymoides DC.

Sanvitalia procumbens Lam.

Ca.

Ca.

Ca.

POWELL & POWELL: CHROMOSOME COUNTS

ty

1 gb

34

12

163

MEX. Baja Calif. Sur: 1 mi N of
Pescadero, S and B 250.

MEX. Chihuahua: 5 mi SW of
Coyame, P, Tu, and S 2476.

TEX. Brewster Co. Alpine,

Sloan 20.

NEW MEX. Otero Co. Sierra
Blanca, R 20348.

TEX. Brewster Co. Alpine,

Sloan 21.

TEX. Pecos Co. N of Ft. Stockton,
Kolle 1415.

COLO. San Miguel Co. 27 mi NE
of Egner, Tu 8051; NEW MEX.
Torrance Co. 8 mi E of Willard,
P2582.

ARIZ. Cochise Co. Benson, R 20341.
MEX. Chihuahua: 10 mi § of
Ojinaga, P, Tu, Magill 2016; 56 mi
S of Ojinaga, P, Tu, and

Magill 2018.

MEX. Chiapas: San Cristobal las
Casas, Br 6742.

MEX. Mexico: 3 km S of
Temascaltepec, Roe 1543; Oaxaca:
Oaxaca, Br 12193 ; Sacatapequez:
5 km N of ject. of roads 14 and 6,
Roe 825; Sinaloa: 70 mi § of
Mazatlan, S and B 207.

MEX. Yucatan: 25 mi S of Uman,
Roe 1317.

TEX. Jeff Davis Co. near Brack’s
Tunnel, S and B 312.

MEX. Oaxaca: 24 mi NE of Sola
de Vega, By 12287.

MEX. Chiapas: 11 miS of La
Trinitaria, Br 132067.

MEX. Vera Cruz: Cumbres de
Acultizingo, Roe 1278.

MEX.: 2 km N of Ixtapan,

Roe 1917.

MEX. Oaxaca: 3 km E of Ixtlan de
Juarez, Br 12227.

TEX. Presidio Co. 2 mi NW of
Marfa, Foster 22.

NEW MEX. Sandovall Co. 20 mi
NE of Bernalillo, Tu 8027.
VENEZUELA. Merida: Potreros
de San Rafael, Lopez and Ruiz
14017.

MEX. Durango: 14 miS of Rodeo.
S and B 367.

MEX. Chiapas: 15 mi E of
Cintalapa, Roe 878.

164 MADRONO

Simsia anplexicaulia (Cav.) Pers. 17

Simsia calva (Gray & Engelm.) Gray 17

*Szmsia sanguinea Gray Casts
Verbesina encelioides (Cav.) Gray 17
*Verbesina hy poglauca Sch.-Bip. CARL.
Viguiera stenoloba Blake Ca.33
Zinnia elegans Jacq. 12
Zinnia littoralis Robins. & Greenm. ca. 10

TAGETEAE
Dyssodia pentachaeta (DC.) Robins. 8

PERITYLINAE
*Peritvle ajoensis Todsen 17

Perityle angustifolia (Gray) ca. 68
Shinners
*Peritvle carmenensis Powell i

Perityle ciliata (L.H. Dewey) Rydb. 17

Perityle emoryi Torr. ca. 34-36
Perityle emoryi Torr. ca. 36
Perityle leptoglossa Harv. & Gray 17
Perityle parryi Gray 17
FLAVERIINAE
Flaveria aiustralasica Hook. ca. 18
Flaverina bidentis (L.) Ktze. 18
Flaveria bidentis (L.) Ktze. 18
*Flaveria sp. 18
Flaveria campestris J. R. Johnst. 18

[Vol. 25

MEX. Chiapas: San Cristobal las
Casas, Br 13283.

TEX. Brewster Co. Alpine, Paul 1,
MEX. Oaxaca: 24 mi NE of Sola de
Vega, Br 12254.

TEX. Presidio Co. 31 mi SE of
Marfa, Sloan 15; Marfa, Foster 12.
MEX. Oaxaca: 3 km E of Ixtlan de
Juarez, Br 12250.

TEX. Presidio Co. 31 mi SE of
Marfa, Sloan 13.

MEX. Mexico: 56 km S of
Temascaltepec, Roe 1652.

MEX. Sinaloa: 2 mi N of Mazatlan,
S and B 196.

TEX. Garza Co. 2.5 mi E of Post,
R 19300.

ARIZ. Pima Co. Organ Pipe Cactus
National Monument, Todsen 2292
(seed progeny).

TEX. Val Verde Co. ca. 25 mi N of
Langtry, P, P, and W 2953.

MEX. Coahuila: Maderas del
Carmen, Adamcewicz and Wendt
525 (seed progeny).

ARIZ. Santa Cruz Co. 3 mi SW of
Patagonia, Watson 234.

SOUTH AMERICA. Chile?, Robres
13-XI-1973 (seed progeny).
SOUTH AMERICA. Chile: Prov.
Atacama: Caldera, Richardson 2176
(seed progeny).

MEX. Sonora: ca. 17 mi SE of
Magdalena. Van Devender s.n.
(seed progeny).

TEX. Presidio Co. 2-3 mi NW of
Fresno Mine, P 2201.

AUSTRALIA. Demarz 5843 (seed
progeny).

DOMINICAN REPUBLIC. Loigier
18314 (seed progeny).

ECUADOR. King 6925.

TEX. San Patricio Co. Port
Aransas, Mustang Island, P 2802;
Nueces Co. Padre Island, Urbatsch
1547 (seed progeny).

NEW MEX. Socorro Co. E of
Socorro, P and P 3011; San Acacia,
Valentine, sn. (seed progeny).

1978]
Flaveria chloraefolia Gray
*Flaveria sp.
Flaveria floridana J. R. Johnst.
Flaveria linearis Lag.

Flaveria mcdougallii Theroux,
Pinkava, & Keil

Flaveria oppositifolia (DC.) Rydb.

Flaveria pringlei Gdegr.

Flaveria pringlei Gdgr.

Flaveria pringlei Gdegr.

Flaveria pubescens Rydb.

Flaveria ramosissima Klatt.

*Flaveria sp.
Flaveria trinervia (Spreng.) Mohr

Sartwellia flaveriae Gray

Sartwellia mexicana Gray
Sartwellia puberula Rydb.
SENECIONEAE
Bartlettia scaposa Gray
*Pelucha trifida Wats.
Raillardella argentea Gray
Raillardella muirit Gray

* Senecio ci. chicharrensis Greenm.

POWELL & POWELL: CHROMOSOME COUNTS

18

18

ca. 18

18

18

18

18

ca. 36

36

18

18

ca. 30

165

MEX. Coahuila: ca 6 mi SW of
Cuatro Ciénegas, Tu 6168.

MEX. Puebla: 5 mi SW of
Tehuacan, Br 14189.

FLOR. Hillsborough Co. Tampa,
Anderson s.n. (seed progeny).
FLOR. St. Lucie Co. 37 mi S of
Ft. Pierce, Luteyn 2885 (seed
progeny).

ARIZ. Mohave Co. Grand Canyon
National Park, Cove Canyon,
Theroux 1675 (seed progeny).
MEX. Coahuila: 3 mi NW of
Nadadores, P and Tu 2710.

MEX. Oaxaca: 30 mi NNW of
Huajuapan de Leon, Tu P-50;
Puebla: 5 km NW of Petalaleingo,
Rzedowski 28942 (seed progeny).
MEX. Oaxaca: 6 mi NW of
Huajuapan de Leon, Dillon 681
(seed progeny).

MEX. Puebla: 16 mi S of
Esperanza, Hartman, et al. 3834.
MEX. San Luis Potosi: 1 mi SE of
Rio Verde, Hartman, et al. 3823
(seed progeny).

MEX. Puebla: 8 mi SE of
Coxcatlan, Anderson and Anderson
5340 (seed progeny).

MEX. Sonora: near San Bernardo,
Martin 5 (seed progeny).

TEX. Culberson Co. Rustler
Springs, P and P 3037.

TEX. Ward Co. 5 mi 5S of Pyote,
Sloan 66. Reeves Co. 1 mi N of
Pecos, P 2927.

MEX. San Luis Potosi: ca. 31 mi S
of Matehuala, S, Olsen, and P 820.
MEX. Coahuila: ca. 22 mi N of
Monclova, S, Olsen, and P 854.

TEX. Hudspeth Co. Malone Hills,
E of Tommy’s Town, P 2418.
MEX. Baja California: Isla San
Pedro Martir, Moran 21745.
CALIF. San Bernardino Co. Mt.
San Gorgonio, R 11148.

CALIF. Fresno Co. Tehipite
Valley, Howell 33960.

MEX. Chiapas: 17 km NW of
Ocozocoautla, Roe 891.

166 MADRONO [Vol. 25

LACTUCEAE
Pinaropappus roseus Less. 9 TEX. Pecos Co. 5 mi SW of Iraan,
S 441; Terrell Co. ca. 15 mi W of
Longfellow, P 2673; Val Verde Co.
just N of Langtry, P 2680.
Pinaro pap pus roseus Less. 9-154 MEX. Coahuila: 104 mi NW of
Muzquiz, P, Pa, and I 1596.
*Pinaro pappus sp. nov. 9 MEX. Coahuila: ca. 92 mi NW of
Muzquiz, P, Pa, and I 1593.
Stephanomeria pauciflora (Torr.) 8 TEX. El Paso Co. Hueco Mts.,
A. Nels. P and P 2991; Hudspeth Co. 15 mi
W of Van Horn, S 463.
MUTISIFAE

*Acourtia nana (Gray) Reveal& King 27 TEX. Hudspeth Co. 15 mi W of
Van Horn, S 462.

“Occasionally with one ITI and one I.

’Consistently with one round, probably centric, fragment.

“Consistently one I or fragment; also observed in a few cells were 17 II + 3 I; 18 II
+ 5 small fragments.

“Meiosis irregular, suggestive of triploidy on a base of x = 9, or apomixis.

ently a rare and localized species of South Carolina and Georgia, is a cor-
rection of an earlier publication in Solbrig et al. (1972) where the taxon
was erroneously reported as C. pubescens. Cuatrecasas (1976) recently
segregated Carramboa little: and Ruizlopezia bromelioides, both n = 19
(Table 1), from Espeletia (x = 19).

Tageteae — The count of m =8 would appear to establish a diploid
base for Dyssodia pentachaeta, a species for which Strother (1969) lists
only m = 13, 16, or 26. Keil and Stuessy (1975) have reported 1 = 16
for D. pentachaeta var. belenidium. A varietal identification has not been
made for our collection. Strother (1969) records nm = 8 for several other
species of Dyssodia.

Peritylinae — In Table 1 the subtribe Peritylinae is positioned be-
tween the Heliantheae and Senecioneae (the classical but unnatural
Helenieae is no longer recognized) in deference to those who do not wish
to accept the taxon into homogeneous versions of tribes (Turner and
Powell, 1977). Perityle ajoensis (n = 17, Table 1) was described (Tod-
sen, 1974) after a monograph (Powell, 1973) of Perityle sect. Laphamia,
to which it belongs, was completed. Todsen suggested a relationship of
P. ajoensis to P. palmeri (= P. tenella; n = 16), P. inyoensis (n = ca.
18) and P. megalocephala var. megalocephala (n = 17). In our view,
P. ajoensis doubtless belongs with the ‘‘southwestern alliance’ of seven
species including the three above but is perhaps closest to P. villosa (a
triploid from one previous count) and P. inyoensis. The n = ca. 68 for
P.. angustifolia adds a new ploidy level to a species where m = 17 and
n = ca. 51 are previously known (Powell, 1968; Powell, 1973). The
chromosome number for the newly described P. carmenensis (n = 17) is

1978] POWELL & POWELL: CHROMOSOME COUNTS 167

consistent with the base number established for sect. Laphamia (Powell,
1973) and with the base number of its related species P. dissecta, P. cas-
tilloni, and P. lemmonii (Powell, 1976). The two counts for P. emoryi
(n = 34-36) listed in Table 1 are the first reports of this bicontinental
taxon from South America. The principal distribution of P. emoryi is in
Baja California and Sonora, Mexico, and in parts of the southwestern
United States where hexaploid and tetraploid chromosome numbers of
n = 50-56 (the common number throughout the range) and 2 =32-36
(from one collection in Baja California) are known (Powell, 1974).
Raven (1963) suggested that P. emoryi reached Chile and Peru by long-
distance dispersal (in late Pliocene or Pleistocene), a subject which is
discussed further by Powell (1974). The present chromosomal data sug-
gest the possibility that South American P. emoryi (n = 34-36) might
have originated from Baja California P. emoryi (n = 32-36) at a time
when tetraploid populations were more widespread.

Flaveriinae — The species of Flaveria and its allied genera Sartwellia
and Haploésthes are consistently n = 18, except for a single case of
polyhaploidy (z = 9) in F. campestris (Anderson, 1972), and the two
tetraploid (x = 36) collections of F. pringlei in Table 1. The latter spe-
cies, related to F. angustifolia and F. vaginata, is also diploid (Table 1).
The three taxa listed as Flaveria sp. are undescribed. With the data pre-
sented here and elsewhere (Turner, 1975; Turner, 1971; Powell and
Powell, 1977) a base number of x = 18 seems certain for the Flaveriinae.
Chromosome numbers are lacking for only four of the 21 species of Fla-
veria (F. angustifolia, F. intermedia, F. robusta, and F. vaginata) ; chro-
mosome numbers are known for all four species of Sartwellia (Table 1)
and for all but one taxon of Haploésthes (H. fruticosa; Turner, 1975).

Counts for the Flaveriinae in Table 1 were accumulated in connection
with a forthcoming revision of Flaveria by the senior author. The study
has revealed in the earlier literature two chromosome number reports
that were attributed erroneously to certain species, and these are cor-
rected as follows: F. anomala, Graham & Johnston 4818, reported as
F. ramosissima in Turner and Johnston (1961); F. anomala, Powell &
Edmondson 542, reported as F. ramosissima in Powell and Turner
(1963).

Senecioneae — The count in Table 1 for Bartlettia scaposa (n = 11)
is consistent with the only previous report for the species (from Mexico),
which is distributed almost entirely in Chihuahua and Durango, Mexico
(Powell, 1963). Pelucha trifida (n = 19) is an anomalous genus with
uncertain tribal affinity, but its chromosome number and morphology
suggest that it might belong in a group with several other North Ameri-
can Senecioid genera with « = 19.

Mutisieae —- The chromosome number of Acourtia (= Perezia) nana
(n = 27) is the same as its two geographically related species A. wrighti
and Perezia runcinata and consistent with the base number of « = 9 so
far indicated for Acourtia (Powell and Sikes, 1970).

168 MADRONO [Vol. 25

ACKNOWLEDGMENTS

Supported in part, particularly the Flaveriinae, by NSF Grant BMS
73-06851-A0O1 to the senior author. We are grateful to Peter Raven for
contributing numerous counts; Reid Moran for buds and voucher of
Pelucha; B. L. Turner for help with some identifications and for provid-
ing bud material; Dennis Breedlove, Al Richardson, Tom Van Devender,
Loran Anderson, and other collectors as indicated in Table 1 for contrib-
uting seed and bud material; Jim Weedin for assisting with numerous
counts; R. M. King for providing identifications for some Eupatorieae;
Tod Stuessy for identifications of Melam podium; and Don Kolle for con-
tributing the count of Helianthus paradoxus.

LITERATURE CITED

ANDERSON, L. C. 1972. Flaveria campestris (Asteraceae): A case of polyhaploidy or
relic ancestral diploidy ? Evolution 26:671-673.

CuaTRECasas, J. 1976. A new subtribe in the Heliantheae (Compositae) : Espeletiinae.
Phytologia 35:43-61.

DeJonc, D. C. D. and E. K. Loncpre. 1963. Chromosome studies in Mexican Com-
positae. Rhodora 65:225-240.

FEeperov, A. A. [ed.]. 1969. Chromosome numbers of flowering plants. Acad. Sci.
U.S. S. R., Komarov Botanical Institute, Leningrad.

Hartman, R. L. 1976. A conspectus of Machaeranthera (Compositae: Astereae) and
a biosystematic study of the section Blepharodon. Ph.D. Disseration, Univ. of
Texas, Austin.

Ker, D. J. and T. F. Struessy. 1975. Chromosome counts of Compositae from the
United States, Mexico, and Guatemala. Rhodora 77:171-195.

Moore, R. J. [ed.]. 1973. Index to plant chromosome numbers 1967-1971. Regnum
Veg. 90:1-539.

. 1974. Index to plant chromosome numbers for 1972. Regnum Veg.
91:1-108.

Powe 11, A. M. 1963. An emended description of the monotypic genus Bartlettia
A. Gray (Senecioneae) with distributional notes. Southwest. Nat. 8:117-120.

. 1968. Chromosome numbers in Pervityle and related genera (Peritylinae—
Compositae). Amer. J. Bot. 55:820-828.

. 1973. Taxonomy of Perityle section Laphamia (Compositae—Helenieae—
Peritylinae). Sida 5:61-128.

. 1974. Taxonomy of Perityle section Perityle (Compositae—Peritylinae).

Rhodora 76:229-306.

. 1976. A new species of Perityle (Asteraceae) from Coahuila, Mexico. Sida
6:311-312.

and J. Cuatrecasas. 1970. Chromosome numbers in Compositae: Colom-
bian and Venezuelan species. Ann. Missouri Bot. Gard. 57:374-379.

and S. Powell. 1977. Chromosome numbers of gypsophilic plant species of
the Chihuahuan Desert. Sida 7:80—90.

and S. Sikes. 1970. Chromosome numbers of some Chihuahuan Desert Com-
positae. Southwest. Nat. 15:175-186.

and B. L. Turner. 1963. Chromosome numbers in the Compositae. VII. Ad-
ditional species from the southwestern United States and Mexico. Madrono
17:128-140.

Raven, P. H. 1963. Amphitropical relationships in the floras of North and South
America. Quart. Rev. Biol. 38:151-177.

Sotsric, O. T., D. W. KyHos, M. Powett, and P. H. Raven. 1972. Chromosome

1978] NOTES AND NEWS 169

numbers in Compositae. VIII: Heliantheae. Amer. J. Bot. 59:869-878.
STROTHER, J. L. 1969. Systematics of Dyssodia Cavanilles (Compositae: Tageteae).
Univ. Calif. Publ. Bot. 48:1-88.
Topsen, T. K. 1974. A new species of Perityle ““Compositae” from Arizona, J. Ariz.
Acad. Sci. 9:35.
Turner, B. L. 1971. Taxonomy of Sartwellia (Compositae—Helenieae). Sida
4:265-273.
1975. Taxonomy of Haploésthes (Asteraceae—Senecioneae). Wrightia
5:108-115.
and D. Flyr. 1966. Chromosome numbers in the Compositae. X. North
American species. Amer. J. Bot. 53:24—-33.
and R. Hartman. 1976. Infraspecific categories of Machaeranthera pinna-
tifida (Compositae). Wrightia 5:308-315.
and M. C. Johnston. 1961. Chromosome numbers in the Compositae—III.
Certain Mexican species. Brittonia 13:64-69.
and A. M. Powell. 1977. Helenieae — systematic review. In The Biology
and Chemistry of the Compositae. vol. 2. V. H. Heywood, J. B. Harborne, and
B. L. Turner [eds.]. Academic Press, N. Y.

NOTES AND NEWS
New PLAnt DIsTRIBUTION RECORDS FROM THE SOUTHWESTERN UNITED STATES AND
NorTHERN Mexico. — Recent collecting activities in the Southwestern United States
and northern Mexico have turned up several new or otherwise significant plant dis-
tribution records. Some of these have developed significance after passage of the En-
dangered Species Act and the subsequent publication of state lists of threatened or
endangered species.

FABACEAE

Astralagus musimonum Barneby. Arizona, Mohave Co., ca. 30.5 km S of St.
George, Utah, at 1350 m on the W slope of Seegmuller Mountain above Mokiah
Wash, on calcareous bank along read in pihon-juniper community, R. Spellenberg,
R. Delson, J. Syvertsen 3182, 20 May 1973.

In Arizona Flora (T. H. Kearney and R. H. Peebles. 1951. Univ. Calif. Press)
this species was listed as possibly occurring within Arizona, the record based on a
collection from the same vicinity (Ripley and Barneby 4321) as our recent collection.
Their collection was immature and Barneby was uncertain of the correct identity of
that material. In the supplement bound with a later edition of Arizona Flora (1964:
p. 1054) Barneby concluded their collection was misidentified, actually representing
Astragalus amphioxys Gray var. modestus Barneby. He maintained this opinion in
his monumental “Atlas of North American Astralagus” (1964. Mem. New York Bot.
Gard. Vol. 13). Our collection “keyed” easily to A, musimonum since it had good
fruit and a few flowers. The identification was confirmed by Barneby. In our cor-
respondence regarding the collection he noted that his and Ripley’s earlier collection
had been lost. The species appears in the 7/1/75 Federal Register as “threatened” in
Nevada.

Petalostemum scariosum (Wats.) Wemple. New Mexico, Socorro Co., 1.6 km § of
La Joya Game Refuge exit on IH-25, R. Spellenberg, J. M. Willson 4228, 8 Jul 1976.

The species is listed as “endangered” in New Mexico in the 6/16/76 Federal Reg-
ister. In a recent revision of the genus (D. K. Wemple, 1970, Iowa State Coll. J.
Sci. 45:1—102) all collections cited were made about 70 years ago from only two
stations near the Rio Grande in Bernalillo and Valencia counties. However, we find
the species to be more widespread, occurring from the Laguna Indian Reservation
in eastern Valencia Co. to the Sandia Mountains in central Bernalillo Co., south-
ward to east of San Antonio in central Socorro Co. The species may have been orig-

170 MADRONO [Vol. 25

inally restricted to eroded tops and slopes of sandy-clay bluffs but seems to respond
positively to disturbance, occurring sporadically in loose sand near roadsides and
other sandy areas now believed to have been degraded by domestic grazing. Observa-
tions made in the spring of 1977 on a population in central Socorro Co. indicate that
the species is not, or but little, grazed by cattle and horses on otherwise heavily used
rangeland.

NYCTAGINACEAE

Selinocarpus palmeri Hemsl. Coahuila, 4 km by winding road E of El Coyote at
NW end of Sierra de Solis, 25°40’N lat., 103°10’W long., elev. ca. 1100 m, on a large,
almost pure gypsum outcrop, R. Spellenberg and J. Syvertsen 3768, 16 Aug 1974.

Until this new collection the species was known from only two others (Palmer
1118, 1119, May 1880, from San Lorenzo de Laguna). But perhaps more important
is evidence of the precise location of Palmer’s “San Lorenzo de Laguna’, not to be
located on modern maps. McVaugh (1956. Edward Palmer: plant explorer of the
American West. Univ. Okla. Press) indicated that it should be in the near vicinity of
the site from which our collection was taken. That our locality is also that of Palmer
is supported by the occurrence of the newly described Euphorbia fruticulosa Boissier
var. hirtella M. C. Johnston (Wrightia 5:141) with the Selinocarpus, our collection
providing the holotype. M. C. Johnston (pers. comm.) subsequently discovered a
specimen of this variety in the Gray Herbarium collected by Palmer at “San Lorenzo
de Laguna”.

OPHIOGLOSSACEAE

Botrychium matricariifolium A. Br. New Mexico, Catron Co., Gila Wilderness, ca.
16 km by air ESE of Mogollon, on Crest Trail 182, 4 km SE of Sandy Point, in ma-
ture spruce-fir-aspen forest, N slope, 3033 m elev., R. Spellenberg, J. Reitzel, D. Hill
4528, 5 Sep 1976.

This record is the first for this species in the state and is the southernmost record
for the genus in New Mexico.

POACEAE

Muhlenbergia villosa Swallen. New Mexico, Otero Co., Otero Mesa NE of Oro-
grande, Sec 7 or 18, T24W, R11E, elev. 1775 m, R. Spellenberg 4565, 24 Sep 1976.

The species is said to be endemic to Texas and to occur there only extremely lo-
cally, apparently confined to gypsum, near the SW corner of the panhandle (F.
Gould. 1975. The Grasses of Texas. Texas A & M Press). The newly discovered popu-
lation was vigorous, on soil derived from limestone, occurring in scattered but dense
patches over an area of about 25 m*. The species is listed in the 6/16/76 Federal
Register for Texas as “endangered”.

Urochloa panicoides Beauv. New Mexico, Las Cruces, New Mexico State Univer-
sity campus, weedy lawn, R. Spellenberg 4480, 26 Aug 1976.

Mr. José Valls, a student of Dr. Frank Gould, kindly identified this collection. It
appears to be the first U. S. record for this Asian and African grass, now introduced
elsewhere in warm parts of the world. At this point it appears to be only adven-
tive. — RICHARD SPELLENBERG, Department of Biology, New Mexico State Univer-
sity, Las Cruces 88003.

ITRREGULARITY OF PINYON CONE PRODUCTION AND ITs RELATION TO PINYON CONE
Motu Prepation. — Eucosma bobana Kearfott (Lepidoptera: Tortricidae) larvae
are prey-specific predators of maturing ovulate cones of Pinus edulis Engelm. and
P. monophylla Torr. & Frem. (Powell, Hilgardia 39:1-36. 1968). Thus the number
of ovulate cones annually produced in pinyon woodlands may determine, in part,
the potential population densities of cone moths. In turn, the number of pine cone

1978] NOTES AND NEWS 171

moths produced in any one year may affect the production of viable pine seeds in
the next year.

Eight P. monophylla—Juniperus osteosperma (Torr.) Little woodlands in southern
Idaho and northern Nevada (U.S. A.) were visited during the winter and spring of
1977. Because ovulate cones leave abscission scars at the annual whorls of actively
growing leader shoots, I could estimate a cone production sequence for the 1968-
1977 period by the following method: sample five trees within a pinyon community
and count all potential cone-bearing leader shoots within those trees; then subsam-
ple five leader shoots from each tree and count the abscission scars and currently
maturing cones at the annual whorls of each leader shoot. By using the total pinyon
canopy coverage of a 600 m? plot that contained the five sample trees, these data
can be used as a basis for estimation of a 10-year cone production sequence on a
unit-area basis for the entire pinyon population (cf. Gorchakovskii, Bot. Zurn.
43:1445-1459. 1958). The sample numbers used in this method have been shown to
be adequate in a previous study (Forcella, Thesis — Montana State U., Bozeman
59715, p. 53. 1977; Weaver and Forceila, in prep.).

The accuracy of this method can be checked by comparing the abscission scar
estimate of the 1976 cone crop with the number of “freshly” abscised cones found
on the ground within the 600 m? plot (occasionally cones fail to abscise from the
leader shoots, but these can be easily counted in most cases since the trees are only
3-5 m in height). The two methods compare favorably (r2 = 0.95,n = 7, p = 0.01).

These freshly fallen cones can also be used to estimate the extent of predation
upon them. Cones infested with E. bobana larvae or pupa cases (normally one/cone
in the several tens of cones I checked) abort before maturity and possess telltale
entry-exit holes. Rodent-molested cones often exhibit chewing marks and are fre-
quently left in piles at the bases of trees directly below feeding platforms. Cones
ravaged by birds, notably pinyon jays (Gymnorhinus cyanocephalus) and Clark’s
nutcrackers (Nucifraga columbiana) have a generally shredded appearance.

In 1976, four of the sampled stands had cone crops well above their 10-year aver-
ages. Two of these stands also had average or above average cone crops in 1975,
while the other two stands had 1975 cone crops that were only about 50% of their
averages. The 1976 cones in the two stands with consecutively abundant crops were
heavily utilized (80% and 50%) by E. bobana larvae; an additional 10-20% were
preyed upon by other animals. In contrast, the abundant 1976 cones of the two
stands with below average 1975 crops were only mildly infested with cone moth
larvae (20% and 17%) and showed insignificant utilization by other sources. A sig-
nificant correlation exists between the current magnitude of cone moth depredation
and the abundance of the previous year’s cone crop (Fig. 1). A similar situation
occurs with Pinus rigida and its cone predators (Mattson, Canad. Entomol. 103:617-
621.1971).

Such evidence for prey-frequency dependent predation indicates that consecu-
tively good (above average) cone crops in pinyon are not selectively advantageous.
Instead for pinyon, fewer cones, but these produced at strategic intervals, may con-
fer fitness. The overall annual cone crops in pinyon communities are highly irregu-
lar (C. V. = 94 + 32% for the eight Great Basin stands). Above average cone
crops occur in only two or three out of ten years in Great Basin pinyons, and these
good crops appear to be randomly distributed within the 10-year sequence investi-
gated. In the same 10-year period P. monophylla communities south of N 136° 20’
(southern Calif., southern Nevada, and the Santa Ancho Mts., Arizona) had good
cone crops only once or twice, and their C. V.’s are correspondingly somewhat high-
er (112 + 24%, n = 7; unpublished data). Such irregularity in cone production
may have selective advantages by keeping the predator populations small through a
low average annual cone (mast) crop, thus insuring predator satiation and seed
survival in years of good cone crops.

