Auf, Hh 6,0! Pa aah PON BES a a:
Md i ie 4
iy '
<A if |
‘i aa ? “oO ee
oa\ ; aS | 344 ? be evil et ioe ast? | :
a i
fh
‘ul bP a)
Ne Bets pA ee 4 Ee i al
é3 Ty ror
=a | a —- en 3 ad rc)
é Sioa “et ea ‘
ha ed a Lana ey ares y AME
e
Sic
an - Sp,
=
“eo ‘
2 N Ye “age iiss os,
Seite
Steerer
| | ie f & eee
eset sate, | a ? Sees crows Q
nora, EY
e 4 1 |
ray
as
Fs
rye, Ice ‘o, es 0 an LF “e i)
AES Bie EEE i : Ng
a hs “ob yam w a fi
Pe: fas Si
aes
alessio
1m a
| . We
<p %,
a
“fg ws = 4)
x 2) ) G | xe Say er a ay : ~, Ne ce ee ad
or a i 7. 1% f4 | en
24N,
a? a! _
"i
Al | oe oad a j 3 ‘A
a
=
9 parce 7 SE - wa” 7 »# oe XO PA ei |
he
= “ot
Ss, ‘ T =. or ae é Li Me @ cca iA
EE A | 4 i
= as,
oe
Agere
. & Geeta’ ty Y A “¥ wy ae a ete ay ° 8 % a4 - i, “i S 4 & ni a ts, Py c: a i ca Vs, nad g 5 ae e ‘jl Ne - Sh, ; (s fisties t i Aj \ i ia , ge % is ant “ey Vy
ah, nor
Ree ye
.
ae |
i
er 1
iM aS ‘ 25, os cs . Me
i i
‘ ceety <— H eos alk’ WY if Ga _%/ ae |! ‘“ A eet 2a
. CBs AN ape #: S bs Jg nae
oe, cr Dee J 4) Ss Vy, i soak eee a)
|
= fig gas teas Pe es ae eis,
Age
4 ie ye
“a i yy “%
EY i | a a? Sy cy
ns oe
Ih Se “yy, h i re
is me
if, ote Mi - eat ;
oo : i A - th mee ry : 4 : _" = OT ry, it > i
7
* y fr ' i} ’ . + : af = : + phy r i th “ ih A ! 7 ron " me J) 4 ) : a i 7 it : 7 ' i - ity) ii hy “a O : ee Jobe, S ae & yoo Ad tf a 1 is Uy ja a)
4
MADRONO
A WEST AMERICAN JOURNAL OF BOTANY VOLUME XVII
1963-1964
BOARD OF EDITORS
EpGAR ANDERSON, Missouri Botanical Garden, St. Louis LyMAN BENSON, Pomona College, Claremont, California HERBERT F. COPELAND, Sacramento College, California JouN F. Davipson, University of Nebraksa, Lincoln WALLACE R. ErRNsT, Smithsonian Institution, Washington, D.C. HERBERT L. Mason, University of California, Berkeley Mitprep E. Martuias, University of California, Los Angeles ROBERT ORNDUFF, University of California, Berkeley MARION OWNBEY, Washington State University, Pullman REED C. ROLLINS, Gray Herbarium, Harvard University Tra L. WiccINs, Stanford University, Stanford, California
Editor—JoHn H. THOMAS
Dudley Herbarium, Stanford University, Stanford, California
BUSINESS MANAGER AND TREASURER Douglas M. Post Biology Department, San Francisco State College
1600 Holloway Avenue, San Francisco 27, California
Published quarterly by the California Botanical Society, Inc.
2016 Life Sciences Building, University of California, Berkeley
Printed by Gillick Printing, Inc., Berkeley, California
To IRA LOREN WIGGINS. It is with great pleasure and satisfaction that the California Botanical Society dedicates this seventeenth volume of MaproNo to you. We have long benefited by your wise counsel, both during your presidency of the Society and subsequently during your many years as a member of the Council. Your studies of plants of western North America have spanned the latitudes from the Arctic to the Tropics; from these studies have come two impressive works: A Flora of the Alas- kan Arctic Slope and the monumental portion, Flora of the Sonoran Desert, of the Vegetation and Flora of the Sonoran Desert. And it is not only botanists who have been influenced by you. As Director of The Arctic Research Laboratory at Point Barrow for four years, fellow scientists of all kinds—zoologists, meteorologists, ecologists—benefited from their association with you. Too, you have been instrumental in the activities of the California Academy of Sciences, having served as its president, as a member of its board of trustees, and you have taken an active part in its expeditions. Last but not least, there is that host of students, many of whom under your versatile guidance really saw for the first time the flowers, the birds, the beasts and the rocks of the out-of- doors. They return through the years to Stanford University, and go to your office door hopefully to have a visit with their favorite professor. They are always welcome and go away content, their affection renewed.
ni ue a
— ee |
CONTENTS
PAGE
Frontispiece: Ira Loren Wiggins Juglans hindsii, the central California black walnut, native or introduced?
EOE TC VUCMIL GRHLOTIUS CMU cacti ame Re he wn gn dcs acct Se uaa tees ioueieacb te Oe 1 A controlled hybrid between Sitanion hystrix and Agropyron trachycau-
JR) aes a 2 10) (2 aaa Ate ek nla nso Ween aes ca Me eens. a ee 10 Cytotaxonomic observations on Mentzelia, sect. Bartonia (Loasaceae),
CHIN ee LILO P SOM ete ares escheat Ae ee eae re et, en OL en yh. abc ce 16 A revision of the genus Thaxterogaster, Rolf Singer and
PAVE IU Cheol Ls ONUUL IE tee ets ae cet oc Nes R ee ale SS Ose RO 22 Cytophyletic analysis of Hymenoxys anthemoides, Bernice M. Speese and
OLAS (ULC TOLILE Jig eae 2 iran td Neon 2h RnB aula co 1 See ieee 35 1 OY ale a a7 ] SRE SRE i a eB tr oe Oe ere ee 30, 66, 89, 140, 170, 233 IN@teS- amGdeING WS: 22. 2). eee ence rid cet S25 055 Oily ds a7. 14s oe
197, 203, 235, 268, 280, 294 Quaternary closed-cone pine flora from travertine near Little Sur, Califor-
nia, Jean H. Langenheim and J. Wyatt Durham ............22...0....22220000--+- 33 Chromosome numbers of some phytogeographically interesting Chilean POLS VL IV O07 Cnt sera. tan vetoes Aas, tence id it aledone seees tot dast ouch a2
Chromosome counts in section Erythranthe of the genus Mimulus (Scrophulariaceae). II., Robert K. Vickery, Jr., Barid B. Mukherjee,
AMC CV OCT C WY CGIIS texte est ec Aetna oe aster caren Bont eee. Meee ee eee 6) An analysis of variation in Viola nephrophylla, Norman H. Russell and TET S) @ OU OSS 10 IUUCR Peace cg eee eh hes tee sage ees 56
A new species of Isopyrum endemic to the Queen Charlotte Islands of British Columbia and its relation to other species in the genus,
erAt eC GIO CraaTN Gales 1a Ie ON VO) ce coat secu seery sean ee cass nee eee 69 Cytophyletic analysis of Hymenoxys odorata: A recapitulation, B. L.
LUE SARA LO ae A NORAD MOET PIE CORI TRRE SDE PENA? OR SNE SES A PE TED OTS Vd A note on taxonomic characters in Lolium, Frank C. Vasek and J. Kirk
Gg CLUS 1 pane Ocak Pir matin set oe etal an P ce ree