E72 MADRONO [Vol. 25

R*= 0.86

y=0.5 + 0.4(x )

CO
©

% aborted

m
O

1976 Cone Crop
RO
©

20 100 150 200
1975 Cone Crop: % of mean

Fig. 1. The relative utilization of pinyon cone crops by Eucosma bobana (pine-
cone moth) larvae as a function of the magnitude of the previous year’s cone crop.
The y-axis represents the percent of the 1976 cone crop that was aborted due to
larval infestation. The x-axis represents the 1975 cone crop as a percent of the aver-
age of a 10-year cone production sequence for each Pinus monophylla community.
If all cone predators were included in the analysis the slope of the regression line
would be about 0.5.

K. T. Douglass, A. F. Johnson, P. G. Risser, and T. Weaver were all kind enough
to edit and comment on the original version of this report. — FRANK FORCELLA,
Department of Biology, Montana State University, Bozeman 59715 (Current ad-
dress: Department of Botany-Microbiology, University of Oklahoma, Norman
73019).

NOTES ON THE FLORA OF EAST-CENTRAL IDAHO. — Although excellent treatments of
much of Idaho’s flora have been published (Davis, Fl. Idaho, 1952; Hitchcock et al.,
Vasc. Plants Pacific Northw., 1955-1969, and Hitchcock and Cronquist, Fl. Pacific
Northw., 1973), some areas within the state remain relatively unknown floristically.
With funding made possible by a C. R. Stillinger Grant, intensive botanical explora-
tion of parts of east-central Idaho (Lost River, Lemhi, and Beaverhead Ranges of
Custer, Lemhi, Butte, and Clark Counties) was initiated in the summer of 1973. As
a result of three collecting seasons in the region, several plants worthy of mention
have been encountered. Collection numbers are those of the author; specimens are
deposited in ID.

1978] NOTES AND NEWS Ti)

Aquilegia coerulea James var. coerulea. Lemhi Co., E slope Lemhi Range, Targhee
Nat. Forest. Moist, N-facing slope beneath Douglas fir, Mammoth Canyon, 2136
m, 15 July 1973, 1108. Although the var. coerulea is common in the central Rocky
Mountain states, it is rarely encountered in Idaho, and this population, which is
composed of about half var. ochroleuca Hook., may represent the NW limit of this
blue-sepaled columbine. A second population in Butte Co., 7.5 km SE of Mammoth
Canyon in a southern tributary of Meadow Canyon is composed of both varieties of
A. coerulea, A. formosa Fisch., and A. flavescens Wats., as well as a wide array of
putative hybrids apparently combining features of all three taxa, 1601, 1602, 1603,
1604.

Caltha leptosepala DC. var. sulfurea Hitchc. Lemhi Co., E slope Lemhi Range,
Salmon Nat. Forest. Moist ground at head of small lake, headwaters of Middle Fork
Little Timber Creek, 2834 m, 3 July 1973, 932; subalpine meadow above Mill Lake,
2800 m, 30 July 1974, 2096. This variety was previously known only from the Lost
River Range, Custer Co., Idaho, but appears to be a common inhabitant of subalpine
stream banks and lake shores of the central and northern Lemhi Range where sub-
strates are primarily quartzitic.

Draba oreibata Macbr. & Pays. Lemhi Co., E slope Lemhi Range, Targhee Nat.
Forest. Alpine tundra on NE side of peak 10,652, 2.5 km NW of junction of Meadow
Canyon and South Fork, 3121 m, Sec 29, T11N, R28E, 24 July 1975, 2819; crest of
Lemhi Range, summit rocks of Big Windy Peak, 3170 m, at head of Spring Moun-
tain Canyon, 16 July 1973, 1116. This species was previously known in Idaho only
from Pass Creek Gorge, Lost River Range, Custer Co. Its occurrence in the Lemhi
Range is apparently restricted to limestone substrate.

Lesquerella carinata Rollins. Lemhi Co., Crest of Lemhi Range, summit rocks of
Big Windy Peak, 3170 m, head of Spring Mountain Canyon, 16 July 1973, 1118; E
slope Lemhi Range, NE ridge of peak 10,858 just S of junction of Meadow Canyon
and South Fork, 2438 m, 24 July 1975, 2864; Crest of Lemhi Range, 0.7 km N of
Bell Mountain, 3160 m, 29 July 1975, 2882. Butte Co., Crest of Lemhi Range, lime-
stone rocks on summit of peak 10,400 at head of Middle Fork Eightmile Canyon,
3170 m, 3 July 1974, 1442. Clark Co., Crest of Lemhi Range at the head of Surrett
Canyon, 2926 m, 8 July 1975, 2593; E base of Lemhi Range, outwash of Eightmile
Canyon, 1920 m, 4 June 1975, 2226; N side of Eightmile Canyon on limestone out-
crop, 2011 m, 6 June 1975, 2289; W slope of Beaverhead Range, limestone outcrop
in Peterson Canyon, 1804 m, 10 June 1975, 2357. Although apparently restricted to
limestone in east-central Idaho and previously considered rare, this species is com-
mon in Custer, Lemhi, Butte, and Clark Counties from lower elevations to the high-
est summits, and has been reported recently from Montana by Lackschewitz (Ma-
drono 23:361. 1976).

Saxifraga debilis Engelm. Lemhi Co., W slope Beaverhead Range, Salmon Nat.
Forest, moist rocks near summit of Mountain Peak, 3050 m, near the head of Bull
Creek, 22 July 1975, 2794, Although this species is listed by Davis (op. cit.), there
is no discussion concerning its occurrence within the state. Hitchcock and Cronquist
(op. cit.) state that it is probably in Idaho. It has been encountered but once in
three field seasons of intensive exploration in the region and can most probably be
considered rare for the state.

Trifolium haydenii Porter. Lemhi Co., E slope Lemhi Range, Targhee Nat. Forest,
limestone rocks at junction of Meadow Canyon with South Fork, 2316 m, Sec 33,
T11N, R28E, 7 July 1974, 1559; limestone rocks on summit of peak 10,652, N side
of Meadow Canyon, 3246 m, 24 July 1975, 2844; W slope Beaverhead Range, Sal-
mon Nat. Forest, quartzite talus on east side of Mountain Peak, 2926 m, 22 July
1975, 2761. Butte Co., E slope Lemhi Range, Targhee Nat. Forest, limestone rocks
near summit of peak 10,858, 1 km § of junction of Meadow Canyon and South
Fork, 3185 m, 10 July 1975, 2623. This plant is common in the Meadow Canyon
drainage of the southern Lemhi Range and occurs from canyon bottoms with limber

174 MADRONO [Vol. 25

pine and Douglas fir to the highest summits. It is particularly abundant on rocky
alpine slopes and, in some cases, is the most abundant understory plant with Engel-
mann spruce at timberline. This is believed to be the first report of this species for
Idaho.

Gentiana tenella Rottb. Custer Co., E slope Lost River Range, Challis Nat. For-
est, moist, grassy bank of Pass Lake near outlet, head of W Fork Pahsimeroi River,
3079 m, 15 August 1973, 1169. This species is apparently rare in Idaho for searches
of Pacific Northwestern herbaria (ID, IDS, WS, and WTU) and NY have failed to
produce a single Idaho specimen. Hitchcock and Cronquist (op. cit.) indicate that it
is reported from Idaho.

Langloisia punctata (Cov.) Goodd. Clark Co., Dry, rocky soil at Reno Point, S
end of Beaverhead Range, 1646 m, 9 July 1975, 2604. This is a rather substantial
population of several hundred plants and represents a significant N extension of its
range. It has not previously been reported from Idaho although its occurrence has
been noted in SW Idaho by Dr. Patricia Packard (pers. comm.).

Phacelia incana Brand. Clark Co., W slope of S Beaverhead Range Targhee Nat.
Forest, moist limestone talus, Bare Canyon, 2103 m, 22 June 1975, 2482; limestone
rocks at Reno Point, 1648 m, 9 July 1975, 2606. This species was previously known
in Idaho only from a few collections and is rare in the state. The collections cited
are populations composed of very few, widely scattered plants.

Phacelia lyallii (Gray) Rydb. Lemhi Co., E slope Lemhi Range, Salmon Nat. For-
est, talus slope on saddle connecting Mill Lake with E Fork Hayden Creek, 2877 m,
1 August 1974, 2735. Gillett (Rhodora 62:205—222. 1960) reported two locations for
this species in N Lemhi Co. Neither Davis (op. cit.) nor Hitchcock and Cronquist
(op. cit.) indicate its presence in Idaho.

Pedicularis contorta Benth. var. ctenophora (Rydb.) Nels. & Macbr. Lemhi Co.,
E slope Lemhi Range, Targhee Nat. Forest, grassy slope of peak 10,652. 3 km NW
of junction of Meadow Canyon and South Fork, 2746 m, Sec 21, T11N, R28E, 24
July 1795, 2859. The collection cited represents a W extension of the range of this
variety of only about 75 km, but it has not previously been reported for Idaho.
This plant is abundant in the Meadow Canyon area but has not been encountered
elsewhere in the region. — Doucrass M. HEeNpErson, Department of Biological Sci-
ences, University of Idaho, Moscow 83843.

REVIEWS

Terrestrial vegetation of California. Ed. by MicHart G. BarBour and JAcK Ma-
yor. + map of the natural vegetation of California (1:1,000,000) by A. W. Kiichler.
John Wiley and Sons, New York. 1977. $47.50.

California is one of the most diverse of states geologically, climatically, and bio-
logically. In what other state than California grow plants of more than 5,000 native
species? And in how many intricate ways do these plants become arranged into
vegetational] communities from tall forests to the almost barren bajadas of extreme
deserts? From chaparral to alpine fell fields? Who would dare to make order out of
such ecological chaos? Two people did, and to them we are indebted for this book.
Without the leadership of Barbour and Major, Terrestrial Vegetation of California
could not have been produced, even given the wealth of material and the cooperation
of knowledgable colleagues. . .

The wealth of plant species in the California Floristic Province, the endemics, the
intricate genetic relationships, and the combination of rock and climate led to the
dominance of taxonomy in California at mid-century and to the rise of biosyste-
matics. Because of this natural pre-occupation with flora and its evolution, it is not
too surprising that research on vegetational structure and classification in California
was somewhat neglected, especially by academic scientists. This book could not
have been written even as recently as twenty years ago. It comes none too early

1978] REVIEWS 1/5

in a region where even its ecologically enlightened people have overwhelmed many
of the native biological communities.

The book is packed with information: 26 chapters, more than 1000 pages, and
the large useful map. The introductory overview alone consists of six chapters and
220 pages, a book in itself. Do you want to know something about the California
climate ? Read Jack Major’s Chapter 2. How about the flora? Peter Raven’s Chapter
4. And what about California’s paleoecology? Dan Axelrod’s Chapter 5 comes as
close in short space to summarizing the great changes that have taken place in west-
ern North America during the Tertiary and Pleistocene as could be done without
writing a separate book. I notice only one omission in the overview section that
would be helpful in such a relatively dry region — a chapter on the hard rock ge-
ology and perhaps another on the building and shaping of the mountains. Vegeta-
tion is almost as closely allied to substratum in much of California as it is to climate.
Fortunately, there are very good books on California geology such as Jahn’s “The
Geology of Southern California” and Bailey’s “Geology of Northern California”’.

There are 20 chapters on the main vegetational “types” of California. These are
arranged into six large floristic provinces based upon their geographic affinities. Each
chapter is an authoritative compilation. There is a uniform approach in all chapters,
and there is cross-reference and integration among the chapters. Fortunately, no
attempt was made to impose a hierarchical classification. Each chapter has one or
more tables of actual vegetational data. Even though these are sometimes small sam-
ples, they provide some feeling for the vegetational and floristic structure.

There are several other things about each chapter that I like. For example, the
title page of every chapter that deals with a vegetational type displays a physio-
graphic map of California. On this map, the areas of the particular vegetation type
are clearly marked. Details can be checked in Kuchler’s excellent map. Each chapter
includes a sampling of data (of varying richness) from physiological ecology. This
is unusual in books on vegetation. Yet, if we are to understand vegetational compo-
sition, pattern, and function, we must have such information. This leads logically to
another uniformity among chapters: suggestions for future research. What a gold
mine of ideas for graduate students and active ecologists. All this, and the literature
reviewed, too!

Further weak points are relatively minor. In such a book, one expects an abun-
dance of clear photographs of the vegetation. There are relatively few, and the re-
production of some lacks clarity. Another money-saving device is the use of type-
writer script for camera-ready tables and reference lists, which I don’t begrudge.

Here, then, is a true classic that everyone interested in the Californian country-
side, mountains, and deserts should have. The price? Entirely reasonable when it is
realized that it is an encyclopedia of western vegetational ecology, and the book has
scientific permanence. Its like will not appear again within the lifetime of its present
readers. — W. D. Bitirncs, Department of Botany, Duke University, Durham,
North Carolina 27706.

Ed. Note: No more than one review of a single volume is normally published in
Madrono. However, Terrestrial vegetation of California is sufficiently significant as
a landmark that more than one perspective is useful.

As members of (we suspect) the first upper division class to utilize Terrestrial
vegetation of California as a text, we wish to communicate our reactions to the book.
After a ten-day field trip through northern California, the class read and discussed
14 of the 20 chapters dealing with vegetation. As informed students we are meters
of evaluation somewhere between the initiate and the professional botanist.

Th editors have handled burdensome layout problems well. However, there are a
few prominent difficulties that must be mentioned. Most of the tables have been
presented lengthwise so that they are not easily examined, and the spacing is so ex-

176 MADRONO [Vol. 25

panded that unnecessary pages are added to an already large volume.

A. W. Kiuchler’s “Native Vegetation of California’ map is enclosed with an ex-
panded legend in an appendix. In addition, each chapter is preceded by a diagram-
matic map depicting the locations of vegetation types discussed. The latter could
have been larger and more clearly contrasted for easy reference. Photographs tend
to be dark, lacking in contrast, and of little assistance. Vegetation types are not con-
sistently named throughout the book. Names should have been standardized or al-
ternatives listed at the beginning of each chapter.

The “Research Natural Areas” and “Status of Vegetation Mapping” chapters are
peripheral to the purpose of the book. The chapter on climate is essential, but
lengthy and not easily read. It could have been a shorter review, as stated in the
book’s objectives.

Raven’s discussion of California’s flora in the introductory chapters is excellent,
offering a good introduction to the highly endemic nature of California taxa and
providing a firm basis for understanding the processes involved. Zinke’s chapter,
“Redwoods and Associated North Coastal Forests’, lacks organization. He incom-
pletely synthesizes a large amount of data, thus failing to guide the reader in specula-
tion about redwood ecology.

Griffin’s chapter, “Oak Woodland”, is the best organized and most readable of all
the contributions. He takes a ubiquitous vegetation type with little or no floristic
unity and molds it into a presentation that at once gives a clear explanation of what
is known (very little) and what is not known (a great deal).

“The Closed Cone Pines and Cypresses” displays the numerous problems in deal-
ing with a difficult assemblage of arbitrary communities. Although the bibliography
is superb, the chapter lacks cohesiveness and reflects a southern California viewpoint.
The same bias is seen in the chaparral chapter.

Chapters on “Southern Coastal Scrub” by Mooney and “Vernal Pools” by Holland
and Jain outshine others in the presentation of autecological studies. It is regrettable
that knowledge of other vegetation types such as “North Coastal Scrub and Prairie”’
is not up to this level. The inconsistency of this last chapter may well be a reflection
of its multiple authorship.

The “Montane and Subalpine Vegetation of the Sierra Nevada and Cascades”
chapter presents a fine, although occasionally flawed, summary. In spite of contro-
versy in the literature, only one interpretation of tree stand dynamics is offered.

The most important function of the volume will be as a reference rather than as
a text — a kind of manual of vegetation types. Although cumbersome in the field or
car, as most floras are, this book will be a valuable part of an herbarium, library, or
botanist’s reference shelf. A valuable feature is the extensive and current bibliogra-
phy that follows each chapter.

This book is easily the best available review of California’s vegetation, assembling
diverse opinions and backgrounds from California’s premier plant ecologists into a
sizeable yet usable compendium that stimulates the reader not only to read and di-
gest but to challenge with personal observation and research. — PETER STEKEL and
Epwarp A. Copr, Department of Biology, Humboldt State University, Arcata, CA
55712

MAP OFFER

The new. map of the natural vegetation of California in full color at the scale of
1:1,000,000 will become available on 15 March 1978 with a brochure containing a
commentary on the map and a detailed description of every vegetation type. Price:
$8 incl. brochure and postage; orders of 20 or more at a discount of 15%. Rolled
maps, shipped in a tube cost $12, postage extra. Make checks payable to A. W.
Kiichler and send orders with checks to A. W. Kichler, Dept. of Geography, Uni-
versity of Kansas, Lawrence, KS 66045.

Membership in the California Botanical Society is open to individuals ($12.00 per
year, regular; $8.00 per year, student). Members of the Society receive MapRroNo
free. Institutional subscriptions to MaproNo are available ($20.00 per year).

Back issues of Madronio are available at the following rates (some issues are
out of print):

Vol. 1 (1916-1929) and Vol. 2 (1930-1934, each consisting of 17 numbers: $1.50
per issue and $25.50 per volume for members; $3.00 per issue and $51.00 per volume
for institutions.

Vol. 3 (1935-1936) through Vol. 23 (1975-1976), each biennial, consisting of 8
numbers: $3.00 per issue and $24.00 per volume for members; $5.00 per issue and
$40.00 per volume for institutions.

Vol. 24 (1977) et seq., one volume per year, each consisting of 4 numbers: $3.50
per issue and $14.00 per volume for members; $5.00 per issue and $24.00 per volume
for institutions.

Applications for membership (including dues), orders for subscriptions, requests
for back issues, changes of address, and undelivered copies of MApRONO should be
sent to the California Botanical Society, Inc., Department of Botany, University of
California, Berkeley 94720.

INFORMATION FOR CONTRIBUTORS

Manuscripts submitted for publication in MaproNo should be sent to the Editor.
Membership in the California Botanical Society is prerequisite to review.

Manuscripts and review copies of illustrations must be submitted in triplicate for
articles and in duplicate for short items intended for NOTES AND NEWS. Follow
the format used in recent issues for the type of item submitted and allow ample
margins all around. All manuscripts MUST BE DOUBLE SPACED THROUGH-
OUT. For articles this includes title (all caps, centered), author names (all caps,
centered), addresses (caps and lower case, centered), Abstract, text, Acknowledg-
ments, Literature Cited, Tables (caption on same page), and Figure Captions
(grouped as consecutive paragraphs on one page). Order parts in the sequence listed
ending with Figures, and number each page. Do not use a separate cover page,
“erasable” paper, or footnotes. All measurements should be in S.J. (metric) units.

Line copy illustrations should be clean and legible, proportioned (including their
captions) to the MAprRoNo page, and designed for reduction to 2/3 original size.
Scales should be included in Figures, as should explanation of symbols. In addition,
maps must include latitude and longitude references. Halftone copy should be de-
signed for reproduction at actual size. In no case should original illustrations be sent
prior to the acceptance of a manuscript. When needed, they should be mounted on
stiff board and mailed flat.

Presentation of nomenclatural matter (accepted names, synonyms, typification)
should follow the format used for Rhus integrifolia in MADRONO 22:288. 1974. Insti-
tutional abbreviations in specimen citations should follow Holmgren and Keuken’s
list (Index herbariorum, Part 1. The herbaria of the world. Sixth edition. 1974.
Regnum Veg. vol. 92). Abreviations of names of journals should be those in Botantz-
co-Periodicum-Huntianum (Lawrence, G. H. M et al. 1968. Hunt Botanical Library,
Pittsburgh). If the correct abbreviation cannot be determined, the full title of the
journal should be used. Titles of books should be given in full, together with the
place and date of publication, name of publisher, and edition, if other than the first.

All members of the California Botanical Society are allotted twelve pages in the
journal over a two-year period. Beyond that number of pages, a required editorial
fee of $40.00 per page will be assessed. The purpose of this fee is not to pay directly
for the costs of publishing any particular paper, but rather to allow the Society to
continue publishing MaproNo on a reasonable schedule, with equity among all mem-
bers for access to its pages. Subject to approval of the Editor, articles may be pub-
lished ahead of schedule, as additional pages of an issue. Editorial fees for such
papers will be larger and will be determined on an individual basis.

ANNOUNCEMENT
CALIFORNIA BOTANICAL SOCIETY GRADUATE STUDENT MEETINGS

The California Botanical Society Graduate Student Meetings will be held at Cali-
tornia Polytechnic State University, San Luis Obispo, on November 11 and 12, 1978.
The meetings will take the form of a series of short research papers, or reports on
work in progress, contributed by graduate students in all botanical fields. Members
and non-members of the society are invited to participate. If you wish to receive
further announcements concerning these meetings, please send your name and ad-
dress to Malcolm McLeod, Biological Science Department, California Polytechnic
State University, San Luis Obispo, CA 93407.

MADRONO

VOLUME 25, NUMBER 4 OCTOBER 1978
Contents
THE VEGETATION OF Two CALirorNIA MounNTAIN SLOPES,

John T. Gray 177
BIOSYSTEMATICS OF PSILOSTROPHE (COMPOSITAE: HELENIEAE).

II. ARTIFICIAL HYBRIDIZATION AND SYSTEMATIC TREATMENT,

Roy Curtiss Brown 187
GERMINATION OF COMANDRA (SANTALACEAE), Job Kuijt 202
CHROMOSOME NUMBERS IN XYLORHIZA NUTTALL (ASTERACEAE —

ASTEREAE) , Thomas J. Watson, Jr. 205
INTERSPECIFIC HYBRIDIZATION BETWEEN NATIVE AND NATURALIZED

CRATAEGUS (ROSACEAE) IN WESTERN OREGON, Rhoda Love and

Marc Feigan 211
A NEw SPECIES OF EUPATORIUM (ASTERACEAE) FROM CALIFORNIA,

Dean Wm. Taylor and G. Ledyard Stebbins 218
A NEw SPECIES OF VIGUIERA (ASTERACEAE — HELIANTHEAE)

FROM Nayarit, Mexico, B. L. Turner 221
A NEw SUBSPECIES OF ABRONIA MARITIMA FROM BAJA CALIFORNIA,

Ann F. Johnson 224
RANUNCULUS GERANIOIDES H.B.K. Ex DC. in Costa RICA AND

PanaMa, Thomas Duncan 228

NOTEWORTHY COLLECTIONS

NEw FoRMAT 231
TEESDALIA CORONOPIFOLIA, Charles F. Quibell and

John L. Strother 231
AGROSTIS HUMILIS, Kurt R. Neisess 232
ASPLENIUM SEPTENTRIONALE, David W. Showers 232

NOTES AND NEWS
LASTHENIA CALIFORNICA (COMPOSITAE), ANOTHER NAME FOR A

CoMMON GOLDFIELD, Dale E. Johnson and Robert Ornduff 227
ENDANGERED SPECIES IN CALIFORNIA: FEDERAL PROCEDURES

AND Status Report, Alice Q. Howard 232
A CORRECTION ON THE INDIGENOUS DISTRIBUTION OF

KwosconeE Prine, Jon E. Keeley 233
GYNODIOECY IN MAMMILLARIA DIOICA (CACTACEAE),

Fred R. Ganders and Helen Kennedy 234
ADDITIONS TO THE FLORA OF THE FARALLON ISLANDS,

CaALiFornNIA, Malcolm C. Coulter 234

A WEST AMERICAN JOURNAL OF BOTANY

Continued on back cover

IBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY

MaproONo is published quarterly by the California Botanical Society, Inc., and is
issued from the office of the Society, Herbarium, Life Sciences Building, University
of California, Berkeley. Established 1916. Second-class postage paid at Berkeley.
Return requested.

Editor — James C. HicKMAN
Department of Botany
University of California, Berkeley 94720
(415) 642-1079

Board of Editors
Class of:

1978—SHERWIN CarRLQvuIsT, Claremont Graduate School
LestiE D. Gottiies, University of California, Davis
Dennis R. PARNELL, California State University, Hayward
1979—Puiripe W. RunNDEL, University of California, Irvine
ISABELLE TAVARES, University of California, Berkeley
1980—JameEs R. GrirFin, University of California, Hastings Reservation
Frank A. Lane, Southern Oregon College, Ashland
1981—DanIeEL J. CRAWForD, Ohio State University, Columbus
James Henrickson, California State University, Los Angeles
1982—Dean W. Taytor, University of California, Davis
RICHARD VocL, California State University, Los Angeles

CALIFORNIA BOTANICAL SOCIETY, INC.

OFFICERS FOR 1978

President: Dr. MicHAEL F. Baap, Department of Biological Sciences,
California State University, Sacramento 95819

First Vice President: Dr. ALAN R. POLANSHEK, Department of Biological Sciences,
San Jose State University, San Jose, CA 95114

Second Vice President: Dr. AMY GILMARTIN, Department of Botany,
Washington State University, Pullman 99164

Recording Secretary: Dr. CHARLES F. QuUIBELL, Department of Biological Sciences,
Sonoma State College, Rohnert Park, CA 94928

Corresponding Secretary: Dr. RUDOLF ScHMID, Department of Botany,
University of California, Berkeley 94720

Treasurer: Dr. JOHN H. THomas, Department of Biological Sciences,
Stanford University, Stanford, CA 94305

The Council of the California Botanical Society consists of the officers listed above
plus the immediate Past-President, WINSLOW R. Briccs, Carnegie Institute of Wash-
ington, Stanford, CA 94305; the Editors of Madrofio; three elected Council mem-
bers: JAMES R. GriFFIn, Hastings Reservation, Star Route 80, Carmel Valley, CA
93924 (1976-1978) ; Harry D. TuHiers, Department of Ecology and Systematic Bi-
ology, California State University, San Francisco, CA 94132 (1977-1979) ; ALAN R.
SmitH, The Herbarium, Department of Botany, University of California, Berkeley,
CA 94720 (1978-1980) ; and a Graduaie Student Representative, NANCY R. Morin,
Department of Botany, University of California, Berkeley, CA 94720.

THE VEGETATION OF
TWO CALIFORNIA MOUNTAIN SLOPES

JoHN T. GRAY
Department of Biological Sciences,
University of California,
Santa Barbara 93106

ABSTRACT

Community characteristics and the distribution of woody species along complex
elevational gradients are described for south-facing slopes in two California moun-
tain ranges. In the inner North Coast Range, the low elevation (780-1250 m) com-
munity on Snow Mountain is a chaparral-woodland on serpentine substrate. At
1250 m, a sharp ecotone marks the change to nonserpentine soils and a mixed-coni-
ferous forest that extends to the peak (2112 m). Along a northern Sierran transect
(600-2040 m) is a series of three, well-defined forest communities. Species are dis-
tributed along this elevational gradient in a broad, overlapping manner with less
correlation between community type and edaphic factors than found on Snow
Mountain. The number of tree species in the Sierra is greater than on Snow Moun-
tain, where serpentine soils and a drier climate are present. The number of shrub
species and their relative importance is greatest in the serpentine chaparral where
tree density and cover are low. On both slopes, overall woody species diversity de-
creases at higher elevations.

Montane forests are distributed in predictable patterns that vary with
elevation. In California, the forests of the Sierra Nevada form a series of
elevational zones dominated by one or several tree species. The distribu-
tion and composition of these forests have been characterized by Rundel,
Parsons, and Gordon (1977), Ornduff (1974), Munz (1959), and
Kylver (1931). Within an elevational zone, each montane community
is composed of a mosaic of site types, based in part on microclimatic
and topographic positions. Diversity of soil and parent material also influ-
ences the composition and distribution of montane communities as shown
by Kruckeberg (1969) and Whittaker (1960).

The objective of this study is to document quantitatively the composi-
tion and structure of montane forest communities along two California
mountain slopes. The study is both a synecological description of several
common forest communities and an example of how ordination tech-
niques can be used to examine community characteristics and variation
in response to environmental factors.

Description of Study Areas. A steep slope was located on Snow Moun-
tain (Lake Co.) in the inner North Coast Range (N39°22’, W122°45’).
The transect extended from Bear Creek Ranger Station to the West Peak.
The elevational gradient was from 780 m to 2040 m, over a linear dis-
tance of 9.6 km. A more gradual slope was located 135 km to the east in

Maprono, Vol. 25, No. 4, pp. 177-240, January 30, 1979
177

178 MADRONO [Vol. 25

the Sierra Nevada (N39°25’, W121°05’ to N39°38’, W120°30’). The
transect began 4 km south of Camptonville (Yuba Co.) and followed
Highway 49 to Yuba Pass (Sierra Co.). The elevational gradient was
from 600 m to 2040 m over a distance of 73 km.

Mean annual precipitation on the Sierran transect is 127 cm, 180 cm,
and 102 cm at low, medium, and high elevations, respectively (U. S.
Weather Bureau, 1964). Climatic records for the Snow Mountain area
from low elevation stations (400 m) east of the mountain show mean an-
nual precipitation of 50-76 cm, and mean annual temperature of 15.7°C
(U.S. Weather Bureau, 1964). Mean annual temperature decreases with
elevation along the Sierran transect from 11.5° to 6.0°C (U.S. Weather
Bureau, 1964). Winter snow is found above 1400 m on the Sierran tran-
sect, and frequently above 1700 m on Snow Mountain.

Soils at low to medium elevations on Snow Mountain are predominantly
red-brown, slightly acidic stony-loams (U. S. Forest Service, 1955).
Above 1800 m, there are large colluvial areas of broken rock with little
soil. Intermixed are areas of shallow, moderately acidic loams (U. S. For-
est Service, 1955). The geology of Snow Mountain and most of the North
Coast Range is dominated by rocks of the Franciscan formation (Page,
1966; Irwin, 1960). At low elevation sites there is a disorderly assem-
blage of sandstone, shale, and chert, with intrusions of serpentine. Above
1250 m is a more homogeneous formation of volcanic and metavolcanic
greenstones with no reported or observed areas of serpentine (Dept. Nat.
Res., 1966). ,

Below 1500 m on the Sierran transect the predominant soils are red-
dish-brown, slightly acidic loams (Soil. Con. Ser., pers. comm.). No data
are presently available for soil types at higher elevations. The parent ma-
terial below 1150 m is marine sedimentary mixtures of clay and shale.
Mid-elevation sites are underlain by glacial deposits consisting of fluvio-
glacial sand and gravel. Granitic rock predominates at higher elevations
(Dept. Nat. Res., 1962).

METHODS

In the spring of 1977, vegetation was sampled at 90 m elevation inter-
vals along both slopes. Only south to southwest slopes were sampled with
inclinations of 5—30°. At each elevation interval a preliminary recon-
naissance was conducted to record the presence of all woody species. A
site was then chosen that had the desired aspect and that contained the
representative species commonly encountered at that elevation. Sites were
homogeneous in the sense that they were undisturbed and not topograph-
ically diverse.

A 20 by 25 m quadrat (0.05 ha) was marked and numbers of all woody
species, including tree seedlings, were recorded. Basal area of stems with
1.0 cm dbh or greater was also recorded for trees, shrubs, and seedlings.
Percent cover of all species was estimated along three 25,m line intercepts.

1978 | GRAY: MOUNTAIN SLOPES 179

Two indices of relative dominance were used in the data analysis. The
relative basal area of stems was calculated for most species at each site.
For all species at each site an importance value (IV) was calculated by
the formula: IV = 0.5 (RD + RC); where RD = relative density and
RC = relative cover. This synthetic index was chosen because it ex-
presses the different types of dominance that are found in the tree and
shrub strata, as well as lessening the over-weighting effect of highly dom-
inant species. Trees tend to dominate in terms of relative cover, even at
low densities; shrubs may show higher densities, but lower cover.

A matrix of pair-wise, unstandardized sample similarities was calcu-
lated by the Bray and Curtis (1957) formula for percent similarity
(PS): PS = = min (x, y); where min (x, y) is the smaller of the two
IV values for a given species in sample sites X and Y. The matrices of
site similarities for both transects were displayed by the polar ordination
method of Bray and Curtis (1957) on two orthogonal axes. The two most
dissimilar sites were chosen as two X axis endpoints. The site with the
lowest mean similarity with all other sites was selected first, followed by
the site least similar to it. The two Y axis endpoints were chosen accord-
ing to the criteria suggested by Cottam, Goff, and Whittaker (1973).

RESULTS

Fig. 1 shows the relative basal area of dominant tree species at each
site along the elevational gradients. In general, the important trees on
Snow Mountain exhibit greater mean dominance as measured by rela-
tive basal area than on the Sierran transect.

The distribution of individual species and their relative importance is
indicated in Tables 1 and 2. On the Sierran transect Pinus ponderosa,
Quercus chrysolepis, Pseudotsuga menziesu, and Calocedrus decurrens
have wide-ranging distributions. In contrast, there are few species on
Snow Mountain with equally broad ranges, resulting in a greater change
of species.

Fig. 2 shows the polar ordination of sample sites. Sites that are located
close together in the ordination field are similar to one another in species
composition. These clusters are considered one community type and
named after the dominant species or vegetation type.

The demarcation of communities is especially apparent on the Sierran
transect where three distinct forest belts are found. A yellow-pine belt
extends from 600-1500 m with Pinus ponderosa dominating, and Pseu-
dotsuga menziesii, Calocedrus decurrens, and Quercus kelloggii exhibit-
ing lesser importance. A white-fir forest with Abzes concolor and Pinus
jeffreyi follows. Above 1860 m is the red-fir community consisting of
mainly Abies magnifica with Pinus contorta ssp. murrayana in small
patches. These elevational belts correspond with the montane communi-
ties described by Rundel et al. (1977) and Munz (1959). The Sierran
transect, though of a greater elevational range, does not include a low-

180 MADRONO [ Vol. 25

SNOW MOUNTAIN .
100 Ee ee

% BASAL AREA PER SITE
4)
oO

780 1050 [500 2040
ELEVATION (m)

SIERRAN TRANSECT
lOO

% BASAL AREA PER SITE

600 1050 1500 2040
ELEVATION (m)

Fic. 1. Relative basal areas of dominant tree species along both elevational gradi-
ents. Only species with 25% or greater relative basal area per site are included.
P.s. = Pinus sabiniana; P.m. = Pseudotsuga menziesii; C.d. = Calocedrus decur-
rens; Pl. = Pinus lambertiana; A.c. = Abies concolor; P.j. = Pinus jeffreyii;
A.m. = Abies magnifica.

elevation foothill community. This vegetation is present below 400-500
m and is characterized by Pinus sabiniana, Quercus douglasi, and Arcto-
staphylos ssp.

Two main community types on Snow Mountain are indicated by the
ordination of sites in Figure 2. Below 1250 m is a chaparral-woodland
with Adenostoma fasciculatum, Pinus sabiniana, and Quercus durata
sharing importance. Above is a montane coniferous forest showing suc-
cessive dominance of Pinus ponderosa, Pinus lambertiana, and Abies con-
color with increasing elevation. The chaparral community is common at
low elevations in the North Coast Range where serpentine is abundant
(Walker, 1954). The montane coniferous forest may be further separated
into a yellow-pine belt (sites 7-10) and a white-fir belt (sites 11-14).
However, sites 7-14 are considered one community type in this study
because of their tendency to cluster together as closely related sites in the
polar ordinations. This similarity is primarily due to the presence of

1978] GRAY: MOUNTAIN SLOPES

181

TABLE 1. IMPORTANCE VALUES (IV) FOR SPECIES AT EACH SITE ON SNOW MOovunNTAIN.

; Sites
Species 12

. Quercus dumosa 2

. Arctostaphylos stanfordiana 4

. Ceanothus cuneatus 1 19 a |

. Umbellularia californica a

. Adenostoma fasciculatum A627 52 62

. Pinus sabiniana 22 18 17 18

. Arctostaphylos viscida 3

. Quercus durata 25,24 +7

. Quercus wislizenii oe ah

10. Pseudotsuga menziesii

11. Ceanothus integerrimus Z

12. Cercocarpus betuloides 22709

13. Arctostaphylos glandulosa

14. Pinus ponderosa 7

15. Quercus chrysolepis 13

16. Arctostaphylos canescens 2

17. Pinus lambertiana

18. Abies concolor

19. Ceanothus cordulatus

20. Abies magnifica

21. Ribes lobbii

O OND NB WHY

3 4 5 6 7 8 9 10 11 12 13.14 15

TABLE 2. IMPORTANCE VALUES (IV) FOR SPECIES AT EACH SITE ON THE SIERRAN

TRANSECT.
Sites

Species i223 4 5 6 57° 8. 910 TTA? 13, 141 16-17
1. Lithocarpus densiflora fis
2. Arbutus menziesii 74 5 4
3. Arctostaphylos viscida 3111 2 1 3
4. Quercus kelloggii 11 20 3125 9 7 13 6 3
5. Ceanothus integerrimus 8
6. Pinus ponderosa 33 31 35.24 28 17-34 23 50 40 14 12
7. Quercus chrysolepis 4-31 2) 1 (84711 204: 6.2073
8. Pseudotsuga menziesii 11 3 13 25 35 48 17 14 11 13 20 8
9. Ribes nevadensis 1
10. Calocedrus decurrens 3 5. 5 of 18°25.37 23:24 22.-4::3
11. Pinus lambertiana 2 © 4 4 2
12. Amelanchier alnifolia 4 4
13. Arctostaphylos patula 4 5
14. Ribes roezlii 4S 3! 22
15. Garrya fremontii 8 2
16. Castanopsis sempervirens 12
17. Abies concolor 3 8 14 41 56 50 58 14 12
18. Ceanothus cordulatus 13 14 22
19. Pinus jeffreyi 2 Gr S13
20. Pinus contorta ssp. murrayana 2. 4Ve3) 16° 6 19
21. Abies magnifica 5, 8 72.56

182 MADRONO [Vol. 25

SNOW MOUNTAIN SIERRAN TRANSECT
100 | &@e@° 100 7 a?
i @° : He, 14 a! en
Montane coniferous I White Fir / a
3 # “7
+ Ve? pee Uc
. @' ae ace
\@’ e@' Cg’
Mo eee Red Fir \ g'é
ae
Y 50
| & ~
2 &, , Chaparral - woodland
& x (serpentine)
\

0 50 100

Fic. 2. Polar ordinations of sites along both elevational gradients. Sample sites are
numbered and the underlying parent material indicated. Site 15 on Snow Mountain
is not included. SNOW MOUNTAIN: triangles = Franciscan formation of sand-
stone and shale, serpentine present; dots = volcanic greenstones of the Franciscan
formation, no serpentine. SIERRAN TRANSECT: triangles = marine sedimentary
rock; dots = glacial deposits; squares = granitic rock.

TABLE 3. SPECIES DIVERSITY AND SHRUB IMPORTANCE VALUES FOR COMMUNITIES
oN BotH ELEVATIONAL GRADIENTS. Values calculated for each community type as
indicated in Fig. 2. Nt = mean number of tree species/site; Ns; = mean number of
shrub species/site; IV; = mean percent IV for all shrub species/site; H’ = Shannon-
Weaver index of diversity.

Community Nt Ns IVs 1
SNOW MOUNTAIN ;

Chaparral-Woodland 2.16 5.83 7355 1.60
Montane-Coniferous 2.39 2.25 47.03 1.40
SIERRAN TRANSECT

Yellow Pine 4.81 2.09 16.17 1.59
White Fir 4/9 215 Zoran 1.49
Red Fir 4.00 0.0 0.0 1.01

Ouercus chrysolepis and Pinus lambertiana in sites 7-14. Quercus chryso-
lepis is found at high densities in the understory of these sites as a small
tree and shrub. The red-fir forest on the peak of Snow Mountain repre-
sents the southwest edge of a more extensive forest community that is
found on the upper north and east slopes.

Table 3 lists the mean numbers of tree and shrub species, and total
woody species diversity for each community. The mean percent of total
importance values contributed by shrubs for each community is given as
an index of relative shrub importance. Woody species diversity decreases
with increasing elevation on both slopes, as does the importance of shrubs
in the vegetation. Tree species richness is greatest in the Sierran tran-
sect; however, the number of shrub species in the serpentine chaparral is
more than twice the number in any Sierran community.

1978] GRAY: MOUNTAIN SLOPES 183

DISCUSSION AND CONCLUSION

The elevational transects in this study are complex climatic gradients
along which temperature, precipitation, and length of growing season
change. Species populations and community types are distributed along
this environmental gradient in response to climatic factors (Whittaker,
1967). Edaphic conditions also change on both slopes and represent addi-
tional environmental factors that may account for variation in species
and community types.

The ecotone between the chaparral-woodland and the montane coni-
ferous forest on Snow Mountain is sharp as indicated by both the ordina-
tion and the individual distribution of the dominants. Quercus durata is
a serpentine indicator (Kruckeberg, 1954, 1969) and is present on all
but one of the chaparral-woodland sites. The discontinuity between this
form of “serpentine chaparral” (Hanes, 1977) and the forest community
at higher elevations corresponds to the edaphic disjunction.

No such disjunction based on edaphic factors is apparent in the Sier-
ran transect. The dominants are distributed in an overlapping manner
and there is less correlation between the ordination of communities and
the underlying parent material. It appears that the distribution of com-
munities along the Sierran transect is primarily a result of species dis-
tributions along a complex climatic gradient.

Although ordination methods are being used with increasing frequency
in the analysis of vegetation, their effectiveness can be limited with cer-
tain types of data. Ordinations are subject to distortions of sample posi-
tion when beta-diversity, the range of community variation, is great.
Polar ordination has been shown to be the most resistant to this prob-
lem in comparison with other techniques (Gauch et al., 1977). There is
no evidence of distortion from curvilinearity in the Snow Mountain ordi-
nation where beta-diversity is high; only a small amount exists in the
Sierran ordination. Indeed, for both ordinations, the X axis or a diagonal
axis parallels the main direction of change in community composition in
response to elevation.

A major limitation of polar ordination is its vulnerability to the effects
of different endpoint choices. When an extreme endpoint is selected with
zero similarity with most other samples, the ordination can often become
ineffective. If less extreme endpoints are chosen, the more extreme sam-
ples will always be found in the center of the ordination field, obscuring
ecological interpretations of sample positions (Cottam et al., 1973). Be-
cause of this problem, site 15 on Snow Mountain was not included in the
final ordinations presented in this paper. Thus, the interpretation of the
Snow Mountain ordination is limited by the incomplete range of commu-
nity variation presented, and a possible over-estimation of the dissimilar-
ity between the two community types.

The following conclusions can be made about the composition and
structure of the vegetation along the two mountain slopes:

184 MADRONO [Vol. 25

(1) The observed pattern of species and community distribution can
be related to an environmental gradient of climatic and edaphic factors.
When environmental factors change abruptly, a sharp ecotone in vegeta-
tion may occur.

(2) Diversity of shrub and tree species decreases with increasing ele-
vation on both slopes, which may result from the harshness of the high
elevation environment.

(3) The numbers of tree species are less in the Snow Mountain com-
munities than in the Sierra, primarily due to the presence of serpentine
soils and a drier climate.

(4) The number of shrub species is greatest in the chaparral-woodland.
Shrubs exhibit relatively high density forming a low, open vegetation
with scattered pine trees. The increase in shrub species richness may be
due in part to the low tree density and cover. This reduction in commu-
nity structure can be attributed to the impoverished serpentine soils that
limit biomass, productivity, and vertical differentiation of the vegetation
(Whittaker, 1954).

ACKNOWLEDGMENTS

I thank P. J. Richerson for his help and encouragement during the study and
S. Schleifer for her aid in field work. I also thank W. H. Schlesinger and J. R. Haller
for reviews of the manuscript and M. Hasey for the illustrations.

LITERATURE CITED

BATEMAN, P. C., and C. WaHRHAFTIG. 1966. Geology of the Sierra Nevada, p. 107-
172. In E. H. Bailey (ed.) Geology of Northern California. Bull. 190, Calif. Div.
Mines, San Francisco.

Bray, J. R., and J. T. Curtis. 1957. An ordination of the upland forest communities
of southern Wisconsin. Ecol. Monogr. 27:325-349.

Cottam, G., G. F. Gorr, and R. H. WuitraAKer. 1973. Wisconsin comparative ordi-
nation, p. 193-222. In R. H. Whittaker (ed.) Ordination and classification of
communities. Junk, The Hague.

DEPARTMENT OF NaTuRAL REsouRCES. 1960. Geologic map of California. 1:250,000.
Ukiah Sheet. O. P. Jenkins Edition. Calif. Div. Mines. San Francisco.

. 1962. Geologic map of California. 1:250,000. Chico Sheet. O. P. Jenkins
Edition. Calif. Div. Mines. San Francisco.

Gaucu, H. G., R. H. Wuirraker, and T. R. WENTWorTH. 1977. A comparative
study of reciprocal averaging and other ordination techniques. J. Ecol. 65:157-
174.

Hanes, T. L. 1977. California chaparral, p. 419-469. Jn Barbour, M. G. and J. Major
(eds.) Terrestrial vegetation of California. Wiley and Sons, New York.

Irwin, W. P. 1960. Geologic reconnaissance of the northern Coast Ranges and Klam-
ath Mountains, California, with a summary of mineral resources. Calif. Div.
Mines. Bull. 19. San Francisco.

KRUCKEBERG, A. R. 1954. The ecology of serpentine soils. III. Plant species in rela-
tion to serpentine soils. Ecology 35:267-274.

. 1969. Soil diversity and the distribution of plants, with examples from
western North America. Madrono 20:129-154.

1978] REVIEW 185

Kytver, F. D. 1931. Major plant communities in a transect of the Sierra Nevada
Mountains of California. Ecology 12:1-17.

Muwnz, P. A. 1959. A California flora. Univ. California Press, Berkeley.

OrnburFF, R. 1974. An introduction to California plant life. Univ. California Press,
Berkeley. 152 p.

Pace, B. M. 1966. Geology of the Coast Range of California, p. 255-276. In E. H.
Bailey (ed.) Geology of Northern California. Bull. 190, Calif. Div. Mines, San
Francisco.

RunNpDEL, P. W., P. J. Parsons, and D. T. Gorpon. 1977. Montane and sub-alpine
vegetation of the Sierran Nevada and the Cascade Range, p. 559-599. In Bar-
bour, M. G. and J. Major (eds.) Terrestrial vegetation of California. Wiley and
Sons, New York.

U. S. Forest SERVICE. 1955. Vegetation-soil maps, legend and supplement. California
Forest and Range Experiment Station, San Francisco.

U. S. WEATHER BurEAv. 1964. Climatic summary of the United States, California.
Supplement for 1951 through 1960. No. 86-4. Washington, DAC.

WALKER, R. B. 1954. The ecology of serpentine soils. II. Factors affecting plant
growth on serpentine soils. Ecology 35:259-266.

WuHitTTakER, R. H., 1954. The ecology of serpentine soils. IV. The vegetational re-
sponse to serpentine soils. Ecology 35:275-288.

. 1960. Vegetation of the Siskiyou Mountains, Oregon and California. Ecol.

Monogr. 30:229-308.

. 1967. Gradient analysis of vegetation. Biol. Rev. 42:207-264.

REVIEW

Northwest Botanical Manuscripts. An indexed register of the papers, 1867-1957, of
Wilhelm Nicholaus Suksdorf, William Conklin Cusick, Charles Vancouver Piper,
Rolla Kent Beattie, and Harold St. John. By TERRY ABRAHAM. 64pp., 7 illus. Wash-
ington State University Library, Pullman. 1976. $5.00.

Botanical exploration of a country moves from travellers to residents, from the
surgeon naturalist with the expedition to the storekeeper, ranchman or local physi-
cian whose keen interest in the local flora leads him afield. We now have a shelf of
biographies of the explorers of Pacific Northwest: Douglas, Tolmie, Nuttall, Wilkes,
Cook, Vancouver, LaPerouse, Lewis, and others. Terry Abraham, archivist at Pull-
man, has provided a checklist of the letters and papers of five important resident
botanists: W. C. Cusick, whose papers date from 1906 to 1924; C. V. Piper, from
1888 to 1926; R. K. Beattie, from 1899 to 1956; W. N. Suksdorf, from 1867 to 1935;
and H. St. John, from 1912 to 1957.

Two botanists had planned to write a history of botany in the Pacific Northwest,
Albert Raddin Sweetser and R. K. Beattie. The Sweetser papers dating from 1897
to 1935, at the University of Oregon Library, are summarized in the National Union
Catalog of Manuscript Collections, 1959-1961 (1962) p. 73. They contain copies of
letters and documents related to such a history, but his death in 1940 curtailed his
plan. The Beattie papers at Pullman, numbering over 350 titles, listed in this publi-
cation, were likewise collected with a history in view. But Beattie died in 1960. The
historian interested in tracing early figures or topics fcr this region will profit par-
ticularly from these two collections. For later twentieth century subjects, the letters
at Pullman will likely prove most useful. The “Correspondence Index,” pages 38-62,
will ease the search since all five collections are integrated in this single index. Some

186 MADRONO [Vol. 25

of the representative multiple entries include: L. R. Abrams, the Brandegees, F. V.
Coville, A. Eastwood, M. L. Fernald, J. B. Flett, Martin Gorman, J. M. Greenman,
H. M. Hall, A. A. Heller, A. S. Hitchcock, T. Holm, T. J. Howell, W. L. Jepson, I.
M. Johnston, M. E. Jones, W. R. Maxon, A. Nelson, F. W. Pennell, B. L. Robin-
son, J. N. Rose, C. S. Sargent, W. A. Setchell, J. K. Small, W. Trelease, and C. A.
Weatherby.

Abraham has prefaced his checklist of manuscripts with “a historic overview” of
Northwest botany and illustrated it with photographs of the five botanists, unfor-
tunately undated, and a group photograph of Harold St. John, then age 32, and
associates, taken at the Pullman commencement of June 9, 1928. The “overview”
includes some erroneous statements, for example, (1) that “Eaton’s manual .
has only an incomplete list . . . of eastern plants.” Beginning with the fifth edition
of the Manual (1829) western species were included, for example, Pinus flexilis (p.
331), Quercus agrifolia (p. 354), Acer macrophyllum (p. 90), Vaccinium ovatum
(p. 434), etc. The eighth edition (1840) included California and Rocky Mountain
species (see p. 16). (2) Abraham’s statement that Pursh’s plants were lost for nearly
a century ignores the fact that Lambert’s herbarium was consulted freely by botan-
ists until its dispersal by auction in 1842. It was the American Philosophical Society
series of Lewis and Clark collections that were “lost”. (3) “Gray’s Manual...
through all its many nineteenth century editions never included the flora of the
state of Washington” is indeed a strange statement! The scope of Gray’s Manual
was patently stated to be the northeastern United States. (4) To say T. J. Howell’s
Flora was “deficient in the identifying keys needed by the practicing botanists”
and that it “did not prove very successful” can only reflect a want of understand-
ing of Howell the botanist. “Considering the circumstances of its production,” wrote
W. L. Jepson in the Dictionary of American Biography, “his [Flora] is balanced,
judicious, and highly useful.” That Howell’s Flora was not a mere compilation but
rested on his own field studies—more than 50 species were new—was the mark of
a vigorous mind.

To overlook Louis F. Henderson in an overview of Northwest botany would be
unfortunate. He was in the first scientific party to enter the rain forest of the
Olympic Peninsula (E. P. Thatcher in Call Number [Library University of Oregon]
22(1):20. 1960). A carefully prepared thesis entitled “A Contribution toward a bib-
liography of Oregon botany with notes on the botanical explorers of the state” by
Katherine W. Hughes (Oregon State College Thesis Series 14, 1940), 93 pages,
mimeographed, supplements Abrahams’ very useful checklist —Jos—EPpH Ewan, De-
partment of Biology, Tulane University, New Orleans, LA 70118.

BOOKS RECEIVED AND LITERATURE OF INTEREST

Nevada postal history 1861-1972. By Robert P. Harris. Nevada Publications, Box
15444, Las Vegas, NV 89113. 1973. $9.50 buckram. Of interest to botanists because
of the large number of local and now “extinct” place names.

Native plants for use in the California landscape. By Emile L. Labadie. Sierra
City Press, Box 2, Sierra City, CA 96125. 1978. $8.95 paper. A quote from p. 93
perhaps characterizes the botanical sophistication of this potentially useful book:
“Chamaecy paris lawsoniana .. . Flowers: Male flowers are bright red catkins.”

BIOSYSTEMATICS OF PSILOSTROPHE
(COMPOSITAE: HELENIEAE). II.
ARTIFICIAL HYBRIDIZATION AND
SYSTEMATIC TREATMENT

Roy CurTIss BROWN
Department of Biology
University of Southwestern Louisiana, Lafayette 70504.

Psilostrophe comprises seven species that grow in the semiarid regions
of the southwestern United States and adjacent Mexico. Members of
the genus are often conspicuous roadside weeds and are commonly known
as paperflowers. In spite of their conspicuousness in the field, their sus-
pected importance in livestock poisoning (Kingsbury, 1964; Schmutz,
Freeman, and Reed, 1968), and their weedy nature, these species have
remained poorly understood taxonomically.

Psilostrophe was first proposed by de Candolle (1838) from material
collected by Berlandier from the vicinity of San Luis Potosi, Mexico.
Three years later, the genus Riddellia, which would later prove to be
synonymous with Psilostrophe, was erected by Nuttall (1841). Riddellia
was typified by R. tagetina described from specimens collected by James
in the Rocky Mountains. In 1849, A. Gray described a second species of
Riddellia, R. arachnoidea, which he later realized (Gray, 1852, 1874)
was the same as Psilostrophe gnaphalodes. Greene (1891) transferred
the species of Riddellia to Psilostrophe. The first revisionary treatment
of Psilostrophe came in 1903 when A. Nelson recognized six species.
Rydberg (1914) reviewed the genus and named three additional species,
bringing the total number of described species to ten. In the last compre-
hensive treatment of the genus, Heiser (1944) reduced the number of
species to six. Heiser’s work was revisionary and left room for a modern
biosystematic study aimed at improving our understanding of the evo-
lutionary relationships within Psilostrophe, which in turn provides the
necessary background for further studies into adaptive strategies of
species and roles of translocation heterozygosity, B-chromosomes, and
polyploidy in the evolution of Psilostrophe.

DISTRIBUTIONS AND HABITATS

Populations of Psilostrophe are common elements of the summer flora
in North American deserts. All species are weedy under the right cir-
cumstances and certain of the species exhibit disjunct populations that
probably represent recent introductions. The variable P. tagetina is the
most widespread of the species and is found throughout much of New
Mexico, southeastern Utah, extreme eastern Arizona, Texas, mostly
west of the Pecos River, and south to the Mexican states of Chihuahua
and Coahuila (Fig. 1). It occurs in desert shrub, prairie grasslands, and
pinyon—juniper woodlands, often in sandy soils from 600 to 2400 m in

187

188 MADRONO [ Vol. 25

F © P. gnaphalodes * P. villosa
@P. tagetina @P. tagetina-
F | tagetina Interm © P. mexicana
]

j ~~ tok
i ee fage sires *
7 2% | | x
! © oe % oe 0% x *
° j + | x kK, |
e! e e e@ . * *y i
! e e @ * ee _~Vm eo
! Fe * * a Uv an a es
! ® @ ae ue tr * \
id a re ky
™ | e i * :
. fa, % °e@ phe
onl fe ee, °° a 100 km
Oo
~. % 6

Fic. 1. Distributions of Psilostrophe gnaphalodes, P. tagetina, P. gnaphalodes-P.
tagetina intermediates, P. villosa, P. tagetina-P. villosa intermediates, and P. mexicana.

elevation. The distribution of P. tagetina is overlapped by P. gnapha-
lodes and P. villosa in western Texas. Mixed populations containing P.
gnaphalodes and P. tagetina occur in Brewster County, Texas. Psilo-
stro phe villosa grows in prairie grassland from southwestern Kansas south
to Texas at 300 to 900 m. Psilostrophe gnaphalodes grows in western
Texas south to the Mexican state of San Luis Potosi from 300 to 2100 m.
The remaining species with a Chihuahuan Desert distribution is the tetra-

BROWN: PSILOSTROPHE

1978]
ae a
~~ {
i |
, |
H Eee SRNR ig) (ee
j |
i
!
l Proll
! ie’
j eS xe ee
tee ! OO i *
‘ je op i
\. jee o | *
\ e e je O Ber: O ont ase e a eee
\ efi O O amano pears
~ FY On ad je
ex 5 C6 | |
e.. le O | i
See \y ee O O | i
\ CO O | i
te i
o ff eee O
e f © se j * P= bakeri
- . 2 ° [ — @P. cooperi i
meenireicc tir) e A ! ; i
ee es e; OP. sparsiflora
e e

Fic. 2. Distributions of Psilostrophe bakeri, P. cooperi, and P. sparsiflora.

ploid, P. mexicana. It is encountered in hot desert shrubland to cool juni-
per woodlands in the states of Chihuahua and Durango, Mexico.

The three remaining species occur in or near the Mojave, Sonoran,
and Great Basin Deserts (Fig. 2). Psilostrophe cooperi is encountered in
the Mojave Desert and south and east to the northern Sonoran Desert
in Arizona and Mexico along washes, rocky flats and hillsides, often in
calcareous soils from 150 to 1500 m. Psilostrophe sparsiflora is generally
found in sandy soils in sagebrush or pinyon—juniper communities in the
cold Great Basin Desert from 900 to 2300 m. Psilostrophe bakeri is re-
stricted to arid, rocky hills in western Colorado from 1350 to 2000 m

190 MADRONO [Vol. 25

and is disjunct in Idaho. Housing development and increased cultivation
through irrigation threaten the habitat of this species.

HYBRIDIZATION EXPERIMENTS

A program of experimental hybridization was initiated to study the
crossing relationships of the species. A total of 2318 pistillate florets
were crossed in two years. Parental strains used were plants transplanted
from their native habitats or collected as seed and grown to maturity
in the greenhouse.

Capitula to be used as ovulate parents were emasculated by removal
of the hermaphroditic disc florets prior to anthesis. Pistillate ray florets
were then dusted with the pollen of the desired pollen parent. Heads of
both pollen and seed parents were covered with glassine bags through-
out the entire crossing procedure beginning with the bud stage to in-
sure against possible contamination. After pollination the cross was
tagged and allowed to mature under cover of the glassine bag. Achenes
matured in about six weeks. Crossability is expressed as percent filled
achenes following artificial hybridization. Estimates of parent and hy-
brid fertility were based on percent pollen stainability in aniline blue-
lactophenol. At least 500 grains were counted wherever possible. Pollen
stainability of parent plants ranged from 91 to 99 percent.

Intraspecific crossability was measured in interpopulational crosses
in P. cooperi, P. gnaphalodes, and P. tagetina and ranged from 48 to
57 percent achene set, with an overall crossability of 51 percent (Table
1). Interfertility among species was variable, ranging from 0 to 53 per-
cent achene set (Table 1; Fig. 3). Table 2 summarizes morphological
features of the species. In general the hybrids were intermediate to the
parental strains with respect to critical morphological characters. In all
cases reflexed ligules were dominant to horizontal ligules in fruit. Other
characters examined apparently were quantitatively inherited. Complete
data on pollen stainability, morphology, and meiosis in individual pa-
rental and hybrid plants are found in Brown (1974).

When coupled with morphological and geographical evidence, the
degree to which plants cross and produce vigorous hybrids and the
degree of fertility of the F, offspring have long been used as indicators
of genetic relationships of plants (Clausen, Keck, and Hiesey, 1941;
Long, 1975). Crossability and fertility as determined in the greenhouse
experiments are only of relative value and several limitations exist.
Ability of species to cross in the greenhouse is a poor measure of their
ability to hybridize in the field. Hybrids that apparently are vigorous in
the greenhouse may not be competitive in the habitats of their parents.
Furthermore, plants utilized in the experimental hybridizations repre-
sent only a minute sample of the total gene pool. That different plants
in the same population may have different interspecific crossabilities is
suggested by the crossing abilities of two individuals of P. tagetina col-
lected from the same population in southeastern Arizona. All attempts

1978]

Ovulate Parent

P. cooperi (3/3)*

P. gnaphalodes (3/3)

P. tagetina (5/4)

P. coopert (5/3)
Reciprocal
P. cooperi (6/3)
Reciprocal
P. cooperi (4/3)
Reciprocal
P. cooperi (3/3)
P. cooperi (2/1)
Reciprocal

P. sparsiflora (4/2)

Reciprocal

P. sparsiflora (2/1)

Reciprocal

P. sparsiflora (2/1)
P. sparsiflora (2/1)

Reciprocal

P. tagetina (3/3)
Reciprocal

P. tagetina (3/3)
P. tagetina (3/3)
Reciprocal

P. villosa (1/1)
Reciprocal

P. villosa (1/1)
Reciprocal

BROWN: PSILOSTROPHE 191
TaBLeE 1. CrossaBiLity IN Psilostrophe. *individuals/populations utilized.
Total Ovules % Achene
Pollen Parent Crossed Set Range
INTRASPECIFIC CROSSES
P. cooperi 29 48 (38-52)
P. gnaphalodes 37 57 (50-78)
P. tagetina 163 50 (46-78)
INTERSPECIFIC CROSSES
P. sparsiflora (1/1) 28 25 (19-50)
26 0 (0)
P. tagetina (3/2) 45 7 ( 0-50)
79 5 ( 0-13)
P. villosa (2/1) 24 0 (0)
24 0 (0)
P. bakeri (1/1) 18 0 (0)
P. gnaphalodes (1/1) 55 29 ( 5-45)
46 0 (0)
P. tagetina (4/3) 94 6 ( 0-23)
50 6 ( 0-22)
P. villosa (2/1) 36 0 (0)
ot 0 (0)
P. bakeri (1/2) 23 4( 0-17)
P. gnaphalodes (3/3) 27 7 ( 0-22)
98 4 ( 0-12)
P. villosa (2/1) 44 64 (44-72)
24 42 (11-60)
P. bakeri (1/1) 43 21 ( 0-42)
P. gnaphalodes (3/3) 137 50 (36-77)
195 44 ( 0-49)
P. bakeri (1/1) 25 4 (4)
20 0 (0)
P. gnaphalodes (4/4) 35 26 (26)
59 56 (33-69)
P. bakeri (1/1) 30 0 (0)

P. gnaphalodes (2/2)

P. coopert X
Sparsiflora

P. sparsiflora
tagetina

P. villosa
tagetina

P. tagetina «
gnaphalodes
P.c ooperi
gnaphalodes

F, Crosses
P. cooperi X
sparsiflora
P. sparsiflora
tagetina
P. villosa
tagetina
P. tagetina
gnaphalodes
P. cooperi *
gnaphalodes

only 1 plant flowered

TS

89

(Es)

0

1

only 1 plant flowered

192

TABLE 1, continued.

P. cooperi X
sparsiflora
Reciprocal
P. cooper: X
Sparsiflora
Reciprocal

P. sparsiflora <
tagetina
Reciprocal
P. villosa X
tagetina
Reciprocal
P. villosa <
tagetina
Reciprocal
P. villosa X
tagetina
Reciprocal
P. tagetina X
gnaphalodes
Reciprocal
P. tagetina
gnaphalodes
Reciprocal
P. cooperi X
gnaphalodes
Reciprocal
P..coopert <
gnaphalodes

MADRONO

BACKCROSSES

. Sparsiflora

. cooperi

. Sparsiflora

. tagetina

. tagetina

. villosa

. gnaphalodes

. tagetina

. gnaphalodes

P. cooperi

16
28

11

[Vol..25

oo

oO

TABLE 2. SUMMARY OF MORPHOLOGICAL SIMILARITIES AMONG SPECIES OF Psilostvophe.

Lower Pedun-
Vestiture Lvs cles Ligules
=
ee oe fe &
0) 9 3 = co ers =
gee ey ee
Se 8 Se ee Nt to
Sos &s & 2 3 Ea S ore S
oS We, Oo con ee ee ee
Species
P. cooperi + + + + +
P. sparsi-
flora a = ae ae a=
P. bakeri + oe + + +
P. tagetina + + + + +
P. villosa + + + + +
P. gnapha-
lodes aig oe: Sats 1 =r
P. mexicana ob + + + +

Pappus
Achenes 3
as aos
ee
o wo
cid
o oOo
; Popa etre
i 2’ 8
Z w ‘ 3 OSU
228 2e8XX
= = TS ies
Sr ls 07 VE
=n ae 35
+ + Le
+ + +
++ -.
+ + +
as + +
ate ote i

1978 | BROWN: PSILOSTROPHE 193

cooper!

7 “a
bad
sparsi -
flora
Ff t
e rt
a
< a rt
. 4 r
3% ‘y 1
; oe :
/ . :
rs ‘; :
en. 4 e F
oS s a

Fic. 3. Crossing diagram of the diploid species of Psilostrophe showing relation-
ships based on percent interfertility. Mature F, hybrids were not obtained from
crosses with less than 5 percent interfertility. Percentages above cross-bars indicate
mean pollen stainability in F, hybrids (N.A.* = no data, achenes failed to germi-
nate; N.A.** = no data, hybrids failed to flower).
to cross one individual with P. sparsiflora failed while attempts with
the other were 23 percent successful (Brown, 1974). This suggests that
all interspecific crossabilities would be higher if the ovulate parents were
covered with pollen from a large number of plants (genotypes) as might
occur under natural conditions. In addition, the backcrossing of F;,
hybrids to their immediate parents increases the likelihood of poor seed
set due to the operation of possible sporophytic incompatibility systems.
Recently, Jones (1976) presented evidence that environmental factors
may exert a considerable influence on the production of stainable and
presumed normal pollen. Nevertheless, data on crossability and pollen

194 MADRONO [Vol. 25

stainability in Psilostrophe seem in accord with morphological and geo-
graphical data.

Partial or complete internal barriers to gene exchange among species
of Psilostrophe occur in the initial crossability and fertility of first-
generation hybrids. Internal isolating mechanisms in Psilostrophe appar-
ently are genic rather than chromosomal. Even though translocation
heterozygosity is common in the genus, structural rearrangement has
not been a major factor in speciation for interspecific hybrids possessed
no more evidence of translocation heterozygosity than did their parents
(Brown, 1977). In no case was evidence for inversions detected in the
hybrids.

On the bases of morphology, geographical distribution, and crossing
relationships, the diploid species of Psilostrophe fall into two categories:
three distinct species in the Mojave, Sonoran, and Great Basin Deserts
and a species alliance of P. gnaphalodes—P. tagetina—P. villosa in the
southeastern portion of the range.

The southeastern species are very similar morphologically and are
not always readily distinguishable. The ranges overlap considerably in
western Texas, where intermediates occur. Artificial hybridizations
showed initial crossabilities in any combination of these three species
to be quite high (41 to 53 percent). Meiosis in the F, hybrids was regu-
lar with pollen stainability ranging from 44 to 93 percent. The inter-
specific hybrids showed an ability to backcross to either parent, but a
marked inability to produce second generation plants. Limited gene
flow between species seems possible in areas of sympatry and may pro-
vide an important source of genetic variability. Field studies of Pszlo-
strophe in trans-Pecos Texas should contribute to an understanding of
the extent and consequences of natural hybridization among these
species.

The morphology and ecology of the tetraploid P. mexicana suggests
that it is closely related to the P. gnaphalodes—P. tagetina—P. villosa
diploid complex. Psilostrophe mexicana was recognized late in the study
and its crossing relationships have not been studied. Without this infor-
mation additional speculation on the phyletic relationships of this inter-
esting taxon is unwarranted. Further work emphasizing artificial hy-
bridization is needed to trace the origin of P. mexicana.

Psilostrophe bakeri, P. cooperi, and P. sparsiflora occur in specialized
habitats in the northern and western portions of the genus range. Each
species is ecologically and geographically distinct. Major morphological
gaps involving several characters are suggestive of major genetic dis-
continuities among these three species. The three species are well isolated
reproductively by low initial crossability in addition to external factors.
Taxonomic decisions within this group are relatively simple and mis-
identification rare.

The species are allied genetically to P. gnaphalodes and P. tagetina

1978 | BROWN: PSILOSTROPHE 195

of the southeastern alliance. Of 101 crosses attempted between P.
cooperi and P. gnaphalodes 16 percent produced filled achenes. Six full
achenes resulted from 125 hybridizations between P. sparsiflora and P.
gnaphalodes, an overall crossability of 5 percent. Meiosis was regular in
F, hybrids of either combination with 16 bivalents undergoing an orderly
segregation (Brown, 1977). Pollen stainability ranged from 44 to 60
percent. Complete chromosomal pairing occurred in hybrids of species
widely divergent in geography, ecology, and morphology. The overall
weakness and low pollen fertility in the hybrids suggest some genetic
disharmony.

Psilostrophe cooperi and P. sparsiflora showed much less crossing
affinity to P. tagetina: 5 and 6 percent overall crossability respectively.
These figures may be misleading for individuals of P. tagetina from the
same population exhibited differing abilities to hybridize as mentioned
earlier. None of the achenes resulting from crosses between P. cooperi
and P. tagetina germinated. Pollen stainability in four hybrids between
P. sparsiflora and P. tagetina ranged from 51 to 55 percent. Low hybrid
fertility in attempts to produce an F: generation and backcrosses is evi-
dence for internal isolation in addition to the eco-geographical isolation
of these species.

Psilostrophe bakeri showed a 21 percent crossability with P. tagetina
but little or no crossability with any other species. The simplest interpre-
tation of this is to regard P. bakeri as derived from P. tagetina or its
immediate ancestor. This distinct taxon is apparently reproductively as
well as geographically isolated.

TAXONOMIC TREATMENT

The treatment of Psilostrophe is based on extensive field and herbar-
ium studies in addition to studies of cytotaxonomy (Brown, 1977) and
artificial hybridization. Most taxonomic difficulties in the genus arise
because of substantial morphological similarities among the species.
While in essential agreement with the capable revision of Heiser (1944),
this study has brought to light additional characters useful in delimiting
the species. A key to the species incorporating these characters is pro-
vided. In addition to reflecting increased knowledge about the geo-
graphical range and ecology of the species, this revision differs from
that of Heiser in the recognition of P. mexicana and in the treatment of
intraspecific variation in P. tagetina.

PstLOSTROPHE de Candolle, Prodr. 7:261. 1838. Type: Psilostrophe
enaphalodes de Candolle.

Riddellia Nuttall, Trans. Amer. Philos. Soc. n.s. 7:371. 1841. Tver:
Riddellia tagetina Nuttall.
Perennial or biennial, leafy-stemmed herbs. Stems simple below, clus-

tered from a woody taproot often covered with old leaf bases, branching

196 - MADRONO [Vol: 25

above. Leaves both rosulate and cauline or all cauline, alternate. Lower
leaves spatulate to oblanceolate or linear, margins entire or pinnately
lobed. Upper leaves much reduced, sessile and usually entire. Vegetative
organs sparingly pilose to densely woolly or sometimes floccose. Capitula
subsessile to long pedunculate, solitary or in corymbose clusters. Invo-
lucres cylindric; phyllaries 5-10, elliptic or lanceolate, herbaceous and
uniseriate, or, if a second series is present, then the inner 1-7 bracts
smaller than the outer and scarious. Corollas yellow to orange. Ray
florets pistillate, fertile, 3-7 in a single series; the ligules becoming
papery and persistent on the ripe achenes. Disc florets perfect, fertile;
corolla tubes cylindric, the 5 triangular lobes glandular-pubescent with-
out. Achenes terete or slightly angled, sublinear, conspicuously striate,
glabrate to villous. Pappus of 4-6 nerveless, lanceolate, subequal, hya-
line squamellae; the margins frequently erose, sometimes lacerate-
dissected. Base chromosome number, « = 16.

KEY TO PSILOSTROPHE

Plants suffrutescent to subshrubby; pubescence of the stem densely
white-pannose; lower leaves linear; heads scattered on peduncles
3-8 cm long. . . si pl SPecooper
Plants herbaceous; pubeccence of ‘lat en aol to sparingly pilose;
lower leaves Soanleee to oblanceolate; heads in corymbose clusters,

the peduncles less than 5 cm long.
Stems green, sparingly pilose, upper foliage glandular-dotted; ligules

tightly reflexed against the involucre at maturity .

ve BART) oe arinere
on loosely milled: fo wane Fooly Panis remaining horizontal in

fruit.
Ray florets 5-6; disc florets 10-20; pappus scales less than one
half the length of the disc corolla . . . . . 3.P. bakeri

Ray florets 2—5; disc florets 5-12; pappus scales one half to equal-
ing the disc corolla in length.
Achenes and pappus glabrate; pappus scales entire or merely

erose.
Heads in open corymbs; peduncles 1-4 cm long; ligules 5-14
mm long, shallowly 3-lobed . . . . . 4. P. tagetina

Heads in congested corymbs; peduncles nearly wanting to 10
mm ie ligules 3-6 mm long, deeply 3-lobed ;

: UD aa villosa
Achenes and SEI aes vllode: eine scales peeraree dissected.
Heads in congested corymbs; peduncles nearly wanting to 10
mm long; disc corollas 3.5—-4.0 mm long. 6. P. gnaphalodes

Heads in open corymbs; peduncles 8-30 mm long; disc corollas
4°5-5:0 mm long . 9.) . « 2). « a 4. Pemexicana

1. PSILOSTROPHE COOPERI (A. Gray) E. L. Greene, Pittonia 2:176.

1978] BROWN: PSILOSTROPHE 197

1891.—Riddellia cooperi A. Gray, Proc. Amer. Acad. Arts 7:358.
1868.—Typr: Gray cited the following specimens as representative,
“Gravelly banks at Fort Mohave, Dec. 1861, Dr. J. G. Cooper. On
the Colorado, Dr. Newberry. Camp Grant, &c., Arizona, Drs. Elliot
Coues and Edward Palmer.’ Gray (1874): clearly stated that the
species was first collected by Cooper, and Heiser (1944) designated
the Cooper specimen (GH!) lectotype.

Perennial suffrutescent plants, 20-60 cm tall. Stems densely white-
pannose, becoming less so with age, freely branched. Leaves cauline,
lanate to glabrate, linear, entire, 1-8 cm long, less than 0.5 cm wide.
Heads scattered, terminating the many branches; peduncles slender,
3-8 cm long. Involucres loosely to densely lanate, cylindric, 6-8 mm
high and 3—5 mm in diameter. Corollas yellow. Ray florets 3-6, ligules
8-18 mm long and nearly as broad, 3-lobed, enlarging and becoming
reflexed in fruit. Disc florets 10-25, corollas 4-5 mm long. Achenes gla-
brous to sparsely glandular with sessile glands; pappus scales oblong-
lanceolate, entire to erose, obtuse to acute, less than ™% the length of
the disc corolla. Chromosome number: 27 = 32. Flowering throughout
the year; mainly March to June.

2. PSILOSTROPHE SPARSIFLORA (A. Gray) A. Nelson, Proc. Biol. Soc.
Wash. 16:23. 1903.—Riddellia tagetina var. sparsiflora A. Gray, Syn-
opt. Fl. N. Amer. 1(2):318. 1884.—TyprE: Gray cited two specimens
from southern Utah in his description of this variety. Heiser (1944)
selected the specimen of Captain Bishop s.n. as lectotype (GH! ).—
Psilostrophe tagetina var. sparsiflora (A. Gray) E. L. Greene, Pittonia
2:176. 1891.

Psilostro phe divaricata Rydberg, North Amer. Flora 34:8. 1914.—TYPE:
United States: Arizona: ‘Grand Canyon of the Colorado,” 1897, D.
T. Allen s.n. (Holotype, NY!; isotypes NY! UC!).

Perennial herbs, 10-50 cm high. Stems sparingly pilose to glabrate,
single or clustered from a woody caudex, often twisted and zigzag at the
nodes; freely branched, the branches more or less strongly divergent.
Basal leaves rosulate, villous when young, becoming glabrate with age,
spatulate to linear, up to 14 cm long, usually less than 1.5 cm wide,
entire or rarely pinnately lobed, frequently lacking in mature specimens.
Cauline leaves smaller, sparsely villous to glabrate, narrowly oblanceo-
late to linear, entire, acute to obtuse, dotted with sessile glands particu-
larly near the inflorescence. Heads born in loose cymes of 3—6 on slen-
der peduncles up to 3 cm long. Involucres lightly villous, cylindric, 5 mm
high and 3 mm in diameter. Corollas yellow. Ray fore 1-4, often 2 or
3, ligules 6-10 mm long and decidedly broader, 3-lobed, enlarging and
becoming sharply reflexed in fruit. Disc florets fewer than 10, corollas
3—5 mm long. Achenes glabrous to sparsely glandular with sessile glands;
pappus scales lanceolate to linear, subequal, frequently erose, 1% to 24

198 MADRONO [Vol. 25

the length of the disc corolla. Chromosome number: 2” = 32. Flower-
ing from April to October.

3. PSILOSTROPHE BAKERI E. L. Greene, Pl. Baker. 3:29. 1901.—T ype:
United States: Colorado: Montrose, Baker 14 (NDG). Greene based
his description of this species on two collections, Baker 14 from near
Montrose, Colorado and Baker 106 from near Grand Junction, Colo-
rado. Both collections are large gatherings with many duplicates and
either could have provided the characters used in the diagnosis. In
citing the type locality as Montrose, Colorado, Rydberg (1914), with-
out stating his reason, implied that Baker 14 is type, hence the speci-
men in the Greene Herbarium is designated as lectotype, (isolecto-
types, GH! MO! POM! US!).

Riddellia tagetina var. pumila M. E. Jones, Proc. Calif. Acad. Sci. ser.
2, 5:700. 1895.—Typr: United States: Colorado: Grand Junction,
in gravel, in open places, 21 Jun 1894, M. E. Jones 5474 (Lectotype
here designated POM!; isolectotypes BM! MO! NY! POM! UC!).—
Psilostrophe pumila (M. E. Jones) A. Nelson, Proc. Biol. Soc. Wash.
16222. 1903.

Perennial herbs, 5-30 cm high. Stems long-villous, one to several from

a branched woody caudex. Lower leaves rosulate, loosely villous, spatu-

late to oblanceolate, up to 8 cm long, entire or rarely pinnately 3- to

5-lobed. Cauline leaves smaller, oblanceolate, entire. Heads in loose
corymbs at the end of the branches on peduncles 1.5—5.0 cm long. In-
volucres lightly villous, cylindric, 7-10 mm high and 4-6 mm in diam-
eter; of 9 distinct bracts. Corollas yellow-orange. Ray florets 5—6, ligules

8-15 mm long and nearly as broad, 3-lobed. Disc florets 10—20, tubular

corollas 4-5 mm long. Achenes glabrous, striate; pappus scales unequal,

short erose, decidedly less than % the length of the disc corolla. Chro-
mosome number: 2” = 32. Flowering May to July.

4. PSILOSTROPHE TAGETINA (Nuttall) E. L. Greene, Pittonia 2:176.
1891.—Riddellia tagetina Nuttall, Trans. Amer. Philos. Soc. n.s.
7:371. 1841.—Type: “The southern range of the Rocky Mountains,
towards the sources of the Platte.” Probably collected by Dr. James
on Long’s expedition (Gray, 1849) (Holotype BM!; isotype GH!).
As pointed out by Heiser (1944) Nuttall’s spelling in the original pub-
lication, “Tagetinae,’ was probably a misprint and subsequent work-
ers have used the gramatically correct spelling, “tagetina.”

Psilostro phe tagetina var. lanata A. Nelson, Proc. Biol. Soc. Wash. 16:21.
1903.—Tvpr: United States: Texas: El] Paso, Apr 1881, G. R. Vasey
s.n. (Holotype US!).—P. lanata (A. Nelson) Hay, Miller & White,
Proc. Biol. Soc. Wash. 16:186. 1903.

Psilostrophe hartmaniu Rydberg, North Amer. Flora 34:8. 1914.—TyPE:
Mexico: Chihuahua: near Laguna de Guzman, C. V. Hartman 726
(Holotype NY!; isotype GH!).

1978] BROWN: PSILOSTROPHE 199

Psilostrophe grandiflora Rydberg, North Amer. Flora 34:8. 1914.—TyPE:
United States: Arizona: Cochise Co.: near Cedar Gulch, Paradise, 21
Sept 1907, J. C. Blumer 1709 (Holotype NY!; isotypes FM! GH!
MO! ARIZ! UC!).—P. tagetina var. grandiflora (Rydberg) C. B.
Heiser, Ann. Missouri Bot. Gard. 31: 292. 1944.

Perennial herbs, 15-50 cm tall. Stems loosely to densely villous, be-
coming less so with age, clustered from the crown of a woody taproot or
single-stemmed in the first year of growth, usually freely branched above,
forming globose clumps. Lower leaves rosulate, loosely to densely white-
villous, narrowly oblanceolate to more commonly spatulate, up to 15 cm
long, less than half as wide, entire to pinnatisect with lanceolate to
broadly linear segments. Upper leaves smaller, oblanceolate to linear,
lightly villous, eglandular or nearly so. Heads terminating the many
branches in open corymbs; peduncles 1—4 cm long. Involucres densely
white-villous, 5-6 mm high and 2—4 mm in diameter. Corollas yellow to
orange. Ray florets 3-6, commonly 3, ligules 5-14 mm long and usually
noticeably broader, broadly 3-lobed at the apex. Disc florets 5—12, corol-
las 3-5 mm long. Achenes glabrous or with a few stout trichomes when
young, striate, terete or slightly angled. Pappus scales broadly to nar-
rowly lanceolate, acute to obtuse, one half to equalling the disc corolla in
length. Chromosome number: 2” = 32. Flowering May to October.

Nelson (1903) and Heiser (1944) have commented on the morpho-
logical diversity of this species. It is also variable cytologically (Brown,
1977). In addition to the typical form, two varieties, var. /anata and
var. grandiflora, were recognized by Heiser (1944) and others. Both
varieties are based on differences in size, particularly of the ligules and
peduncles, and degree of pubescence, characters that are extremely vari-
able. My observations lead me to believe that recognition of infraspecific
taxa is unjustified. I have visited the type locality of var. grandiflora on
several occasions and found the population to be morphologically and
chromosomally variable. The type collection is definitely extreme in
size. Comparable plants as to size of parts and degree of pubescence are
uncommon. No cytological character such as translocation heterozygosity
or number of supernumerary chromosomes could be correlated with any
morphological character. Unless further investigations prove otherwise,
I favor a conservative treatment of this variable and widespread species.

5. PSILOSTROPHE VILLOSA Rydberg in Britton, Manual Flora North.
States Canada. 1006. 1901.—TyPr: Original description and illustra-
tion (no authentic type specimen located).

Psilostrophe cerifera A. Nelson, Proc. Biol. Soc. Wash. 16:21. 1903.—
Type: “Cheyenne Country, Indian Territory.” Jun 1891, M. A.
Carleton 201 (Holotype RM, paratype US!).

Psilostrophe cerifera var. biennis A. Nelson, Proc. Biol. Soc. Wash.
16:21, 1903.—Typr: United States: Kansas: Meade Co.: prairie
near Crooked Creek 16 Aug 1890, B. B. Smyth 140 (Holotype US!,

200 MADRONO [Vol 25

isotype NY!).—P. biennis (A. Nelson) Hay, Miller & White, Proc.

Biol. Soc. Wash. 16:186, 1903.

Biennial or perennial herbs, 20-60 cm high. Stems loosely to densely
long-villous, becoming less so with age. Basal leaves rosulate, densely
long-villous, spatulate to oblanceolate, up to 10 cm long and less than
5 cm wide, entire or occasionally 3- to 5-lobed. Heads several at the ends
of the branches in congested corymbs; peduncles subsessile to 5 mm
long. Involucres densely white-woolly, 5-7 mm high and 3-4 mm in
diameter. Corollas yellow-orange. Ray florests usually 3, ligules 3-6 mm
long usually broader than long, deeply 3-lobed. Disc florets 5—10, corol-
las 3-5 mm long. Achenes glabrate; pappus scales linear-lanceolate,
acute, one half to equaling the disc corolla in length. Chromosome num-
ber: 2n = 32. Flowering from April to October.

6. PSILOSTROPHE GNAPHALODES de Candolle, Prodr. 7:261. 1838.—
Type: Mexico: San Luis Potosi, Berlandier 1336 (Holotype G, photo-
graphs FM! US!; isotype GH! probable isotypes BM! MO! ).—Rid-
dellia gnaphalioides (de Candolle) O. Hoffman in Loesner, Bull. Herb.
Boissier 3:628. 1895.

Riddellia arachnoidea A. Gray, Mem. Amer. Acad. Arts ser. 2, 4:94.
1849.—_Typr: Three specimens are cited by Gray, “dry soil around
Buena Vista and Saltillo, Dr. Gregg, Dr. Wislizenus: also near Mon-
terrey, Dr. Edwards.’ All three specimens are mounted on a single
sheet in the Gray Herbarium. The Gregg specimen is here selected as
lectotype (GH!).

Perennial herbs, 20-50 cm high. Stems loosely to densely villous, one
to many from a woody taproot. Lower leaves rosulate, loosely to densely
villous, spatulate to oblanceolate, up to 6 cm long and 1.5 cm broad,
entire or occasionally pinnately 3- to 7-lobed. Upper leaves smaller,
linear to oblanceolate, entire. Heads several at the ends of the branches
in congested corymbose clusters; peduncles subsessile to nearly 1 cm
long. Involucres densely white-woolly, 5-6 mm high and about 3 mm
in diameter. Corollas yellow-orange. Ray florets 2—4, generally 3, ligules
4-7 mm long, usually slightly broader than long. Disc florets 5-9,
corollas 3.5—4.0 mm long. Achenes and pappus scales long-villous, pappus
scales linear-lanceolate, margins lacerate-dissected, about one half the
length of the disc corolla. Chromosome number: 2” = 32. Flowering
throughout the year; mainly March to September.

7. PSILOSTROPHE MEXICANA R. C. Brown, Brittonia 26:115. 1974.—
Type: Mexico: Chihuahua: desert shrub community dominated by
Larrea, Prosopis and Parthenium, 8.1 mi N of junction to Jimenez
along Hwy. 45, Pinkava, McGill & Brown 788 (Holotype ASU!; iso-
types to be distributed).

Perennial herbs, 15-50 cm tall, with one to many ascending, long-
villous stems. Lower leaves rosulate, loosely to densely villous, oblan-

1978 | BROWN: PSILOSTROPHE 201

ceolate to linear-oblanceolate in outline, entire or rarely lobed to sub-
pinnatifid. Cauline leaves smaller, narrowly oblanceolate to linear.
Heads several at the ends of the branches in open corymbs, peduncles
10-30 mm long. Involucre cylindric, woolly, 3-4 mm in diameter and
5-8 mm high. Corollas yellow-orange. Ray florets 3—5; ligules obovate
to orbicular in outline, 5-8 mm long, 3-lobed. Disc florets 7-10; corollas
4.3-5.0 mm long. Achenes sublinear-oblong, long-villous with trichomes
similar to those of the herbage; pappus scales lanceolate to linear, mar-
gins lacerate-dissected into long hairs. Chromosome number: 2” = 64.
Flowering from July to November.

ACKNOWLEDGMENTS

I thank the curators of the following herbaria for loans of specimens or courtesy
extended during my visits: ARIZ, ASU, BM, DHA, F, GH, MO, MSC, NY, POM,
SMU, UC, UMO, US. I am grateful to Donald J. Pinkava for his guidance in this
study, which formed part of a thesis submitted in partial fulfillment of the require-
ments of the Ph.D. degree at Arizona State University, Tempe. The manuscript has
benefited from critical reviews by John L. Strother and Billie L. Turner. Special
thanks are due to many people who assisted me in obtaining living materials of
Psilostrophe. In particular, I gratefully acknowledge the help of M. G. McLeod,
D. J. Keil, and L. A. McGill.

LITERATURE CITED

Brown, R. C. 1974. Biosystematics of Baileya and Psilotrophe (Compositae,
Helenieae). Ph.D. Thesis, Arizona State University, Tempe.

. 1977. Biosystematics of Psilostrophe DC. (Compositae). I. Chromosome
variability. Rhodora 79:169-189.

CLaAuUsEN, J., D. D. Keck, and W. M. Hiesey. 1941. Experimental taxonomy. Carne-
gie Inst. Wash. Yearbook 40: 160-170.

DE CANDOLLE, A. 1838. Prodromus Systematis Naturis Regni Vegetabilis 7:261.

Gray, A. 1849. Plantae Fendlerianae Novi-Mexicanae. Mem. Amer. Acad. Arts. ser.
2, 4:1-116.

. 1852. Plantae Wrightianae Texano-neo-Mexicanae. Smithsonian Contr.
Knowl. 3(5) :1-146.

———\. 1874. Notes on Compositae and characters of certain genera and species.
Proc. Amer. Acad. Arts 9:195.

GREENE, E. L. 1891. Some neglected priorities in generic nomenclature. Pittonia
2:173-184.

Hetser, C. B. 1944. Monograph of Psilostrophe. Ann. Missouri Bot. Gard. 31:279-300
+ 1 plate.

Jones, A. G. 1976. Environmental effects on the percentage of stainable and pre-
sumed normal pollen in Aster (Compositae). Amer. J. Bot. 63:657-663.

KincsBury, J. M. 1964. Poisonous plants of the United States and Canada. Pren-
tice-Hall, Inc., Englewood Cliffs.

Lonc, R. W. 1975. Artificial interspecific hybridization in temperate and tropical
species of Ruellia (Acanthaceae). Brittonia 27:289-296.

NEtson, A. 1903. Psilostrophe, a neglected genus of southwestern plants. Proc. Biol.
Soc. Wash. 16:19-24.

NvuTTALL, T. 1841. Descriptions of new species and genera of plants in the natural
order Compositae, collected in a tour across the continent to the Pacific, a resi-
dence in Oregon, and a visit to the Sandwich Islands and upper California,
during the years 1834 and 1835. Trans. Amer. Philos. Soc. n.s. 7:371.

RypbBeErG, P.A. 1914. Carduaceae: Helenieae: Riddellianae. North Amer. Flora 34:6—11.

ScHMuvTZ, E. M., B. N. FREEMAN, and R. E. REEp. 1968. Livestock-poisoning plants
of Arizona. Univ. Arizona Press, Tucson.

GERMINATION OF COMANDRA (SANTALACEAE)

Jos Kurjt
Department of Biological Sciences, University of Lethbridge
Lethbridge, Alberta T1K 3M4, Canada

In all parasitic angiosperms germination corresponds to a_ brief
period of independence from host plants. During this short span of time,
when growth is facilitated by nutrients present in endosperm and/or
cotyledons, the seedling has an opportunity to establish structural and
physiological contact with a host plant. It is a rather precarious transi-
tional period of considerable biological interest. Surprisingly little is
known, however, about the seedling stage of many parasites including
members of Santalaceae, all of which are parasitic. The following struc-
tural details relate to the early establishment of Comandra umbellata
(L.) Nutt. subsp. pallida (DC.) Piehl, as based on seedlings found in
loose sand in Lethbridge, Alberta, 18 May 1976.

Seedlings of Comandra show exceedingly slender, erect stems with
narrow, erect leaves when first emerging from the soil (Fig. 1A). They
are difficult to spot among older plants, which normally have numerous
sterile, unbranched shoots. About 6 or 7 cm below soil level the entire
fruit was still recognizable, the fruit wall turned nearly black, and both
it and the hard mesocarp split open on the apical pole. From this pole
there protruded the withered endosperm still enclosing the 10 mm long
cotyledons, which were separate at the base but closely adnate in the
upper one-third (Fig. 1B, 1F). The cotyledons, which were only a frac-
tion of a mm at the time of fruit dispersal (Piehl, 1965), remain in that
position until they decay.

The initial stages of germination are clearly described and illustrated
by Piehl. The precise origin of roots and rhizomes, however, needs fur-
ther comment. Piehl states that the first rhizomes originate “in the
transition or cotyledon region” of the axis. In my material there was
no sign of lateral buds associated with cotyledons, while buds were
clearly recognizable in the axils of the first one or two reflexed scale
leaves (Fig. 1C). It is thus at this level, several cm below the soil surface,
that the rhizomes originate in axillary positions. In several plants axillary
roots had also formed in association with both these scale leaves and the
cotyledons (Fig. 1C, 1D). Where both a bud and a root are formed in an
axillary position, they already show the same regular position with re-
spect to each other that characterizes the mature rhizome system (Fig.
1C; cf. Kuijt, 1969, Fig. 3-6).

Haustoria are formed even two weeks following germination. Many
haustoria in the seedling are closely similar to the so-called “haustorial
rudiments” that Simpson and Fineran (1970) described for Mida. They
are rather elongate, spur-like outgrowths clearly continuous with that

202

1978] KUIJT: COMANDRA 203

Fic. 1. Comandra umbellata. A. Seedling; soil level indicated by broken line.
B. Cotyledonary zone. Within the cracked fruit wall the yellowish shell is visible,
from which the tubular remnant of the endosperm, containing the twisted cotyle-
dons, protrudes. Compare D and F. C. One of several subterranean nodes, about 4
cm above cotyledons, showing recurved scale-leaf and axillary root, between which
is the axillary bud (arrow). The axillary root probably develops into a rhizome.
D. Cotyledonary zone, showing twisted cotyledons still in exhausted endosperm, one
cotyledon having an axillary root. E. Haustorial “rudiment’’, 4 mm along first lateral
root, 5 mm below cotyledons. F. Cotyledonary zone with endosperm removed, the
cotyledon tips cohering. Two axillary roots have formed, each bearing one young
haustorium.

204 MADRONO [Vol. 25

of the mother root (cf. Kuijt, 1969, Fig. 5). These rudiments thicken
upon contact with host organs, and many eventually differentiate into
haustoria. They develop on any root, whether below the cotyledonary
node or above it, sometimes only a few mm away from the shoot in the
latter cases (Fig. 1F), except for the upper portion of the primary root,
which always seems to lack haustoria.

The only other Santalaceae where germination has been described
are Exocarpos (Stauffer, 1959; Fineran, 1962) and Santalum album
(Barber, 1906). Both these plants, and quite probably also Buckleya
(cf. Kusano, 1902) are epigaeous and non-rhizomatous. Because Co-
mandra is hypogaeous and rhizomatous it may be predicted that other
rhizomatous genera such as Arjona, Geocaulon, Nanodea, and possibly
Nestronia also have a hypogaeous type of germination.

LITERATURE CITED

BarBer, C. A. 1906. Studies in root parasitism. The haustorium of Santalum album.
1. Early stages, up to penetration. Mem. Dept. Agriculture India, Bot. Ser.
1:1-30.

FINERAN, B. A. 1962. Studies on the root parasitism of Exocarpus (sic) bidwillii
Hook. f. I. Ecology and root structure of the parasite. Phytomorphol. 12:339-
S55,

Kurt, Jos. 1969. The biology of parasitic flowering plants. Univ. California Press,
Berkeley and Los Angeles.

Kusano, S. 1902. Studies on the parasitism of Buckleya quadriala, B. et H., a Santa-
laceous parasite, and on the structure of its haustorium. J. Coll. Sci., Imp. Univ.
Tokyo 17:1-12.

Pireut, M. A. 1965. The natural history and taxonomy of Comandra (Santalaceae).
Mem. Torrey Bot. Club 22:1-97.

Smrpson, P. G. and B. A. FIneran. 1970. Structure and development of the hausto-
rium in Mida salicifolia. Phytomorphol. 20:236—248.

STAUFFER, H. U. 1959. Revisio Anthobolarum. Santalales-Studien IV. Mitt. Bot.
Mus. Univ. Zurich, No. 213.

CHROMOSOME NUMBERS IN XYLORHIZA NUTTALL
(ASTERACEAE — ASTEREAE)

Tuomas J. WATSON, JR.
Department of Botany, University of Montana, Missoula 59812

X ylorhiza is a genus of eight species of the western United States and
Mexico. The plants are suffruticose perennials or small shrubs that
grow and flower in early spring. Most of the taxa have limited distribu-
tions in remote areas. Previously, the genus has been studied largely
from the few specimens available in herbaria. Chromosomal data pre-
sented in this paper were obtained during the course of a biosystematic
investigation of the genus (Watson, 1977).

Chromosome counts have been reported previously for only four
species of the genus: X. tortifolia (Raven et al., 1960); as Machaeran-
thera tortifolia), X. wright (Turner, 1964; Powell and Sikes, 1970; as
M. wrightii; Urbatsch, 1974), X. glabriuscula (Solbrig et al., 1969; as
M. glabriuscula) and X. frutescens (Anderson et al., 1974; as M. fru-
tescens). All plants counted previously were diploids with 2n = 12.

MATERIALS AND METHODS

Achenes and/or immature capitula for chromosome counts were col-
lected from populations throughout the range of each species of Xylo-
rhiza. Fruits and/or inflorescences were taken from one to five plants at
each site. Immature heads were fixed in modified Carnoy’s solution (4
chloroform: 3 ethanol: 1 glacial acetic acid; v/v/v). Aceto-carmine
squashes of anthers were obtained by the method of Turner and Johnston
(1961).

Seeds were germinated on moist filter paper in petri dishes. Emerging
root tips were pretreated for four hours in a saturated solution of para-
dichlorobenzene. The root tips were then fixed, hydrolyzed, stained, and
squashed by the technique of Huziwara (1957).

RESULTS AND DISCUSSION

Chromosome counts from 118 stands of Xvlorhiza are recorded in
Table 1. In addition, chromosome numbers for species previously thought
to belong in Xylorhiza are included here or have been published else-
where (Watson, 1973). Meiotic chromosome behavior was studied in the
available taxa; chromosomes of most species were studied at mitotic
metaphase.

Chromosome numbers of all taxa in Xylorhiza are now known; the
base number for the genus is x = 6. Populations of most taxa are uni-
formly diploid with 2m = 12. Tetraploids (2m = 24) were found only
in X. tortifolia, X. venusta, and X. glabriuscula var. linearifolia.

In most instances, meiosis in the diploids was regular with the forma-

205

206 MADRONO [Vol. 25

TABLE 1. CHROMOSOME NUMBERS OF X ylorhiza spp. AND Aster kingiz. Chromo-
some counts determined from mitotic cells are denoted by an asterisk (*); other
counts are from pollen mother cells. Populations with individuals having fragments
are indicated by the superscript f. Collection numbers refer to T. J. Watson; vouch-
ers are in TEX.

X ylorhiza cognata (H. M. Hall) T. J. Watson
2n = 12 CALIF.: Riverside Co., 365, 366, 606.
Xylorhiza confertifolia (Cronquist) T. J. Watson
2n = 12 UTAH: Garfield Co., 312*, 313, 696, 697.
X ylorhiza glabriuscula Nuttall var. glabriuscula
2n = 12 COLO.: Moffat Co., 891. MONT.: Carbon Co., 470**. UTAH: Daggett
Co., 452. WYO.: Albany Co., 488, 489; Carbon Co., 479, 480, 481, 482, 484, 485,
487 ; Natrona Co., 473, 478; Sweetwater Co., 455, 458; Uinta Co., 462; Washa-
kie Co., 463*, 464, 465.
X ylorhiza glabriuscula var. linearifolia T. J. Watson
2n = 12 UTAH: Grand Co., 308*, 435, 436, 908.
2n = 24 UTAH: Grand Co., 680, 905, 914, 916.
X ylorhiza orcuttii (Vasey & Rose) Greene
2n = 24 CALIF.: San Diego Co., 364*, 603; Imperial Co., 604.
X ylorhiza tortifolia (Torrey & Gray) Greene var. tortifolia
2n = 12 ARIZ.: Mohave Co., 385, 386; Yavapai Co., 387, 388; Yuma Co., 610.
CALIF.: Inyo Co., 376, 377, 727; Kern Co., 372, 373; 374, 375; Riverside Co.,
607; San Bernardino Co., 384, 722, 723. NEV.: Clark (Co.5 380).712, 715,716;
717, 718, 720, 721; Nye Co., 378, 379, 728, 729, 730. UTAH: Grand Co., 316*.
2n = 24 NEV. Clark Co., 387, 382, 353, 713.
X ylorhiza tortifolia var. imberbis (Cronquist) T. J. Watson
2n = 12 UTAH: Grand Co., 309%, 310, 911, 912, 913, 915.
X ylorhiza venusta (M. E. Jones) Heller
2n = 12 COLO.: Delta Co., 429*, 430*, 431*, 665, 666, 667; Moffat Co., 655, 656,
894; Montrose Co., 427*, 428*, 668, 669; Rio Blanco Co., 449*, 654, 898.
UTAH: Carbon Co., 918; Emery Co., 304* ; Grand Co., 305*, 306* ; Uinta Co.,
451*, 652°, 653, 895.
2n = 24 COLO.: Mesa Co., 432*, 662*. UTAH: Garfield Co., 900; Grand Co.,
433*, 682, 683, 687, 690, 901, 904.
X ylorhiza wrighti (A. Gray) Greene
2n — 12 TEX.: Brewster Co., 401*, 403*, 626; Jeff Davis Co., 411*; Presidio Co.,
408*, 409.
Aster kingit D. C. Eaton
2n. = 18: UTAH: Salt Lake Co., 766,

tion of six bivalents (Figs. 1-10) followed by normal disjunctions. In a
few individuals of X. glabriuscula from Wyoming, a bridge at anaphase
I was observed, suggesting that the plants were heterozygous for seg-
mental rearrangements on one pair of chromosomes. Also, in a few
individuals of X. glabriuscula and X. venusta, a pair of centric frag-
ments was observed in pollen mother cells (Fig. 1) and/or in root tip
cells. The fragments synapse and disjoin during meiosis I. At mitotic
metaphase, the fragments are approximately one micrometer long and
appear to be telocentric. The normal chromosome complement consists
of submetacentrics that are 2.5—5.0 »m long at mitotic metaphase.

The tetraploids characteristically form multivalents at meiosis and
are morphologically indistinguishable from: diploids of the respective

1978] WATSON: XYLORHIZA 207

a2

& 4

Fics. 1-10. Camera lucida drawings of meiotic metaphase chromosomes of X ylor-
hiza spp. and Aster kingii. All collections are those of T. J. Watson. 1. X. glabrius-
cula, 2n = 6,,; + synapsed fragments, 470. 2. X. glabriuscula var. linearifolia, 2n =
63; 908. 3. X. confertifolia, 2n = 6y,, 697. 4. X. venusta, 2n = 64, 669. 5. X. torti-
folia, 2n = 6), 372. 6. X. tortifolia var. imberbis, 2n = 61, 911.7. X. wrighit, 2n =
6;;, 620. 8. X. cognata, 2n = 6,,, 365.9. X. orcuttii, 2n = 64, 603. 10. Aster kingit,
2n = 94, 766.

9 [-10»m—|

taxa, suggesting that the plants are autotetraploids. The tetraploids of
X. tortifolia are located at the northern distribution limits in Nevada.
Tetraploids of X. venusta are found at the southwestern margin of the
range in Utah and Colorado. Diploids and tetraploids of X. glabriuscula
var. linearifolia grow intermixed over the small range of the taxon in
western Utah.

The taxonomic status and placement of Xylorhiza have varied (for a
complete taxonomic history see Watson, 1977). Recent investigators
(Cronquist and Keck, 1957; Turner and Horne, 1964) feel that X ylo-
rhiza Nutt., Machaeranthera Nees (sensu stricto), and Haplopappus
Cass. section Blepharodon DC. are closely allied. In their view, Machaer-
anthera series Originales Cronq. & Keck is a pivotal, infrasectional taxon
through which the three groups are related. Cronquist and Keck (1957)
consider Xvlorhiza and the remainder of Machaeranthera to have been
derived from an Originales-like ancestry and construe M. blephariphylla
of this series to be the most primitive extant taxon in the X ylorhiza—
Machaeranthera alliance. Chromosome numbers of the taxa involved
seem instructive in these regards.

208 MADRONO [Vol. 25

Most species of Originales and those species of Blepharodon related to
Originales are reported to be on a base of x = 4. Although 2” = 10 has
been reported from M. blephariphylla (Jackson, 1959; as M. gymno-
cephala), Hartman (1976) feels that the count was erroneous. He has
recorded 2m = 8 from three populations of this species and has found
plants from one of the populations to have 2 or 3 pairs of small, super-
numerary chromosomes in addition to the normal complement of four
pairs. Thus, Hartman regards Orviginales and related members of Ble-
pharodon to be unibasic with x = 4. If Xylorhiza (x = 6) has evolved
from an M. blephariphylla-like ancestor or from another extinct or ex-
tant member of Originales, the chromosome number of the former is a
result of an aneuploid gain. However, it is noteworthy that plants serv-
ing to link Xylorhiza to Originales by morphology and phenology and
having a documented base number of « = 5 are unknown. Species on a
base of x = 5 are found in Machaecranthera but these taxa belong to
other subgeneric groups (i.e., series Variables and series Verae of Cron-
quist and Keck, and section Psilactis Turner and Horne) that cannot
be related directly to Xylorhiza.

Some investigators (Raven et al., 1960; Solbrig et al, 1969) hold that
the primitive base number for the Astereae as a whole is x = 9 and that
the lower chromosome numbers in the tribe were generally derived
through aneuploid reduction (for a contrasting viewpoint see Turner et
al., 1961). According to this hypothesis, the chromosome level at which
X ylorhiza diverged would precede that of Machaeranthcra. Also, Solbrig
et al. (1969), Anderson et al. (1974), and Hartman (1976) have noted
the frequent occurrence of « = 6 in taxa that have been included in or
bear relationship to Haplopappus and Machaeranthera (e.g., Grindelia,
Prionopsis, Xanthocephalum, Isopappus, Pvrrocoma, Isocoma, Hazardia,
Xylorhiza and the ‘‘phyllocephalus group” sensu Hartman, 1976).
This observation led Hartman (1976) to suggest that x = 6 is a more
primitive number for this alliance and that the lower base numbers (i.e.,
« = 4,5) are derived. It is interesting to note that one of the few docu-
mented cases of descending aneuploidy in natural populations is known
from Haplopappus sect. Blepharodon (i.e., Haplopappus gracilis—H.
raveni,; Jackson, 1962; 1965).

The foregoing observations suggest that X ylorhiza diverged early from
the line that gave rise to Machaeranthera and Haplopappus in North
America. However, before any credible phylogenetic interpretations can
be made, it appears to me that the relationships of the North American
Astereae to the poorly known Haplopappus sect. Haplopappus (sect.
Euhaplopappus of Hall, 1928) of South America need to be explored.
The latter taxon seemingly connects various elements in the X vlorhiza—
Machaeranthera—Ha plopappus alliance of North America (see Watson,
1977). Chromosome numbers for only ten of the South American species
are known: most have 2m = 10 (Grau, 1976; L. C. Anderson. personal

1978] WATSON: XYLORHIZA 209

communication), but H. cuneifolius has 2n = 12 (B. L. Turner and J.
Bacon, personal communication).

The chromosome number of Aster kingit (2n = 18) is reported here
for the first time. This taxon was included in Machaeranthera sect. Xy-
lorhiza by Cronquist and Keck (1957). However, it is phenologically,
ecologically, morphologically, and chromosomally anomalous there. The
plants of A. kimgi flower in mid-summer and are found in coniferous
forests in cracks of granitic outcrops at subalpine elevations in the
Wasatch Mountains of Utah. Members of Xylorhiza flower in early
spring and are distributed in relatively deep soils of deserts and semi-
arid grasslands. Although plants of A. kingi have taproots surmounted
by a caudex, the roots are small and resemble those of the alpine Asters,
e.g., A. alpigenus (T. & G.) Gray. Individuals of A. kingi are cespitose
and have relatively small capitula with phyllaries that have anthocyanic
margins and squarrose tips; the disc florets are anthocyanic. None of
these features is found in species of X ylorhiza.

The presence of taproots and squarrose phyllaries in A. kingi sug-
gests Machaeranthera. However, with the exception of M. brevilingulata
(2m = 18; Turner and Horne, 1964; Powell and King, 1969), which is
better placed in Aster or Conyza (Hartman, 1976), Machaeranthera
consists of diploids with 2m = 8 or 10 and tetraploids with 27 = 16 (see
Hartman, 1976).

Aster kingii is probably most closely allied with species of Aster in
which 2m = 18 is a common number (Raven et al., 1960; Solbrig et al.,
1964; Solbrig et al., 1969; Anderson et al., 1974; and others). Some
members of Aster have taproots (e.g., A. alpigenus) and others have
squarrose phyllaries (e.g., A. conspicuus Lindl.). The florets and capit-
ula of A. kingi resemble those of the more widespread 4. integrifolius
Nutt., although the latter lacks squarrose phyllaries and differs in habit.
This similarity was noticed by Gray (1884), who treated the two species
together within Aster proper.

ACKNOWLEDGMENTS
This study is a portion of a dissertation submitted in partial fulfillment of the
requirements for the degree of Doctor of Philosophy at the University of Texas,
Austin. I am grateful to Drs. A. M. Powell and R. L. Hartman for evaluating the
manuscript and for encouragement during the study.

LITERATURE CITED

Anpverson, L. C., D. W. Kuyos, T. Mosguin, A. M. Powe Lt, and P. H. RAveEn.
1974. Chromoscme numbers in Compositae. IX. Haplopappus and other Aster-
eae. Amer. J. Bot. 61:665-671.

Cronouist, A. and D. D. Keck, 1957. A reconstruction of the genus Machaeranthera.
Brittonia 9:231-239.

Grav, J. 1976. Chromosomenzahlen von Siidamerikanischen Haplopappus. Mitt. Bot.
Staatssamml. Miinchen 12:403-409.

Gray, A. 1884. Synoptical Flora of North America. Vol. 1, part 2. Smithsonian Inst.,
Wash., D. C.

TiO MADRONO [ Vol. 25

Hatt, H. M. 1928. The genus Haplopappus, a phylogenetic study in the Compositae.
Publ. Carnegie Inst. Wash. 389.

Hartman, R. L. 1976. A conspectus of Machaeranthera (Compositae: Astereae) and
a biosystematic study of the section Blepharodon. Ph.D. Dissertation, Univ.
Texas, Austin.

Huziwara, Y. 1957. Karyotype analysis in some genera of Compositae II. The kary-
otype of Japanese Aster species. Cytologia 22:96—112.

Jackson, R. C. 1959. In Documented chromosome numbers of plants. Madronio
L252.

. 1962. Interspecific hybridization in Haplopappus and its bearing on chro-
mosome evolution in the Blepharodon section. Amer. J. Bot. 49:119-132.

. 1965. A cytogenetic study of a three-paired race of Haplopappus gracilis.
Amer. J. Bot. 52:946-953.

Powe 1, A. M. and R. M. Krnc. 1969. Chromosome numbers in the Compositae:
Columbian species. Amer. J. Bot. 56:116-121.

and S. SrkeEs. 1970. Chromosome numbers of some Chihuahuan Desert
Compositae. Southw. Naturalist 15:175—-186.

Raven, P. H., O. T. Sorsric, D. W. Kyuos, and R. Snow. 1960. Chromosome num-
bers in Compositae. I. Astereae. Amer. J. Bot. 47:124-132.

SHinneErs, L. H. 1950. Notes on Texas Compositae V. Field and Lab. 18:32-42.

Sorsric, O. T., L. C. ANpErson, D. W. Kyuos, and P. H. RAven. 1969. Chromosome
numbers in Compositae VII: Astereae III. Amer. J. Bot. 56:348-353.

—, and L. RUDENBERG. 1964. Chromosome numbers in Compositae V. Astereae
II. Amer. J. Bot. 51:513-519.

TurRNER, B. L. 1964. Karyotype of Xylorhiza wrightii Gray (Compositae). Southw.
Naturalist 9:315.

, W. L. Exrison, and R. M. Kinc. 1961. Chromosome numbers in the Com-
positae. IV. North American species with phyletic interpretations. Amer J. Bot.
48:216-223.

and D. Horne. 1964. Taxonomy on Machaeranthera sect. Psilactis (Com-
positae-Astereae). Brittonia 16:316-331.

and M. C. Jounston. 1961. Chromosome numbers in the Compositae. III.
Certain Mexican species. Brittonia 13:64-69.

Urgpatscu, L. E. 1974. In IOPB chromosome number reports XLV. Taxon 23:624.

Watson, T. J. 1973. Chromosome numbers in Compositae from the southwestern
United States. Southw. Naturalist 18:117-124.

. 1977. The taxonomy of Xylorhiza (Asteraceae-Astereae). Brittonia
29:199-216.

INTERSPECIFIC HYBRIDIZATION BETWEEN NATIVE
AND NATURALIZED CRATAEGUS (ROSACEAE)
IN WESTERN OREGON

Ruopa Love and Marc FEIGEN
Department of Biology, University of Oregon, Eugene 97403

ABSTRACT

Morphological evidence for hybridization between Crataegus douglasi var. suks-
dorfii and C. monogyna in the southern Willamette Valley, Oregon, has been derived
from a population with intermediate leaf morphology. Experimental hybridization
resulted in an average fruit set of 42 percent in 2 C. douglasti X & C. monogyna,
and 7 percent in 2 C. monogyna X 6 C. douglasii crosses. Hybrid pollen was 66
percent stainable with aniline blue-lactophenol compared with 95 percent for C.
douglasit and 96 percent for C. monogyna. This is the first documented hybridiza-
tion between a European and a northwestern North American species.

Hawthorns (Crataegus, Rosaceae) are known to hybridize where spe-
cies are sympatric (Standish, 1916; Bradshaw, 1954; Robertson, 1974;
Byatt, 1975, 1976). None of the documented cases of hawthorn hybridi-
zation, however, has involved a species native to North America and a
European species. In western Oregon, the native hawthorn, Crataegus
douglasu Lindl. var. suksdorfi Sarg., and an introduced European spe-
cies, Crataegus monogyna Jacq., have come together within the last 100
years. The native species has black fruit, five styles, and mostly unlobed
leaves; the introduced species has red fruit, a single style, and deeply
lobed or laciniate leaves. Hybridization has produced a population of
intermediate plants with black fruits, variable style number, and a
wide range of leaf shapes. The evidence presented here for the hybrid
origin of these intermediate individuals is based on leaf morphology, the
results of crossing experiments, and pollen stainability tests. All infor-
mation gathered thus far supports the hybridization hypothesis.

The study was conducted on the Cogswell-Foster Reserve, a 36-ha
tract 40 km north of Eugene in Linn County, Oregon. The Reserve was
acquired by the Nature Conservancy in 1969. Crataegus douglasi, the
black hawthorn, is native to the Reserve and is found throughout the
Pacific Northwest, especially along streams, ditches, and valley bottoms.
English hawthorn, C. monogyna, was introduced onto the Reserve about
100 years ago (Lucile Foster, pers. comm., 1976). It has spread vigor-
ously throughout the area, forming dense thickets under canopies of
Quercus garryana Dougl. Intermediate individuals can be found through-
out the Reserve but are especially common along fence lines in open
areas, where many individuals are large annd produce abundant flowers
and fruit. Increment cores suggest that most of these trees are less than
20 years old.

In May, both species and their putative hybrid bloom simultaneously,
C. monogyna having the most abundant flowers. Hymenopterans and

Zu

212 MADRONO [Vol..25

A = Leaf margin to base of lowest sinus
B = Half leaf width

C = Lowest leaf lobe width

D = Leaf length

es = Midvein of lowest leaf lobe perpendicular
to "C"

Fic. 1. Leaf measurements used in the scattergram (Fig. 2).

dipterans are attracted in large numbers. Honey bees (A pis mellifera L.)
move among all three types on a single collecting flight. Frugivorous
birds are known to be the principal agents of hawthorn seed dispersal
(Hitchcock et al., 1969; Robertson, 1974) and are commonly seen de-
vouring hawthorn fruits on the Reserve.

MATERIALS AND METHODS

Leaf Variability. Ten to 15 leaves from 114 randomly selected plants
were collected in May and October, 1977. Four measurements were made
on each leaf (Fig. 1).

Artificial Crosses. Hand crosses were made on May 1-13, 1976, and
May 11-19, 1977. Crossing was done on calm, clear days, during mid-
mornings and early afternoons. A corymb of 1—16 flowers was considered
the “crossing unit” for each hybridization. Eighteen C. douglasi X* C.
monogyna crosses were made involving 203 flowers on 18 different plants.
The pistillate parent was C. douglas for ten corymbs and C. monogyna
for eight corymbs. Hawthorn stigmas become receptive about two days
before the petals open. Pollen is shed at anthesis. Therefore, flowers of
the pistillate parent were chosen in the ‘‘popcorn” stage, just before bud
opening. Stamens were removed with fine forceps, and pollen was trans-
ferred directly from the flowers of the male parent to the stigmas of the
emasculated flowers. Corymbs were then bagged with cheesecloth and
tied with string. Bags were opened in July to check fruit set and then
reclosed to allow fruit ripening. All of the resulting fruits were saved for
testing of seed viability.

Nine corymbs were bagged before bud opening to test for the necessity
of pollen vectors. On eight plants, self-crosses were made to test for
self-incompatibility, which has been reported in some Rosaceae (East,
1940),

Pollen Viability Tests. Pollen from 22 pressed hawthorn specimens
was examined using aniline blue-lactophenol as an indicator (see Byatt,
1977). Percentages were based on microscopic examination of an aver-

1978] LOVE & FEIGAN: CRATAEGUS 213

[) eno data on style number :
YU jeone style '
= 4|¥two styl
D3 Oo styles cn .
¥three styles 4
yfive styles oe ,
@@@ position of leaves Ue caves 4
d f 3 oe Wee © meee e
ictured in Figure :
2 ica is ee : M 4 @ oon 4
ede 4
e oe

°
—

Lowest Leaf-lobe Width/Leaf Len

| 2 3 4 DigOu, 7.
Leaf Margin to Lowest Lateral Sinus/Half Leaf Width (A/B)

8 9 10

Fic. 2. Scattergram illustrating range of hawthorn leaf variability at the Cogswell-
Foster Reserve. Each point represents the mean of leaves measured from an individ-
ual plant. One leaf each of C. douglasit (X), C. monogyna (Z), and an intermediate
type (Y) from Fig. 3 are plotted to help indicate the relationship of leaf morphology
to position on the scattergram.

age of 250 grains per plant. Grains that stained dark blue were assumed
to be viable and unstained grains to be inviable.

RESULTS

Leaf Variability. Mean relative distance from the leaf margin to the
base of the lowest sinus (A/B) was plotted against the mean ratio of the
lowest leaf lobe width to leaf length (C/D) for each plant (Fig. 2).
Distribution of points on the scattergram suggests the presence of a
“swarm” of hybrid types that overlap the parental types in leaf mor-
phology. Crataegus douglasii leaves are unlobed or shallowly lobed; C.
monogyna leaves are usually deeply lobed or laciniate; while leaves of
intermediate plants show wide morphological variation (Fig. 3).

Artificial Crosses. There is cross compatibility between C. douglasii
and C. monogyna as judged by the high percentage of fruit set in 2 C.
douglasii * 4 C. monogyna crosses. Fruit formation on ten corymbs
ranged from 25 to 73 percent of treated flowers, with mean fruit set at
42 percent. The mean fruit set for C. douglasi corymbs left open for
insect pollination was 29 percent.

Fruit set was much reduced in 2 C. monogyna < ¢ C. douglasii
crosses. Five out of eight corymbs set no fruit. The highest fruit set on

214 MADRONO [Vol. 25

C. douglasii

pees.

C:18

C. douglasii X C. monogyna

rm i a

Nor! 10 ~ C-72

C. monogyna

“ Vv &£

\ oA

C-63 C:28 C38

Fic. 3. Representative hawthorn leaf types from the Willamette Valley, Oregon.
Letters and numbers refer to individual plants. Leaves X, Y, and Z have been
plotted on the scattergram (Fig. 2).

1978 | LOVE & FEIGAN: CRATAEGUS Z1>

one corymb was 25 percent and mean fruit set was 7 percent. Mean fruit
set for C. monogyna corymbs left open to insects was 50 percent. The
low fruit set is not thought to be an artifact of the experimental handling
of C. monogyna flowers, because in other crosses, C. monogyna set sig-
nificant amounts of fruit.

There was no fruit set on any corymb bagged before bud opening
nor on any selfed corymb.

Pollen Viability Tests. Pollen from presumed hybrids showed signifi-
cantly lower percentages of stainable grains than parental pollen. For
pollen from nine individuals with hybrid morphology, the percentage of
stainable grains ranged from 42 to 75 percent (mean = 66 percent).
Stainable pollen from five specimens of C. douglas ranged from 93 to
98 percent (mean = 95 percent); and from eight specimens of C.
monogyna, stainability ranged from 93 to 98 percent (mean = 96 per-
cent).

At least one hybrid individual at the Cogswell-Foster Reserve is com-
pletely male-sterile, all flowers having vestigial stamens that lack an-
thers. This plant is vegetatively vigorous, flowers heavily, and sets some
fruit.

A comparison of the Willamette Valley hawthorns is given as Table 1.

DISCUSSION AND CONCLUSIONS

Evidence that the native Crataegus douglasii var. suksdorfi hybrid-
izes with the introduced Crataegus monogyna in western Oregon can be
summarized as follows:

1. There are many plants intermediate in leaf shape and style number
and possessing a novel combination of parental characters: distinctly
lobed leaves and black fruits.

2. Interspecific crosses resulted in substantial fruit set when C. doug-
lassii was the pistillate parent. Partial unilateral sterility appears to be
occurring because reciprocal crosses resulted in reduced fruit set. Insect
vectors appear necessary for pollination, and all plants tested exhibited
self-sterility.

3. Pollen stainability tests indicated significantly lower percentages of
viable pollen grains in the putative hybrids than in the parents.

Most of the C. douglastt < C. monogyna hybrids on the Cogswell-
Foster Reserve have 3-5 lobed leaves and 2-3 styles and, when in
bloom, match the key description of Crataegus oxyacantha L., another
hawthorn of European origin that is naturalized in the Pacific North-
west. |C. oxyvacantha is now known in Europe as C. laevigata (Poiret)
DC. See Byatt, 1974.| Hawthorn samples from the Reserve were sent
to J. Byatt, Westfield College, London, whose determinations support the
hybrid nature of the intermediate plants and confirm that they are not
C. oxvacantha. Dr. Byatt reports (pers. comm., 1977) that C. oxyacan-
tha is never black-fruited. She also writes, “It has already been sug-

216

MADRONO

[Vol. 25

TABLE 1. SOME DIAGNosTIC CHARACTERS OF Crataegus douglasii vAR. suksdorfii, C.
monogynda, AND THEIR HYBRIDS FROM THE WILLAMETTE VALLEY, OREGON.

Character

Style and

C. douglasii Hybrids C. monogyna
Petal color White White White or pink
5 (1) 2-3 (4) 1
pyrene number
Receptacle Glabrous Glabrous to hairy Mostly hairy
to woolly
Mature fruit color Black Black (Imm.: Red
purple or red)
Fruit shape Globose Globose Ovoid
Mature leaf shape Elliptic Elliptic to obovate Ovate or obovate
Unlobed or Variously lobed; Deeply lobed

Leaf lobing

Leaf length

Leaf pubescence

Leaf margins

Lowest lateral
leaf veins

Termination of
lateral veins

Fruit ripens

Chromosome number

Geographic range

shallowly lobed

2-9 cm

Both surfaces
pubescent to
glabrate

Serrate or biserrate
Straight
At tooth apices

Jul-Aug

2n = 34 (Calder
et al., 1968)

B. C.t0' S OR,
W of Cascades

sinuses of
intermediate depth

1.5-6 cm

Glabrous to
somewhat hairy,
or with some hairs
on veins below

Serrate or toothed
Straight or
slightly recurved
Variable

Aug—Sep

Willamette Valley,
OR;; possibly
elsewhere

or laciniate

15-35 cm
Glabrous except for
patches of hairs in
axils of veins
beneath

Entire or sparingly
serrate

Strongly recurved

At apices and
sinuses

Sep-Oct
2n = 34 (Clapham
et al., 1962)

Eurasian native;
naturalized
sparingly but

widely in N. A.

gested that black fruit colour is dominant in crosses between red and
black-fruited taxa.” This agrees with our observation that all presumed
hybrids on the Cogswell-Foster Reserve are black-fruited.

Hitchcock et al. (1969) list both C. monogyna and C. oxyacantha as
naturalized elements of our flora, distinguishing between them on the
basis of leaf lobing and style number. Both are described as red-fruited.
Crataegus douglas < C. monogyna hybrids and C. oxyacantha may be
distinguished in future treatments on the basis of fruit color.

Crataegus monogyna may be hybridizing with another North Ameri-
can species. J. B. Phipps (pers. comm., 1977) has noted probable hybrid-
ization between C. monogyna and the native C. punctata Jacq., in On-
tario, Canada.

1978 } LOVE & FEIGAN: CRATAEGUS 217

In this study we describe what we believe to be the first documented
case of hybridization between a western North American species and a
European native. At the present time, C. douglas X C. monogyna hy-
brids are known only from Linn County in the Willamette Valley; how-
ever, they can be expected to occur in other locations in western British
Columbia, Washington, and Oregon, where the parent species have co-
existed for some time.

ACKNOWLEDGMENTS
We thank Stan Cook, David Wagner, and Kenton Chambers for their support,
reading of the manuscript, and helpful suggestions. We also thank Debra Ayres and
Stan Love for their help in the field. Supported in part by NSF Grant DEB-7709472.

LITERATURE CITED
BrapsHAw, A. D. 1954. Man’s influence on hybridization in Crataegus. VIII Cong.
Int. de Botanique. 9/10:217.
Byatt, J. I. 1974. Application of the names Crataegus calycina Peterm, and C. oxya-
cantha L. J. Linn. Soc., Bot. 69:15-21.

. 1975. Hybridization between Crataegus monogyna Jacq. and C. laevigata
(Poiret) DC. in south-eastern England. Watsonia 10:253-264.

. 1976. The structure of some Crataegus populations in north-eastern France
and south-eastern Belgium. Watsonia 11:105~-115.

, I. K. Ferguson, and B. G. Murray. 1977. Intergeneric hybrids between
Crataegus L. and Mespilus L.: a fresh look at an old problem. J. Linn. Soc.,
Bot. 74:329-343.

Caper, J. A., R. L. TAytor, and G. A. MuLtican. 1968. Flora of the Queen Char-
lotte Islands. Canad. Dept. of Agriculture Monograph 4. Vol. IT.

CrapHaAM, A. R., T. G. TuTiIn and E. F. Warpurc. 1962. Flora of the British Isles.
2nd ed. Cambridge University Press, Cambridge, England.

East, E. M. 1940. The distribution of self-sterility in the flowering plants. Proc.
Amer. Philos. Soc. 82:449-518.

Hitcucock, C. L., A. Cronguist, M. OwnsBeEy, and J. W. THompson. 1969. Vascu-
lar plants of the Pacific Northwest. University of Washington Press, Seattle.
RoBertson, K. R. 1974. Genera of the Rosaceae in southeastern U. S. J. Arnold

Arbor. 55:626-633.
STANDISH, L. M. 1916. What is happening to the hawthorns? J. Heredity. 7:266-279.

A NEW SPECIES OF EUPATORIUM (ASTERACEAE)
FROM CALIFORNIA

DEAN WM. TAYLOR
Department of Botany, University of California, Davis 95616
G. LEDYARD STEBBINS
Department of Genetics, University of California, Davis 95616

Eupatorium (Asteraceae, Eupatorieae) is a highly diverse and wide-
spread group of well over 1000 taxa. Numerous species are found in the
American tropics, and the genus is well represented in the floras of more
northerly areas of the globe, including North America. However, Califor-
nia is relatively depauperate in members of this large and complex genus,
with only two native species recorded by Munz (1959). Consequently,
the existence of a new, highly distinct, and narrowly endemic member
of the genus in the state is of considerable interest.

Eupatorium shastense Taylor & Stebbins, sp. nov.; a E. occidentale
differt foliis basi opposita, folia caulina alterna. Capitulum plerumque
solitarium, terminale. Flores albae. (Fig. 1)

Perennial herb from woody, enlarged or occasionally rhizomatous base.
Stems clustered, 1.5—4.5 dm long, puberulent to pubescent, often with
some glandular hairs above. Leaves opposite at base, alternate above;
the juvenile orbicular, entire to slightly dentate; the adult ovate, obtuse
to truncate at base, acute to accuminate at tip, coarsely serrodentate,
often ciliate on margins with short hairs 0.3-0.9 mm long, glabrous to
puberulent on laminar surfaces. Petioles 4-6 mm long, blades 15-31 mm
long, 11-20 mm wide. Capitula mostly solitary (1-3) at ends of
branches, often subtended by a small leaf-like bract, 12-16 mm diameter
when pressed. Phyllaries 9-13, green, often ribbed at base, in two series,
glabrous to pubescent. Phyllaries glabrous to pubescent, 9-14 mm long,
1.1-2.5 mm wide. Flowers all tubular, 30-60 per capitulum. Corollas 5-8
mm long, white. Style branches clavate, elongate, 3-5 mm long, with
short stigmatic lines; appendage elongate, papillate. Achenes brown to
black, puberulent to pubescent, 3.0-5.5 mm long, 5-nerved. Pappus of
20—40 whitish barbellate bristles. Chromosome number 7 = 17.

Type: CA, Shasta Co., 1.66 km E of Squaw Creek, T35N R2W
(MDM), 762 m. Stebbins & Ehrendorfer 5968, 20 Jun 1959, Holotype
(UC). Isotypes: MO, NY, GH, DAV.

Additional specimens examined: CA, Shasta Co., Shasta Lake, Mc-
Cloud Arm opposite Bailey Cove, 487 m, on North Gray Rocks, Stebbins
and Gajewski 5949, 18 Oct 1959 (UC, DAV); Devil’s Rock, $28 T35N
R2W, 670 m, along Low Pass Creek, Stebbins 6177, 9 Sep 1967 (DAV);
N face Hirz Mt., S7 T35N R3W, 1066 m, Taylor 2409, 2410, 1 Jul 1972,
chromosome voucher 2430, 1 Jul 1973 (DAV).

218

1978] TAYLOR & STEBBINS: EUPATORIUM 219

rly Way Z
> Ms
ve 3 @ WE

Q, % 4 :
NZ
= i :
>

Fic. 1. Eupatorium shastense. Upper left: habit of plant. Upper right: detail of
corolla. Lower left: camera lucida drawing of chrosome voucher * 3100. Lower
right: detail of single capitulum.

Eupatorium shastense is obviously related to the more widespread E.
occidentale Hook., because these two taxa are similar in a number of
vegetative characters. Both have numerous clustered stems that arise
from an enlarged woody base and have very similar foliage and canopy
architecture. Heteroblastic leaf development is typical of nearly all indi-
viduals of E. shastense observed, but this character is either absent or of
rare occurrence for E. occidentale. The most striking differences separat-
ing these two species are characters of the capitulum. Eupatorium shas-
tense has a solitary, terminal capitulum (occasional individuals possess a
single subtending pair), whereas EF. occidentale typically has >18 capit-
ula in a terminal corymbose cluster. The capitula of E. shastense are
larger in most dimensional characters than those of E. occidentale, and
the corollas of E. shastense are invariably white whereas the corollas of
all E. occidentale we have observed are tinged with purple. Eupatorium
shastense differs significantly at the 0.05 level of probability (based on ¢
statistic comparisons) from E. occidentale for the following characters:

220 MADRONO [Vol. 25

capitula per branch, capitulum width when pressed, and corolla width
when pressed. The large, solitary heads of E. shastense are somewhat
reminiscent of several species of Brickellia, but the 5-nerved achenes and
chromosome number of m = 17 of the former show that this resemblance
is superficial.

Eupatorium shastense is a strict calcicole, being restricted to nearly
vertical limestone cliffs of the Hosselkus formation from 450 to 1200 m
between the Pit and McCloud river drainages in Shasta County. Plants
of E. shastense establish on these cliffs wherever there is sufficient soil or
organic matter lodged in cracks or on ledges to support their growth.
Eupatorium occidentale and E. shastense are sympatric at all popula-
tions of the latter known to us. Ecologically, the two taxa occupy differ-
ent habitats at a given site. Eupatorium occidentale occurs as an un-
derstory herb in rocky sites in a dense woodland dominated locally by
Pseudotsuga menziesii, Pinus ponderosa, Quercus chrysolepis, and Q.
garryana var. brewert. Plants of E. occidentale are absent from the verti-
cal cliffs that are occupied by E. shastense. Elsewhere, E. occidentale is
often a plant of cliff-faces, but it is not a strict calcicole. At all popula-
tions studied, no individuals of E. shastense have been found in the seem-
ingly more favorable habitats in the surrounding woodland occupied by
E. occidentale.

The Hirz Mountain population of E. skhastense was surveyed during
the summers of 1972-1975. The population during this period was stable
and consisted of less than 100 individuals. Other known populations of
the new species are similarly small. Individuals in small size-classes were
observed in sufficient numbers on Hirz Mountain to suggest a stable
population size.

ACKNOWLEDGMENTS
We thank Gerald D. Carr for the chromosome number determination.

LITERATURE CITED
Mowz, P. A. 1959. A California flora. Univ. California Press, Berkeley.

A NEW SPECIES OF VIGUIERA
(ASTERACEAE-HELIANTHEAE) FROM NAYARIT, MEXICO

B. L. TURNER
Department of Botany, University of Texas, Austin 78712

Viguiera is a notably complex genus with relationships extending to
such well-known genera or generic segregates as Helianthus, Tithoma,
and Hymenostephium.

Viguiera websteri, described below, belongs to a group of species that
Blake (1918) recognized as a distinct genus, Hymenostephium. D’Arcy
(1975) sunk the latter into synonymy with Vigwiera; this was also ac-
cepted by H. Robinson (1977). Neither author went so far as to include
Hymenostephium within sect. Diplostichis of Viguiera, but I believe that
its relationship is in, or near, this group and consequently concur with
their generic disposition.

Detection of the relationship of Hymenostephium to Diplostichis (and
consequently Viguiera) by D’Arcy and H. Robinson is not surprising, for
Blake, himself, (1918, p. 7) notes, “They [Hymenostephium| are dis-
stinguished from the section Diplostichis of Viguiera, which they closely
resemble in all other features, solely by their pappus .. .”. The same may
be said for the genus Haplocalymma Blake, which H. Robinson (1977)
reduced to synonymy under Viguiera, a submersion to which I also sub-
scribe, but again, this is not unexpected since Blake (p. 8) went on to
state that Haplocalymma “. . . is clearly a lateral offshoot of the Dzplo-
stichis-H ymenoste phium line,...”.

And that is one of the problems in dealing with Blake’s otherwise very
scholarly treatment of the Compositae: he tended to let the absence of
one, or perhaps two, characters make a genus. As noted by Cronquist
(1968, p. 10) ‘‘. . . the absence of a character is a less reliable guide to
taxonomic affinities than its presence”. In the case of the pappus, very
simple but loose, genetic control of its absence has been amply demon-
strated by Clausen (1951) and many others.

Viguiera websteri B. L. Turner, sp. nov. V. hintonit H. Robinson
simulans sed capitulis pluribus majoribus, pedunculis longioribus, paleis
receptaculi majoribus, plantis parvis erectis perennis caulibus comparata
tenuibus, rhizomatibus ligeis cormoideis. (Fig. 1)

Perennial herb 30-65 cm tall, the stems slender, sparsely appressed
pubescent to nearly glabrate, arising from woody, corm-like rootstocks,
ca 20 mm long, 15—20 mm thick. Leaves opposite throughout, except for
the several much-reduced leaves that subtend each of the flowering pe-
duncles; petioles short, 3-8 mm long; blades ovate, 3-6 cm long, 1.5—2.5
cm wide, sparsely appressed-pubescent above and below, the margins
crenate-serrate. Inflorescence loose, the heads remote, mostly (2)3—5 per

221

222 MADRONO {Vol. 25

Fic. 1. Sketch of isotype (LL) of Viguiera websteri Turner: left, habit; upper
right, involucre; middle right, individual head; lower right, partial head showing
florets.

1978] TURNER: VIGUIERA 223

secondary branch, the ultimate peduncles (1)2—5 cm long. Involucre
campanulate, 2-seriate; bracts, lanceolate-ovate, 2.0-3.5 mm long, ca 1
mm wide, moderately appressed-pubescent, 2—3 striate. Receptacle con-
vex, knobby; bracts linear-ovate, acute or apiculate, ca twice as long as
the involucre. Ray florets 8, yellow, neuter; ligule ca 5 mm long, 2—3 mm
wide, the tube ca 0.8 mm long. Disk florets 20-30, yellow; corolla ca 4
mm long, the tube sparsely pubescent, ca 1 mm long; the lobes 5, ca 0.6
mm long; stylar appendages caudate, pubescent beneath, especially at
the apex. Achenes black, sparsely appressed pubescent, somewhat tetra-
gonally-flattened, ca 2 mm long, 1 mm wide; pappus absent.

Type: Mexico. Nayarit; “oak woods on volcanic rock 25 km by road
S of Tepic” (between Tepic and Compestela), ca 1000 m, 18 Oct 1970,
G. L. Webster and G. J. Breckon 15744. (Holotype DAV; isotype, LL).

The closest relationship of Viguzera websteri is probably with the re-
cently described V. Aintonii H. Robinson (1977) from Michoacan and
Guerrero. It has the foliage and achene characters of this taxon but its
habit is markedly different (low perennial from corm-like rootstocks vs.
shrub 1—2 meters tall). Further, the heads of V. webster are much larger
(6-7 mm vs. 3-4 mm) with more numerous, larger florets.

The typification of Viguiera hintonii (McVaugh 22637, US) is unfor-
tunate since H. Robinson has apparently described a plant whose inflo-
rescence (to judge from the illustration accompanying his description)
has been badly affected by insect egg deposition and larval development
among heads, so that the peduncles of the capitula are described as 1-10
mm long. This range holds for isotypic material at TEX, but the heads
are also badly infested by insects. Paratypic material (Hzuton et al.
14182, LL), however, is relatively free of insects, possessing peduncles
up to 30 mm long.

The section Diplostichis (including Hymenostephium) is in much
need of detailed study. There is a perplexing array of variation in the
group as noted by H. Robinson, especially along the Pacific Mountain
slopes from Guatemala to Durango. Field work should do much to help
unravel the complex and I suspect that several additional undescribed
species will come to light in the process.

LITERATURE CITED
BLAKE, S. F. 1918. A revision of the genus Viguwiera. Contr. Gray Herb. 54:1-205.
CLAUSEN, J. 1951. Evolution of plant species. Cornell Univ. Press, Ithaca.
Cronguist, A. 1968. The evolution and classification of flowering plants. Houghton
Mifflin Co., Boston.
D’Arcy, W. G. 1975. Helianthinae Jn Flora of Panama. Ann. Missouri Bot. Gard.
62:1101-1173.
Rosinson, H. 1977. Studies on the Heliantheae (Asteraceae). VIII. Notes on genus
and species limits in the genus V7guzera. Phytologia 36:201-213.

A NEW SUBSPECIES OF ABRONIA MARITIMA
FROM BAJA CALIFORNIA, MEXICO

ANN F. JOHNSON
Department of Botany, University of California, Davis 95616

In the course of a survey of vegetation along the coasts of Baja Cali-
fornia, Sonora, and Sinaloa, Mexico, I noted a pink-flowered A bronia
maritima (Nyctaginaceae). It is presumed to be of hybrid origin.

Abronia maritima Nutt. ex Wats. ssp. capensis A. F. Johnson, ssp.
nov. differt a A. maritima ssp. maritima foliis persucculentis, aggregatis,
ovalibus v. subrotundis, crenatis, 1-3 cm longis, 0.8—1.5 cm latis, 3-5 mm
crassis; perianthio bracteis 1.0—-1.5plo longiore, tubo perianthii viridi-
albido, 7-9 mm longo, limbo perianthii roseo pallido centro albido et
lobis non reflexis; anthocarpo glanduloso-puberulo omnino.

Abronia maritima Nutt. ex Wats. ssp. capensis A. F. Johnson differs
from A. maritima ssp. maritima in that the leaves are very succulent,
crowded, oval to suborbicular with crenate margins, 1-3 cm long, 0.8—1.5
cm wide, and 3—5 mm thick, the perianth is 1-1.5 times longer than the
bracts, the tube greenish-white, 7-9 mm long, the limb pale pink with
white center and unreflexed lobes, anthocarp glandular-puberulent
throughout.

Type: Mexico, Baja California Sur, beach south of lighthouse at Cabo
Falso NW of town of San Lucas, 18 Mar 1974, A. F. Johnson 711 (holo-
type, DAV). Paratypes: San Pedrito, B. Cfa. Sur, Dec 1972, A. F. John-
son 514 (UC); Migrino, B. Cfa. Sur, Dec 1972, A. F. Johnson 516
(CAS).

A. maritima capensis differs from A. maritima maritima primarily in
perianth color and in having smaller, more crowded leaves (Fig. 1). Al-
though obvious in the field, these characteristics tend to be lost in the
preparation of herbarium specimens of this succulent species.

I have found subspecies capensis only along the Pacific shore of the
Cape region of Baja California, Mexico, between Todos Santos and San
Lucas (Fig. 2). A search of the collections of A. maritima in four her-
baria (UC, CAS, DS, SD) failed to turn up any specimens collected
along this 80-km segment of coastline. The purple-flowered ssp. maritima
was not seen to co-occur with ssp. capensis along this stretch of coast,
although it was observed on beaches 50 km north of Todos Santos and
at San José del Cabo, 35 km northeast of San Lucas.

Several features, besides the morphological ones mentioned, serve to
distinguish ssp. capensis from ssp. maritima. Estimated pollen viability
of ssp. capensis, determined by staining with aniline blue-lactophenol,
averaged 50 percent, much lower than that of ssp. maritima, which aver-
aged 85 percent. The pollen of ssp. capensis when brushed on stigmas of

224

1978 | JOHNSON: ABRONIA 220

Fic. 1. Abronia maritima Nutt. ex Wats. ssp capensis A. F. Johnson, drawn from
live specimen growing in greenhouse. A) branch with flower head (x 2/3), B)
detail of leaves and flower head (x 1), C) detail of flower (x 3), D) anthocarp
Cx 2),

ssp. maritima never produced seed, whereas the reverse procedure read-
ily did so, a phenomenon often noted in autogamous species.

Secondly, ssp. capensis, grown in isolation in the greenhouse, is auto-
gamous and produces copious seed, whereas ssp. maritima from Califor-
nia does not produce seed under these circumstances. However, ssp.
maritima from the shores of the Gulf of California was found to be auto-
gamous when grown in the greenhouse.

A third distinction is in the amino acid complement of the nectar,
which includes tryptophan in ssp. capensis and lacks it is ssp. maritima.
Baker and Baker (1976, 1977) have found the set of amino acids in the
nectar to be constant within a species and useful for detecting hybrids,
the nectar of which contain amino acids from both parents. This is the
case with the nectar of hybrids between ssp. capensis and ssp. maritima
which contains tryptophan.

Given the propensity of species of Abronia to hybridize (Tillett,
1960), it seems reasonable to suppose that ssp. capensis is the end result,
after introgression and selection, of an initial cross between 4. maritima
maritima and another species of Abronia. The closest candidate is
Abronia gracilis Benth., a desert annual having a bright pink perianth
with a white center, unreflexed limb, and small leaves with crenulate
margins. Although this species does not presently occur in the Cape Re-
gion (ranging only as far south as the Magdalena plain), its nectar amino
acid complement supports the hypothesis. Its nectar contains tryptophan

226 MADRONO [Vol. 25

Ss’

CALIFORNIA

=
S.Jeerr”
ce

‘
‘
‘
‘

!

@ abronia maritima- Pacific type

O A maritima-Gulf type

© a. maritima ssp. capensis

0

Fic. 2. Distribution of the variants of Abronia maritima.

and no other amino acids not found in A. maritima maritima. On this
basis, another candidate, 4. wmbellata Lam., a northward-ranging annual
morphologically similar to A. gracilis, can be considered less likely since
its nectar contains lysine and leucine, not found in the nectar of A.
maritima capensis. However, several attempts to cross A. gracilis with
A. maritima capensis and A. maritima maritima from both the Pacific
and Gulf coasts did not produce seed.

Populations of ssp. maritima from the Gulf of California resemble ssp.
capensis and differ from Pacific coast populations of ssp. maritima, not
only in being autogamous, but also in several morphological features
(Johnson, 1978). The differences are maintained when plants from Pa-
cific and Gulf coasts grown from seed in the greenhouse. Plants of ssp.

1978 | NOTES AND NEWS 227

maritima from the Pacific coast north of Todos Santos have a reddish-
purple perianth with reflexed limb and leaves with generally entire mar-
gins. Gulf coast plants have a shorter, bright purplish-pink perianth with
unreflexed limb, and smaller, crenately—lobed leaves. The latter features
are duplicated in some of the F, progeny resulting from crosses between
ssp. capensis and ssp. maritima from the Pacific coast (Silver Strand
State Beach, California). It appears possible that the morphological dif-
ferences between Pacific and Gulf coast populations of ssp. maritima
are also the result of an earlier episode of hybridization, perhaps the
same one that produced ssp. capensis.

ACKNOWLEDGMENTS
I thank Irene Baker for the nectar analyses and Annetta Carter for helpful
suggestions.

LITERATURE CITED

Baker, H. G. and I. Baker. 1977. Intraspecific constancy of floral nectar amino acid
complements. Bot. Gaz. 138:183-191.

Baker, I. and H. G. BAKeEr. 1976. Analyses of amino acids in flower nectars of hy-
brids and their parents with phylogenetic implications. New Phytol. 76:87-98.

Jounson, A. 1978. Some aspects of the autecology of Abronia maritima. Ph.D. the-
sis, Botany Department, University of California, Davis.

TiLteTT, S. S. 1960. The maritime species of Abronia. Ph.D. thesis, The Claremont
Graduate School.

NOTES AND NEWS

Lasthenia californica (COMPOSITAE), ANOTHER NAME FOR A COMMON GOLDFIELD.—
In conjunction with a systematic study of certain Californian Helenieae (Johnson.
Systematics of Eriophyllinae (Compositae). 1978. Ph.D. dissertation. Univ. Califor-
nia, Berkeley.), we had opportunity to examine type specimens that were previously
unavailable (see Ornduff. 1966. Univ. Calif. Publ. Bot. 40.). The holotype of Las-
thenia californica DC. ex Lindley, 1 Aug 1835, Edwards’s Bot. Reg. 21: facing pl.
1780 (“HHS [Hort. Horticultural Society of London]”, J. Lindley. CGE!) was
found to be referable to what until now has been called L. chrysostoma (Fischer &
Meyer) Greene (basionym: Baeria chrysostoma Fischer & Meyer, Jan 1836, Index
Seminum Hert. Petrop. 2:29) and not L. glabrata Lindley, Edwards’s Bot. Reg. 21:
facing pl. 1780 (lectotype designated here by Ornduff: California, “1833 [22 Dec
1830-18 Aug 1832], D. Douglas. CGE!). Because publication of the epithet calzfor-
nica predates that of chrysostoma, the name in use for this conspicuous plant of
southwestern Oregon, California, southern Arizona, and northern Baja California
must be changed to L. californica. — Date E. JouHnson, Hunt Institute for Botani-
cal Documentation, Carnegie-Mellon University, Pittsburgh, PA 15213, and RoBErtT
OrnpurF, Department of Botany, University of California, Berkeley 94720.

PUBLICATION OFFER — Robert Ornduff (address above) has a number of copies of
“A Biosystematic Study of the Goldfield Genus Lasthenia” Univ. Calif. Publ. Bot.
40, 1966. He would be glad to send gratis copies to individuals or libraries upon
request.

RANUNCULUS GERANIOIDES H.B.K.EX DC.
IN COSTA RICA AND PANAMA

THomas DUNCAN
Department of Botany and University Herbarium
University of California, Berkeley 94720

Authors of recent floras and previous authorities on Ranunculus have
reported R. repens L. and R. pilosus H.B.K. ex DC. from Costa Rica
and Panama (Standley, 1937; Benson, 1948; Duke, 1962). Recent
studies of the Ranunculus hispidus Michx. complex have revealed that
R. geranioides H.B.K. ex DC. has been treated erroneously as R. pilosus
H.B.K. ex DC. (Benson, 1948; Duke, 1962) and occurs in both Costa
Rica and Panama. The purpose of this paper is to note these corrections
to the floras of Costa Rica and of Panama and to document the distri-
bution of R. geranioides and its differences from R. repens in these
countries.

Sepals reflexed; petals lanceolate, commonly 7-10, rarely 5; achenes

papillate . .. ¢ Oe epee geranivides
Sepals patent; petals ahorate commonly 5 aes: many as 10; achenes
smooth . . eee teror R. repens

1. Ranunculus Pee HB. K. ex DC. eyee ie 286. 1818

Plants stoloniferous and rhizomatous; sepals reflexed, 4.0-7.0 mm
long, 2.0-3.5 mm wide, one-half to fully as long as the fails petals
yellow, (5)7-10, 6.0-11.7 mm long, 2.8-6.0 mm wide, the widest point
above the middle; nectary scale obovate; achene wall with minute to
conspicuous papillae, achene beak one-third to one-half as long as the
body, the tip often slightly curved and tapering from a broad base;
n = 16 (Panama, Duncan 2331).
FLOWERING AND FruiTING Dates: Throughout the year.
Hasitats: Paramos, meadows, fields, streambanks, and roadsides.
DISTRIBUTION: Primarily South American from Colombia to northern
Peru; northern limit of range on Volcan Baru, Chiriqui Province, Pana-
ma, and near Orosi and on Volcan Turrialba in Costa Rica; 1300-3300 m
in Costa Rica and Panama.
SPECIMENS EXAMINED: Costa Rica: Cartago: At foot of Orosi waterfall,
5 May 1957, Rodriguez C. 428 (UC, MICH). Volcan Turrialba, 2000 m,
Jan 1899, Pittier 7550 (GH). Volcan Turrialba, ca. 3300 m, 13 Feb
1922, Greenman and Greenman 5582 (MO). Volcan Turrialba, 2900 m,
17 Sep 1969, Weston and Kincaid 6136 (UC). Volcan Turrialba, ca. 20
km by road above Turrialba and 8 km by road N of Pastora, 7000 ft,
16 Feb 1974, Duncan 2314 (MICH, UC). San José: Potrero of Finca
Santa Rosa north of E] Alto de Cabeza de Vaca on Rio Sucio, 14 Nov
1929, Dodge and Thomas 4949 (MO). Sables du Rio Parrita au Copey,

228

1978] DUNCAN: RANUNCULUS 229

Fig. 1. (Leit) Ranunculus geranioides trom Colombia. A. Whole plant. B. Flower.
C. Petal. D. Sepal. E. Infructescence. F. Achene. G. Achene (from Volcan Baru,
Panama). H. Receptacle. Scale line represents ca. 3 cm in A, 1.5 mm in C and D,
5 mm in B, 2 mm in E and H, and 1 mm in F and G.

Fig. 2 (Right) Ranunculus repens from Volcan Turrialba, Costa Rica. A. Whole
plant. B. Flower. C. Petal. D. Sepal. E. Infructescence. F. Achene. Scale line repre-
sents ca. 3 cm in A, 1.5 mm in C and D, 5 mm in B, 2 mm in E, and 1 mm in F.

1800 m, Feb 1898, Tonduz 11873 (GH). PANAMA: Chiriqui: Valley of
the upper Rio Chiriqui Viejo, vicinity of Monte Lirio, 1300-1900 m, 27
Jun-13 Jul 1935, Siebert 159 (MO). Along the trail between Cerro Punta
and the Quebrado Bajo Grande, 2000-2100 m, 28 May 1970, Wilbur
11924 with Luteyn and Armond (GH, DS, MO). Along Quebrado Bajo
Grande below road to Cerro Punta and ca. 1 km from Cerro Punta, 6000
ft, 24 Feb 1974, Duncan 2331 (MICH, UC). Vic. of Bajo Chorro, 1900
m, 20-22 Jul 1940, Woodson and Scherry 646 (GH). Boquete District,
Bajo Chorro, 7000 ft, 26 Mar 1938, Davidson 444 (GH).

The name most frequently applied to these populations, R. pilosus
H.B.K. ex DC. is based on material collected by Humboldt and Bon-
pland in Colombia. This name is currently treated as a synonym of R.
praemorsus var. praemorsus (Duncan, 1979), which is widely distributed
in the Andean paramos from Venezuela to Argentina. Earlier authors
treated R. pilosus as conspecific with R. petiolaris H.B.K. ex DC. var.
petiolaris. The latter is a widespread Mexican, Central American, and
northern South American taxon and is what earlier workers thought R.
geranioides to be. Benson (1948) emphasized the lack of stolons for R.
petiolaris var. petiolaris. However, the plants from Costa Rican and

230 MADRONO [Vol. 25

Panamanian populations are distinctly stoloniferous. Additionally, the
short, stout-based, slightly curved achene beaks, clavate receptacles, and
fibrous roots readily distinguish R. geranioides from R. petiolaris var.
petiolaris, which possesses long, easily broken, straight achene beaks,
conical receptacles, and tuberous roots. The illustration in Duke (1962)
is not from material of R. geranioides from Panama. Apparently a speci-
men of R. petiolaris var. petiolaris was used. This taxon does not occur
in Panama or Costa Rica.

An additional specimen of Davidson 444 is reported to be at MO. A
search of their collections has not resulted in the discovery of this dupli-
cate. Duke (1962, based on the identification of Benson) reports that
this specimen is R. repens. I doubt this report because R. repens is cur-
rently not known to occur in Panama and the duplicate at GH is R.
geranioides.

2. Ranunculus repens L. Sp. Pl. 554. 1753.

Plants stoloniferous and rhizomatous; sepals appressed 4.0-6.0 mm
long, 2.0-4.0 mm wide up to two-thirds as long as the petals; petals
yellow 5—7 (10), 6.0-10.0 mm long, 5.0-12 mm wide, the widest point
above the middle; nectary scale flabellate; achene wall smooth, the
margin narrow, inconspicuous, or absent; achene beak less than one-
third as long as the body, the tip slightly curved and tapering from a
broad base; = 16 (Costa Rica, Duncan 2302).

FLOWERING AND FRuITING Dates: Throughout the year.

Hasitats: Disturbed roadsides, fields, and wet meadows.

DISTRIBUTION: Native of Europe with a cosmopolitan distribution; in
Costa Rica in the provinces of Cartago, Heredia, and San José; 1500-
2500 m.

SPECIMENS EXAMINED: Costa Rica: Cartago: Near stream, beyond Pa-
cayas, 9 Jun 1957, Rodriguez C.471 (UC). Lower potrero of Finca Coli-
blanco, 1620-1910 m, 17 Oct 1929, Dodge and Thomas 4530 (MO).
Volcan Turrialba, ca. 2900 m, 17 Sep 1969, Weston and Kincaid 6135
(UC). Volcan Turrialba, ca. 20 km by road above Turrialba at town of
Pastora, 7000 ft, 16 Feb 1974, Duncan 2314 (MICH, UC). Heredia:
Along roadside between Los Cartagos and Vara Blanca on road to Vol-
can Poas, 6000 ft, 20 Feb 1974, Duncan 2322 (MICH, UC). Vara
Blanca de Sarapiqui, north slope of Central Cordillera, 1500-1750 m,
Jul-Sep 1937, Skutch 3249 (MO, GH). San José: Potreros of Rancho
Redondo, 2200-2600 m, 18 Nov 1929, Dodge and Thomas 4946 (MO).
La Palma, 1460 m, Aug 1898, Tonduz 7402 (GH). State Unknown: Vic.
of Los Nubes, 1800 m, 1 Dec 1937-1 Jan 1938, Allen 714 (GH).

The three specimens cited by Duke (1962) as R. repens from Pana-
ma are treated here as R. geranioides. Ranunculus repens is currently
not known to occur in Panama. Standley (1937) included all material of
R. geranioides in R. repens. He was correct in doubting the previous
treatment of these populations as R. petiolaris var. petiolaris (R. pilosus

1978] NOTEWORTHY COLLECTIONS 231

sensu other authors) but apparently considered no possibilties other than
R. repens. Standley (1937) suggested that R. repens was introduced
from Europe with grass seed.

ACKNOWLEDGMENTS
The field work conducted during this project was supported by an NSF Pre-
Doctoral Disseration Improvement Grant. I thank Mr. Jorge Campabadal of the Or-
ganization for Tropical Studies for the use of their facilities, Mr. Alphonso Rodri-
guez for his assistance in field work, and Miss Jan McCarthy for her illustrations.

LITERATURE CITED

Duke, J. A. 1962. Flora of Panama (Ranunculaceae). Ann. Missouri Bot. Gard.
49:143-153.

Benson, Lyman. 1948. A treatise on the North American Ranunculi. Amer. Midl.
Naturalist. 40: 1-268.

Duncan, THomas. 1979. A taxonomic study of the Ranunculus hispidus complex in
the Western Hemisphere. Submitted to Univ. Calif. Publ. Bot.

STANDLEY, P. C. 1937. Flora of Costa Rica (Ranunculaceae). Publ. Field Mus. Nat.
Hist., Bot. Ser. 18:434-435.

NOTEWORTHY COLLECTIONS

Ed. Note: With this issue a new format is inaugurated for “range extensions” and
similar notes. Its purpose is to provide a greater array of useful data in more tele-
graphic style than has been customary. Prospective authors of these notes should
study carefully the conventions of the new format.

TEESDALIA CORONOPIFOLIA (Bergeret) Thellung (CRUCIFERAE) —USA, CA, Sonoma
Co., W edge Santa Rosa, SW of intersection of Fulton and Piner roads, locally com-
mon in wet areas with Blennosperma, 7 Mar 1977, C. F. Quibell 1392 (BM, CDA,
GH, ROPA, RSA, UC). Basionym: Thlaspi coronopifolium Bergeret; for discussion
of synonymy, see Thellung, A. 1912. Repert. Spec. Nov. Regni Veg. 10:289-290.

Previous knowledge—Native to S Europe and N Africa; adventive in N Europe.
(Herbaria consulted: CAS, DS, JEPS, UC; published sources: Clapham, A. R. et al.
1962. Fl. Brit. Isles, 2nd ed.; Tutin, T. G. et al., eds. 1964. Fl. Europaea, vol. 1.)
The only other member of the genus, 7. nudicaulis (L.) R. Br. [= Iberis nudi-
caulis L.], is native to W and central Europe and has been recorded as locally
adventive in B.C. (Taylor, R. L. and B. MacBride. 1977. Vasc. Pls. Brit. Columbia),
WA and OR (Hitchcock, C. L. and A. Cronquist. 1973. Fl. Pacific Northw.), and in
E USA from MA to NC (Fernald, M. L. 1950. Gray’s Man. Bot., 8th. ed.). Diag-
nostic characteristics—Our plants differ from 7. nudicaulis principally in having
acutely (vs. bluntly) lobed leaves, subequal (vs. unequal) petals, and 4 (vs. 6)
stamens.

Significance—Apparently, first record of species for N.A. (In Gleason, H.A. 1952.
New Britton and Brown Ill. Fl., vol. 2, plant figured as T. nudicaulis may be T.
coronopifolia) ; first record of genus for CA (sources in addition to those cited
above: Abrams, L. 1944. JIl. Fl. Pacific Sts., vol. 2; Munz, P. A. 1959. A Calif. FI.;
———.. 1968. Suppl. Calif. Fl.; —————. 1974. A Fl. S. Calif.). The value of
documenting introductions and migrations has been discussed by Shinners (1965.
Sida 2:119-128) and by Strother and Smith (1970. Taxon 19:871—874) —CHARLES
F. QuiIBELL, Department of Biological Sciences, Sonoma State College, Rohnert
Park, CA 94928 and Joun L. StrotHeEr, Botany - Herbarium, University of Cali-
fornia, Berkeley 94720.

232 MADRONO [ Vol. 25

AGROSTIS HUMILIS Vasey (POACEAE).—USA, CA, Tuolumne Co., moist alpine
meadow at outflow of Blue Canyon Lake, 3048 m, (NE% S9 T5N R20E), 22 Jul
1976,, Neisess 67 (OBI, US). Mixed community, including Carex nigricans C. A.
Mey., Pedicularis groenlandica Retz., Potentilla breweri Wats., Dodecatheon alpinum
Greene, Caltha howelii Greene, Aster alpigenus Gray var. andersonii Peck, Salix
anglorum Cham. var. antiplasta C. K. Schneid., Castilleja culbertsoni Greene, Tri-
setum spicatum Richt. var. molle Beal, Juncus longistylis Torr., and Claytonia
nevadensis Wats. Collection verified by T. R. Soderstrom, US, Apr 1977.

Previous knowledge—Range: Cascade and Olympic Mts. of B.C., WA, OR; across
NV and UT (Uinta Mts.) ; Rocky Mts. from MT south to NM (Herbaria consulted:
US; UC and JEPS kindly checked by Alan R. Smith; published sources: Hitchcock,
A. S., Man. Grasses U.S., 1950; Hitchcock, C. L. et al., Vasc. Pls. Pac. Northw.,
Cronquist et al., Intermountain Flora, 1977.). Diagnostic characters—Small tufted
perennials; culms 3-18 cm tall; ligules 0.5-1.5 mm long, obtuse to truncate; blades
0.5-1.2 mm broad, mostly basal; panicles loosely contracted, 1.5-4 cm long; glumes
subequal, 1.5—2.2 mm long, lanceolate, acute, purple; lemma 1.5-1.8 mm long, awn-
less; palea shorter, about 7% its length; rachilla vestige lacking or very short. Blue
Canyon population exhibits maximum dwarfing.

Significance—Previously unlisted in State and local floras. Full distribution in
California unknown. Habitat and range suggest that it is relictual in sierran alpine
tundra.—Kurt R. Netsess, Department of Botany and Plant Sciences, University
of California, Riverside 92521.

ASPLENIUM SEPTENTRIONALE (L.) Hoffm. (ASPLENIACEAE). —USA, CA, Lassen
Volcanic Natl. Park: Raker Peak, SW slope, 21 Jun 1976, D. Showers 3533 (SFSU) ;
1 km W of Lost Creek Camp, 2 Sep 1976, D. Showers 3748 (SFSU, CAS). Rare.
Other localities include: Eagle Peak, Loomis Peak, and the North Domes. Scattered
populations, in crevices of dacite volcanic rock, fully exposed, 1800-2700 m. Fre-
quent associates are Penstemon newberryi, Cryptogramma acrostichoides, and Poly-
stichum scopulinum. Verified by J. T. Howell, Apr. 1977 (D. Showers 3748 ).

Previous Knowledge—Known from SD and OK, W to OR and Baja Calif.; also
WV; Eurasia. Known in CA from, Tulare Co., Columbine Lake, collected by J. T.
Howell in 1942. Single locality in OR, Douglas Co., Copeland Creek on the N Ump-
qua River collected by F. Lang. (Herbaria consulted: CAS, DS, UC, SFSU; pub-
lished sources: Munz, Supplement to a California Flora, 1968; Amer. Fern J. 59:
45-47. 1969). Diagnostic characters—small tufts consisting of grasslike fronds, the
stipe longer than the blade, the latter divided into 2—3 linear segments.

Significance—A second locality in CA. The Lassen populations are between the
two known localities for the southern Cascades-Sierra Nevada axis. They are 310
km SE of the Douglas Co, OR locality and 520 km N of the Tulare Co, CA
locality —Davip W. SHowErs, Department of Ecology and Systematic Biology, San
Francisco State University, San Francisco 94132.

NOTES AND NEWS

ENDANGERED SPECIES IN CALIFORNIA: FEDERAL PROCEDURES AND STATUS REPORT. —
There is considerable confusion about the various federal actions that have taken
place relating to rare plants. This is exemplified by the statement in the April 1978
Madrono (25:107) that Cordylanthus mollis ssp. mollis has Endangered status
under the Endangered Species Act. This is not yet so. It may be well to review the
steps necessary to attain this status.

To be legally recognized as Endangered or Threatened under this act, a taxon must
have been the subject of a proposed and a final rulemaking published in the Federal
Register. Critical habitats are given legal standing in the same manner. These rule-
makings are the responsibility of the U. S. Fish and Wildlife Service. So far it has

1978] NOTES AND NEWS 233

taken the following steps concerning plants: A Notice of Review appeared 1 Jul 1975
and included essentially the national list compiled by the Smithsonian Institution
(H.R. Doc. No. 51, 94th Congress, 1st Session, Report on endangered and threat-
ened plant species of the United States, compiled for the Committee on Merchant
Marine and Fisheries by the Smithsonian Institution, 15 Dec 1974). This is a per-
missible, but not mandatory, step. On 16 Jun 1976 a rulemaking proposing 1783
taxa for Endangered status appeared. There has been no proposed rulemaking for
Threatened status. Final rulemakings based on the 1976 action have appeared spo-
radically since then. The first such action for the nation listed four San Clemente
Island taxa on 11 Aug 1977. It covered Lotus scoparius ssp. traskiae, Malacotham-
nus clementinus, Delphinium kinkiense, and Castilleja grisea. On 26 Apr 1978 a
second group was listed; included, along with the notorious Furbish’s lousewort,
were five more California plants: Oenothera deltoides var. howellit and Erysimum
capitatum var. angustatum, both of the Antioch Dunes in Contra Costa County;
Oenothera avita ssp. eurekensis and Swallenia alexandrae, both of the Eureka
Dunes in Inyo County; and Dudleya traskiae of Santa Barbara Island. On 28 Sep
1978 four more California plants joined the select list: Arabis macdonaldiana of Red
Mountain, Mendocino County; Orcuttia mucronata of a single vernal pool in
Solano County; Pogogyne abramsi, an inhabitant of rapidly disappearing vernal
pools in San Diego County; and Cordylanthus maritimus ssp. maritimus, a coastal
salt marsh taxon from Southern California. Only two Critical Habitats have been
proposed for California taxa so far, for the two Antioch Dunes plants mentioned
above. Both were the subject of a final rulemaking on 31 Aug 1978.

Of the 22 plant taxa now listed for the nation as Endangered or Threatened,
thirteen are from California. This impressive proportion testifies not only to the
large number of very rare taxa in the California flora but also to the hard work of
the many amateurs and professionals that have assisted in the California Native
Plant Society’s Rare Plant Project, begun in 1968 at the instigation of G. Ledyard
Stebbins, president from 1966 to 1971.—ALice Q. Howarp, Chairman, Rare Plant
Committee, CNPS, University Herbarium, University of California, Berkeley 94720.

Note added in proof: Amendments to the Endangered Species Act passed on the
final day of the 95th Congress in mid-October will change somewhat in the listing
process outlined above.

A CORRECTION ON THE INDIGENOUS DISTRIBUTION OF KNOBCONE PINE.—In a recent
note (Madrono 25:106. 1978.) I reported a population of Pinus attenuata Lemm.
along Beasore Road N of Bass Lake as a southward range extension. This popula-
tion was thought to be indigenous based on the confirmation by the Timber Man-
agement Officer for the Sierra National Forest that neither knobcone pine nor any
knobcone mixture had been planted in this area. However, a recent communication
from Frank G. Hawksworth (Forest Pathologist, Rocky Mtn. For. Range Expt.
Sta.) and an article in the Fresno Bee from 1971 describe planting by the Forest
Service in the early 1960’s of a knobcone-monterey pine hybrid (P. * attenuradiata
Stockw. and Right.) along Beasore Road. Much of the present population is appar-
ently offspring from these hybrids, many having lost most monterey pine character-
istics—Jon E. Krerertey, Department of Biology, Occidental College, Los Angeles,
CA 90041.

Ed. Note: Jim A. Bartel, Botanist with the Sierra National Forest, Fresno, has pro-
vided the following further information: Pinus x attenuradiata was planted as a
timber tree in several harshly dry sites in the Sierra National Forest in the early
1960’s. The hybrid pine was promoted for its rapid growth but-has not been a good
timber tree because it is readily bent or broken by snow.

234 MADRONO [Vol. 25

GYNODIOECY IN MAMMILLARIA DIOICA (CACTACEAE).—Virtually every flora and
monograph treating Mammiullaria dioica K. Brandegee describes the species as in-
completely or partially dioecious. In her description of the species Brandegee stated
that many plants were either male or female, and others hermaphroditic or “im-
perfectly dioecious in all degrees” (Erythea 5:115-116. 1897). Our observations of
plants in Anza Borrego Desert State Park in SE San Diego Co., CA in 1968 and
1978 indicate that populations of M. dioica in this area are in fact gynodioecious,
the plants being either hermaphroditic or pistillate. Both hermaphrodites and pistil-
late plants set fruit with apparently normal seed. Compared with hermaphrodites,
the flowers of pistillate plants are smaller, with narrower petals, but larger stigmas
(Fig. 1.). The pistillate flowers bear stamens with indehiscent anthers that contain
no pollen. Self-incompatibility, self-compatibility, autogamy, cleistogamy, and
agamospermy are known in the Cactaceae (Ganders, Cact. Succ. J. Gt. Brit. 38:39-
40. 1976), but M. dioica is apparently the only species in the family with imperfect
flowers. Brandegee’s description was based only on plants from near the coast, so

Fic. 1. Pistillate (left) and hermaphroditic flowers of Mammiullaria dioica (scale
in mm).

it is uncertain whether the breeding system of the species differs in coastal and in-
land populations, or whether she misinterpreted the situation. The distribution of
floral forms in this species merits more extensive observation by botanists in the
San Diego region—Frep R. GANpDERS and HELEN KENNEDY, Department of Botany,
University of British Columbia, Vancouver, B.C., Canada V6T 1W5S.

ADDITIONS TO THE FLORA OF THE FARALLON ISLANDS, CALIFORNIA. — The flora of
the Farallon Islands, San Francisco County, California, was recently described by
Coulter (Coulter, Madrofo 21:131-137. 1971) based on observations made in 1968.
Since 1968, I have noted 14 additions, eight previously unreported and six previously
reported species, possibly missed in 1968. A new variety of a previously noted spe-
cies has also been observed. I have not noted any extinctions. I report here on the
additional species noted from 1968 through the summer of 1975.

Before the 1968 flora only two papers had been published on the plants of the
islands. Blankenship, who was on the islands 3-6 July 1892, collected 28 species, 11

1978 | NOTES AND NEWS 255

native and 17 introduced plants (Blankenship and Keeler, Zoé 3:144-165. 1892).
Ornduff, who was there for a short time in May 1960, found only 20 species, 10 na-
tive and 10 introduced, but noted the addition of 3 new ones to the islands (Ornduff,
Leafl. West. Bot. 9:139-142. 1961). For the 1968 flora, I was present on the islands
for three months during the spring and summer of 1968 and found 36 species, 14
previously unnoted. Of the 36 species, 13 were native and 23 were introduced. With
the 14 additions noted here, the island list contains 50 species, 18 native and 32 in-
troduced.

Following is a list of the additional species. Previous listings of species by Blank-
enship and Ornduff are noted. For a description and map of the islands, see the 1968
flora (Coulter, Madrofo 21:131-137. 1971). Names are given according to Munz
(Munz, A California flora. Univ. of Calif. Press, Berkeley, 1959) and where they
have been changed Munz’s names are in synonomy. Names with asterisks are those
of introduced species. Specimens have been placed in the Dudley Herbarium (DS)
except as noted.

*Anagallis arvensis L. {. caerulea (Schreb.) Baumg. Although this taxon was re-
ported by Blankenship, Ornduff, and Coulter, the population consists almost exclu-
sively of the pin-orange variety. A few plants of the blue variety were found near
East Landing about 2 m south of the tram tracks in 1971. They have not been found
in subsequent years.

Bromus cf. maritima Hitchc. Previously unreported, this grass was found in 1975
in a few patches 15 to 30 south of the lighthouse on Lighthouse Hill. It has not been
collected; a photograph of the plant was identified by B. Crampton.

*Cerastium viscosum L. This is likely the C. glomeratum Thuill. of Blankenship.
It has not been reported since 1892. In 1975 a few plants were found by B. Lewis
along the sidewalk northeast of Heligoland Hill. It has not been collected.

*Gnapalium luteo-album L. A new species on the islands, in 1975 a few plants
were found along the tram tracks by B. Lewis.

Juncus bufonius L. Jim Lewis (PRBO) found viable seeds in the damp area be-
tween the living quarters and the paint locker in 1975. Plants grown from these
seeds were collected and identified. The species was reported by Blankenship and
Ornduff but missed by Coulter in 1968.

*Teontodon leysseri (Wallr.) G. Beck. Observed in 1972, this plant is new to the
islands. It grows along the tram tracks between East Landing and the Power House.

*Malva parviflora L. In 1974 many plants were found by R. Boekelheide around
the living quarters and from the living quarters to the water. This plant was reported
by Blankenship but not listed by Ornduff or by Coulter in 1968.

*Medicago hispida Gaertn. In 1975 scattered plants were found in the southeast
section of the island. Perhaps this is M. denticulata Willd. recorded by Blankenship.
It has not been collected.

Montia hallit (Gray) Greene. Found in 1972 but not observed earlier, this plant
grows along the path near the top of Lighthouse Hill.

*Plantago coronopus L. A new plant to the island, many plaintains were found
along the south slope of Lighthouse Hill near the living quarters and between the
living quarters and the Power House.

*Polycarpon tetraphyllum (L.) L. Found on the island for the first time in 1972,
this plant grows commonly along the tram tracks near the living quarters.

Psilocarphus tenellus Nutt. var. tenellus. Listed by both Blankenship and Ornduff
but missed in 1968, this plant grows commonly where the soil is hard and gravelly
in the southeast part of the island.

*Rumex crispus L. One plant was found among the gull colony in the southeast
part of the islands in 1974. It bore fruit in that year and again in 1975. This plant
has not been found on the islands before.

Sagina occidentalis Wats. This plant grows commonly in the southeast part of the

236 MADRONO [ Vol. 25

island where the soil is hard and gravelly. It was noted by Blankenship and Ornduff
but missed in 1968.

*Vulpia myuros (L.) K. C. Gmelin var. hirsuta Hack. (Festuca megalura Nutt.)
In 1975 many patches of this grass were found near the lighthouse on Lighthouse
Hill. This is a new species to the islands.

Some species such as Psilocarphus tenellus and Sagina occidentalis, reported by
both Blankenship and Ornduff, were probably present but overlooked in 1968. Ce-
rastium viscosum, Malva parviflora and Medicago hispida were recorded by Blank-
enship but not by Ornduff or by Coulter in 1968. In 1892 these plants may have
persisted in fenced gardens, protected from rabbit grazing, as suggested by Ornduff.
The gardens have since been abandoned. Between 1972 and 1975 the Point Reyes
Bird Observatory carried on a program to eliminate the rabbits, which were finally
completely eliminated in 1975. With the reduction in the rabbit population these
plants may have been able to recolonize the islands; or, perhaps, these species per-
sisted as repressed populations, expanding with the reduction in rabbit numbers.

The location where some new species were first recorded suggests the ways in
which these plants came to the islands. Anagallis arvensis forma caerulea, Leontodon
leysseri, and Polycarpon tetraphyllum, found along the tram tracks where there is
much human activity were likely brought by man. Bromus maritima, Montia hallii,
and Vulpia myuros were found near the top of Lighthouse Hill where most migrant
passerine birds first land on the islands. These plants may have been transported by
passerines. Finally, Rumex crispus, found in the gull colony, may have been brought
by gulls, which fly between the islands and the mainland.

I thank R. Boekelheide, J. and B. Lewis, D. Manual, and D. Gaines for pointing
out new plants. Dr. H. Baker, Dr. B. Crampton, and G. True helped in identifica-
tion. This paper has been improved through discussions with J. and B. Lewis and
through comments by Dr. P. Raven and R. Boekelheide on an earlier draft. I very
much appreciate the encouragement of Dr. P. Raven, Dr. H. Baker, and Dr. J. H.
Thomas. The U. S. Coast Guard kindly provided logistic support and the Point
Reyes Bird Observatory generously made possible my stay on the island. I thank the
personnel of the Farallon Island Wildlife Refuge for permission to work on the
island.

This is contribution number 151 of the Point Reyes Bird Observatory. — Mat-
comm C. Courter, Department of Biology, University of Pennsylvania, Philadelphia
19104.

REVIEW

Manual of the vascular plants of Wyoming. By Ropert D. Dorn. Illustrations by
Jane L. Dorn. 2 vols. 1498 pp. 1977. Garland Publ. Co., New York. ISBN 0-8240-
9905-2. $95.

Wyoming now has a flora! A conspicuous blank spot has been filled in for plant
taxonomists, biogeographers, ecologists, resource managers, users of Wyoming’s nat-
ural resources, and those who appreciate and have the opportunity to enjoy its rich
natural beauty.

Many of Wyoming’s political leaders and residents are salivatingly eager to exploit
its coal, oil, forests, rangelands, wildlife, water soils, scenery, and other operationally
non-renewable natural resources. Others wish to apply a conservation ethic, or legal
restrictions, to unregulated use. Both groups have had a most useful tool handed to
them free, more or less, by an independent, dedicated, skillful scientist.

The manual is excellent. Dorn is a practiced, perceptive, industrious plant collector.
He mentions giving himself only three years to do the flora. Thus, some weeds and
all infraspecific taxa are not included, distributions are given only within Wyoming
and in broad categories, habitat information is minimal. 2144 species are well de-
scribed. Leading families are Compositae (with 17.3% of the species), Gramineae

1978] . REVIEW 237

(10.2), Leguminosae and Cruciferae (5.8 each), Cyperaceae (5.5), Scrophulariaceae
(4.6), Ranunculaceae (3.0), Umbelliferae (2.6), and Boraginaceae, Caryophyllaceae,
and Chenopoliaceae (2.4). Dorn’s keys are direct, imaginative, practical for field use,
and have been tested (p. 2). Keys to fruiting plants are provided for some groups
(Cruciferae, Umbelliferae, Astragalus, Salix); vegetative characteristics are often
used. Taxa are arranged alphabetically. References are given to recent monographic
treatments, and sources of original descriptions are given. Synonomy seems adequate.

Improvements in a new edition might include a less lavish use of paper simply to
condense volume and weight. More detailed distributions are desirable, particularly
since the Rocky Mountain Herbarium at Laramie has a file of dot maps for Wyo-
ming plants, and many of the species are mapped in Hultén’s Alaska flora and his
other publications and still others in the monographs Dorn gives as references. Ref-
erence should have been made to Wyoming taxonomic work already done. These
include Beetle and May’s (1971) treatment of the grasses, Beetle’s on the section
Tridentatae of Artemisia (1960), Porter’s series on families through the Fumartaceae
(1962-1972), theses, local floras such as Shaw’s for Teton County (1976), Despain’s
for Yellowstone National Park (1975), Nelson’s unfortunately unpublished one for
the Medicine Bow Mts. (1974). All of these contain valuable information on the
flora of Wyoming that Dorn’s flora does not.

Some comments can be made on individual taxa. Beetle’s treatment of Artemisia
tridentata and its allies is more perceptive, and an exception could have been made
here for including subspecies. Puccinellia-Glyceria-Torreyochloa is less confusing if
attention is paid to the species’ habitats. A few species from Teton County are not
included (Antennaria plantaginifolia, Dodecatheon jeffreyi, Carex subfusca).

A condensed appendix supplies some of the accessory information vital for under-
standing the flora. The map at 1/5.8x10® could have been at 1/3x10® and still fit the
page but supply more information. A paleobotanical discussion adds to the material
on geography, climate and floristic elements given by Porter (1963:6-8). Dorn’s
floristic discussion adds some interesting ideas but omits some of Porter’s details.
“Vegetation types” recognizes 47 plant associations that can be seen in the field. It
avoids to a large extent both problematic casual explanations and physiognomic
groupings. Soils limitations of several species are first mentioned here. Is grassland
diversity slighted? Are data on these kinds of vegetation so lacking that no hierar-
chial arrangement at all is possible? desirable? References here are too selective.
Rare and endangered species are listed by counties, and this list is a first for Wyo-
ming. Early collectors are briefly mentioned. A systematic summary lists numbers of
genera and species by division, subclass, and families. The glossary and drawings, by
Jane L. Dorn, are excellent and very useful. The volumes are fully indexed. Eight
pages of Additions conclude the manual.

Dorn’s book is a valuable gift to the people of Wyoming and to botanists every-
where. Both groups have needed it. The high cost obviously makes the work less
available. Unfortunately no one will benefit from the $95 price. One must conclude
that our present methods of getting floras written, or not written for North America,
are wasteful of a valuable resource, namely skilled botanists — their industry, imagi-
nation and training. — Jack Major, Botany Department, University of California,
Davis 95616.

238 MADRONO [ Vol. 25

DUES INCREASE ANNOUNCEMENT

Because of increasing costs, the Council of the California Botanical Society has
reluctantly voted to increase regular dues for 1979 from $12 to $15 and student dues
from $8 to $10. Institutional subscriptions will rise from $20 to $22.

The size of MaproNo has remained fairly constant since 1973, but cost per page
has increased about 50 percent since then — from $35 to $53. These figures do not
include expenses of the Treasurer, Secretaries, or Editor, which now total about
$1000 per year. Expenses have exceeded income since 1974, and our endowment
fund has dwindled to the point that the current total assets of the Society would
sustain only about six months of normal activity. Meanwhile, the number of life
memberships, from which no further income may be anticipated, has increased to
more than 30.

To insure a firm financial footing allowing the publication of 200-250 pages a
year in MAproNO, a dues increase now is absolutely necessary. Simultaneously, poli-
cies on editorial fees for longer papers are being clarified and made more stringent;
and ways to lower publication costs are being sought. Nevertheless, it seems likely
that small increases will be necessary as long as costs continue to inflate.

REVIEWERS OF MANUSCRIPTS
The Editor thanks Barbara D. and Grady L. Webster, the members of the Board
of Editors, and authors of papers for their cooperation and help in many ways. Spe-
cial appreciation goes to John L. Strother, Alan R. Smith, and Alice Q. Howard,
without whose generous assistance the mid-volume editorial transition ere have

been much more difficult.
Reviewers of manuscripts for Volume 25 are gratefully acknowledged below. Inad-

vertent omissions, if any, are sincerely regretted.

Loran C. Anderson
Herbert G. Baker
Michael G. Barbour
Spencer C. H. Barrett
Jim A. Bartel

Jerry M. Baskin
John H. Beaman
Bert G. Brehm
Annetta M. Carter
Kenton L. Chambers
Tsan L. Chuang
Lincoln Constance
Stanton A. Cook
Beecher Crampton
William B. Critchfield
Robert W. Cruden
Alva Day

Lauramay T. Dempster
Melinda F. Denton
William J. Ferlatte
David Fruchter
Thomas C. Fuller

Fred R. Ganders
Frank W. Gould
J. Robert Haller
Steven N. Handel
Ted L. Hanes
Gary Hannan
Charles B. Heiser
Ann F. Johnson
Daniel H. Janzen
Donald R. Kaplan
Jen E. Keeley
Suzanne Koptur
Steven P. Lynch
Tom Mabry
Richard N. Mack
Jack Major
Deborah Mangis
Mildred E. Mathias
Rogers McVaugh
Richard S. Mitchell
Reid Moran

Lazlo Orloci

Robert Ornduff
Rexford E. Palmer
Robert W. Pearcy
Lawrence H. Pike
Louis F. Pitelka
Duncan M. Porter
Charles F. Quibell
Velva E. Rudd
John O. Sawyer, Jr.
Rudolf Schmid
Alfred E. Schuyler
Richard Spellenberg
Dean Wm. Taylor
Ronald J. Taylor
John H. Thomas
Robbin W. Thorp
Alice M. Tryon
Warren H. Wagner, Jr.
Dieter C. Wasshausen
Thomas J. Watson
Robert Wright

1978]

INDEX 239

INDEX TO VOLUME XXV
Classified entries: major subject headings, including key words from titles; botani-
cal names (new names are in boldface) ; reviews. Incidental references to taxa (in-
cluding lists and tables) are not indexed separately. Names of authors followed by
titles of articles are listed alphabetically in Table of Contents, pp. ii-iv.

Abronia maritima ssp. Capensis, 224
Agrostis humilis, new to California, 232
Announcements, 64, 150, 176, 238
Arctostaphylos, leaf angle and light ab-
sorptance, 133
Asplenium septentrionale, 232
Axiniphyllum, taxonomy, 46; A. sagit-
tilobum, 50; A. pinnatisectum, 52
Biogeography, New York Mountain
ferns, 54

California, 59, 60, 113, 159, 205, 227
Klamath Region, 93, 138, 218
Modoc Region, 9
Mojave Region, 54, 187
North Coast Region, 65, 93, 177, 231,
234
Sierra Nevada Region, 93, 133, 177,
eV)
South Coast Region, 39, 104, 227, 234
Central America, 228
Chaparral, Monterey Bay Region, 65
seed dispersal in, 104
Chemotaxonomy of Jepsonia, 39
Chromosome counts, Asteraceae,
Lomatium, 1; Xylorhiza, 205
Claytonia perfoliata diploids from S.
Mexico, 57
Collecting in Mexico, 111
Comandra umbellata ssp. pallida, germi-
nation, 202
Cordylanthus mollis, rediscovery, 107
Crataegus douglasii hybrids with C. mo-
nogya, 211

160;

Dates of publication, 240

Dedication of volume 25, ii

Dispersal patterns related to abundance
and distribution, 104

Draba juniperina, 101

Endangered, threatened, or rare species,
59, 65, 107, 138, 159, 169, 172

Eriogonum combinations, 60; E. kear-
neyi ssp. Monoense, 61; E. sperguli-
num ssp. reddingianum, 61; E. ela-
tum ssp. villosum, 61

Eschscholzia californica, tetrad pollen, 59

Eupatorium shastense, 218

Farollon Islands flora, 234

Ferns, New York Mountains, 54

Fritillaria phaeanthera, synonym of F.
eastwoodiae, 93

Galapagos Islands, 58

Galinsoga, relationships, 81

Germination of Comandra, 202

Gossypium turneri, 155

Great Basin plants in Montana, 105

Great Basin states, 171, 205

Gunnera kilipiana, synomized with G.
mexicana, 53

Gynodioecy in Mammillaria dioica, 234

Hemizonia conjugens, 159
Hybridization, in Crataegus, 211; in Fri-
tillaria, 93; in Psilostrophe, 187

Jatropha, new species, 30; J. giffordi-
ana, 30; J. moranil, 34; J. mce-
vaughii, 36

Jepsonia, flavonoids, 39

Juncus bufonius var. occidentalis, 104

Lasthenia californica, new name for L.
chrysostoma, 227

Leaf angles and light absorptance in Arc-
tostaphylos, 133

Lomatium farinosum, reproductive bi-
ology, 1

Lycopersicon cheesmanii, yar. minor,
58

Mammillaria dioica, 234

Maritime vegetation, Monterey Bay Re-
gion, 65

Miexico;:30,46, 53.57, Tih g13,, Voi 1555
169, 187, 221
Baja California, 39, 113, 151, 159, 224

Mirabilis, subgenus Quamoclidion, syste-
matics, 113; M. alipes, 120

Montane vegetation, 177

New York Mountains, ferns, 54
Noteworthy Collections, new format, 231

Opuntia megasperma yar. orientalis,
58; O. echios var. gigantea, 58

240 MADRONO

Pacific Northwest, 1, 44, 113, 132, 172,
211

Phacelia dalesiana, 138

Pinus: P. albicaulis — Vaccinium scopa-
rium association, 139; P. attenuata in
Sierra Nevada, 106; correction, 233;
P. jeffreyi, forest vegetation, 9; P.
monophylla, cone predation by moths,
170

Pollen, 59

Predation of pinyon cones, 170

Psilostrophe, biosystematics and taxono-
my, 187

Quamoclidion, systematics, 113

Raillardella pringlei, 138

Ranunculus: R. californicus in Washing-
ton, 132; R. geranioides, 228

Reproductive biology: Crataegus, 211;
Eschscholzia, 59; Fritillaria, 93; Lo-
matium, 1; Mammillaria, 234; Pinus,
170; Psilostriphe, 187

Reviews: T. Abraham, Northwest Bot-
anical Manuscripts, 185; M. G. Bar-
bour and J. Major, Terrestrial vegeta-
tion “of Calitornia, 174, 175; R. °C.
Barneby, Daleae Imagines, 110; J. C.
Beatley, Vascular plants of the Neva-
da Test Site and central-southern Ne-
vada, 109; R. D. Dorn, Manual of the
vascular plants of Wyoming, 236;

[Vol. 25

N. T. Mirov and J. Hasbrouck, The
story of pines, 110; P. H. Raven and
T. E. Raven, The genus Epilobium in
Australasia, 61

Reviewers of manuscripts, Volume 24,
63; Volume 25, 238

Rocky Mountain states, 101, 105, 139,
4

Scientific collecting in Mexico, 111

Scrophularia laevis, 106

Seed dispersal in chaparral, 104

Shrubs, endemic to Monterey Bay Re-
gion, 65

Southwestern states, 107, 113, 169, 187,

205
Spilanthes darwinil, 58
Stephanomeria malheurensis, 44

Teesdalia coronopifolia, new to North
America, 231

Tragus racemosus, 107

Trichostema in Mexico, 151

Vaccinium scoparium with Pinus albi-
caulis, 139

Vegetation, of Modoc National Forest,
9; of montane slopes, 177

Viguiera websteri, 221

Xylorhiza, chromosome counts, 205

Dates of publication of Madrono, volume 25

No. 1, pp. 1- 64: 30 January 1978
No. 2, pp. 65-112: 15 June 1978
No. 3, pp. 113-176: 14 September 1978
No. 4, pp. 177-240: 30 January 1979

Membership in the California Botanical Society is open to individuals ($15 per year;
students $10 per year for a maximum of seven years). Members of the Society re-
ceive MaproNo free. Institutional subscriptions to MADRONO are available ($20 per
year). The membership or subscription year begins on January 1.

Back issues of MApRONO are available at the following rates (some issues are
out of print):

Vol. 1 (1916-1929) and Vol. 2 (1930-1934), each consisting of 17 numbers: $1.50
per issue and $25.50 per volume for members; $3.00 per issue and $51.00 per volume
for institutions.

Vol. 3 (1935-1936) through Vol. 23 (1975-1976), each biennial, consisting of 8
numbers: $3.00 per issue and $24.00 per volume for members; $5.00 per issue and
$40.00 per volume for institutions.

Vol. 24 (1977) et seq., one volume per year, each consisting of 4 numbers: $4.00
per issue and $15.00 per volume for members; $6.00 per issue and $24.00 per volume
for institutions.

Applications for membership (including dues), orders for subscriptions, requests
for back issues, changes of address, and undelivered copies of MApROoNO should be
sent to the Corresponding Secretary. Renewal payments should be sent to the
Treasurer.

INFORMATION FOR CONTRIBUTORS

Manuscripts submitted for publication in MApRONo should be sent to the Editor.
Membership in the California Botanical Society is prerequisite to review.

Manuscripts and review copies of illustrations must be submitted in triplicate for
articles and in duplicate for short items intended for NOTES AND NEWS. Follow
the format used in recent issues for the type of item submitted and allow ample
margins all around. All manuscripts MUST BE DOUBLE SPACED THROUGH-
OUT. For articles this includes title (all caps, centered), author names (all caps,
centered), addresses (caps and lower case, centered), Abstract, text, Acknowledg-
ments, Literature Cited, Tables (caption on same page), and Figure Captions
(grouped as consecutive paragraphs on one page). Order parts in the sequence listed
ending with Figures, and number each page. Do not use a separate cover page,
“erasable” paper, or footnotes. All measurements should be in S.J. (metric) units.

Line copy illustrations should be clean and legible, proportioned (including their
captions) to the MaAproNo page, and designed for reduction to 2/3 original size.
Scales should be included in Figures, as should explanation of symbols. In addition,
maps must include latitude and longitude references. Halftone copy should be de-
signed for reproduction at actual size. In no case should original illustrations be sent
prior to the acceptance of a manuscript. When needed, they should be mounted on
stiff board and mailed flat.

Presentation of nomenclatural matter (accepted names, synonyms, typification)
should follow the format used for Rhus integrifolia in MADRONO 22:288. 1974. Insti-
tutional abbreviations in specimen citations should follow Holmgren and Keuken’s
list (Index herbariorum, Part 1. The herbaria of the world. Sixth edition. 1974.
Regnum Veg. vol. 92). Abreviations of names of journals should be those in Botani-
co-Periodicum-Huntianum (Lawrence, G. H. M et al. 1968. Hunt Botanical Library,
Pittsburgh). If the correct abbreviation cannot be determined, the full title of the
journal should be used. Titles of books should be given in full, together with the
place and date of publication, name of publisher, and edition, if other than the first.

All members of the California Botanical Society are allotted eight pages in the
journal over a two-year period. Beyond that number of pages, a required editorial
fee of $40.00 per page will be assessed. The purpose of this fee is not to pay directly
for the costs of publishing any particular paper, but rather to allow the Society to
continue publishing MApRONO on a reasonable schedule, with equity among all mem-
bers for access to its pages. Printer’s fees for illustrations and typographically diffi-
cult material @ $35.00 per page (if their sum exceeds 30 percent of the paper) and
for author’s changes after typesetting @ $3.00 per line will be charged to authors.

Contents, continued

REVIEWS
TERRY ABRAHAM, Northwest Botanical Manuscripts. An
indexed register to the papers, 1867-1957, of
Wilhelm Nicholaus Suksdorf, William Conklin Cusick,
Charles Vancouver Piper, Rolla Kent Beattie, and

Harold St. John (Joseph Ewan) 185
RoBErT D. Dorn, Manual of the Vascular Plants of Wyoming
(Jack Major) 236
BOOKS RECEIVED 186
DUES INCREASE ANNOUNCEMENT 238
REVIEWERS OF MANUSCRIPTS 238
INDEX TO VOLUME XXV 239

STATEMENT OF OWNERSHIP, MANAGEMENT, AND CIRCULATION
(Act of Oct. 23, 1962; Section 4369, Title 39, United States Code)

Madronio, A West American Journal of Botany, is published quarterly at Berke-
ley, California.

The Publisher is the California Botanical Society, Inc., Life Sciences Building,
University of California, Berkeley, California 94720.

The editor is James C. Hickman, Department of Botany, University of California,
Berkeley, California 94720.

The owner is the California Botanical Society, Inc., Life Sciences Building, Uni-
versity of California, Berkeley, California 94720. There are no bondholders, mort-
gagees, or other security holders.

The average number of copies distributed of each issue during the preceding 12
months is 955; the number of copies of the single issue closest to the filing date is 945.
I certify that the statements made by me above are correct and complete.

James C. Hickman, Editor

September 28, 1978

ie : oF
5 : '

a ‘ ms : a

5 i 0
cen

q :

( “
. i H

“~

or

i SA Ze a / x 8 i ae Pig coal

= = ae
ILALILSNI NVINOSHLINS S3IYVYEIT LIBRARIES

=
SMITHSONIAN INST

S

SMITHSONIAN
NVINOSHLIWS S:

NVINOSHLIWS
SS
AN \
BAN
YS
SMITHSONIAN

SMITHSONIAN

3RARIES SMITHSONIAN INSTITUTION NOILONLILSNI

NVINOSHLIWS S31UNVHE

NOILNLILSNI
LIBRARIES

NVINOSHLINS SS!YVYUdIT LIBRARIES

Y

LLALILSNI INSTITUT

fy
~

SMITHSONIAN

SAJIYVYaIT LIBRARIES

INSTITUTION
ssiuyvugdl)

oil
Ph

INSTITUTION NOILNLILSNI
SUIUVUSIT LIBRARIES

SRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS S31YYV ¢

|

‘

NVINOSHLIWS
NVINOSHLIAS
SMITHSONIAN
NVINOSHLIWS

NVINOSHLINS LIBRARIES SMITHSONIAN INSTITUTI

NOILALILSNI

NOILLNLILSNI
NOILNLILSNI

3RARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS Sa1uVul

S3IYVUGIT LIBRARIES SMITHSONIAN

INSTITUTION
INSTITUTION

INSTITUTION

TLALILSNI NVINOSHLINS S3I1YVYdIT LIBRARIES SMITHSONIAN INSTITUT!

1g

BRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVUE

NVINOSHLINS S31uVveall LIBRAR LEo

SMITHSONIAN
NVINOSHLINS
SMITHSONIAN
SMITHSONIAN

a

TITUTION NOILNLILSNI

LIBRARIES

ILOLILSN!I NVINOSHLINS S3IYVYEGIT LIBRARIES
‘se

ryVYSIT LIBRARIES
TITUTION NOILNLILSNI
lUVUGIT LIBRARIES

Rh 2-8-0

MLILSNI NYUINUSRLINGD oaAIaGVGdt)! LIGRANICS oOMIINMOUNIAN
= w~ <. we 2) =
= S is Rc = <
“fl Lg & r pip 5 QE SE = 5
iy 7) nD tf ty ” KK u? Tp)
£ on So UY = \ 8 +
_ = "GG x s =
GG = > = > =
wn a - eee Ww)
RARIES SMITHSONIAN INSTITUTION NOILOLILSNI
} ts z us 2 & 8
” Wy
a = : 2 WN =
<x} a < = We <
oc a PQAQ a
z 5 e a Ns
3 z ad Pa a}
S3'tuvyugit LIBRARIES SMITHSONIAN _INSTITUTI
e z c = ie
wo — wo == Yip =O
2 = 2 5 $Y:
a > rae J" Ls >
te = 7 nm a Gj, D
rm ” ae 7 a -
oo Fae mM > vr}
rep) = w — a
SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLIWS S3iuVvyE
” Zs ” z ”
= ae = a =
= Zz = = 4
ee © a O va
ep rT she 7
O at wo 2 oe O
= i =, = =
ws = = S 5
: VY) : ie w —
LNLILSNI_NVINOSHLINS SSIYVESIT LIBRARIES SMITHSONIAN INSTITUTI
a ul e us 2
as 72] ep)
: e = : pe
= a: ser oc S
a = =
O \: = O = 5
2 3 S 5 =
FRARI ES. SMITHSONIAN _ INSTITUTION NOILNLILSNI NVINOSHLINS S3IYUVus
‘and
Oo — 5 asd \S S
= = a S
=e = = ee =
‘- ee i = =
2 rm “ mi 2)
= W = ” 4 OS
ESN NV INU SHAINS 2 luVvud mul BRARI ES SMITHSONIAN INSTITUT
<< = ES = RN
mf 5 = 4 LT if Mj z + = = : ASN
B he © LL Wi © <a = O WS
Yi, = oO - i, se O a
Yi, = Z Wi = z ms
GQ s > , s > =
w) a ”) * rt 2p)
3RARIES SMITHSONIAN INSTITUTION NOILNIILSNI NVINOSHLIWS
” — w
= at.
: 2 CE =
~~ 4 ie KS
<5 < SSX
oc oc
faa) co
a _

ILLALILSNI

5

VSTITUTION NOILNLILSNI

SJiuyvVYUaIT LIBRARIES

Jiuvads

LUVUGIT LIBRARI

ISTITUTION NOILNLILS

7 ‘
STINgsia >

wo hela tt ee 8 oo

SMITHSONIAN INSTITUTION LIBRARIES

Th |

3 9088 00691 1440

. wry a F
ete ey : N , ny
‘ 4 \ ts ‘
‘ \ i
co eeg A i \ , ;
it : ,
vor a ‘
Sane Ma eae ‘ : ‘ ,
Ney \ ' na ' ' F
1 ina
beythiges Ne rath 2 '
\ ' t 4 ‘ ‘ a
A bey : ’ : ‘
wa ‘ i : .
Cae ‘ ' epg ; oa :
ray ‘i 4 ir te
Wis ea tee inet ;
1 ‘ to
ae oe i
; ah Hol BERG era 2 . “
, ra i ‘
‘ wo rae : Ys ‘
i . ‘
br eae i Wedhats a
x ‘ fae ‘ \ eh hg
= wi : ear) i
‘ ‘ 1 ‘ ‘
n a ath he hye
‘ any ‘ . u,
gto i a se ‘ ‘
, hare , 4 ThA. : ‘
‘ ‘ ‘ PA et :
Lie? ‘ \ ‘
‘ Fn ‘ \
: tS ria ‘ tt had
en aly ae shi ‘ ha i Raper
x . a ( tiie yer ‘ bested Seeietee
oe a on eon ot , ' Ef Wea
fy Sado ert ut “ ho A
Sor ay Poteet ks ‘ , oe coer f
' ma MY ire ae | . ar ‘
por | fick iy ne ran
wey ‘ 1
ae er \ bay te ie F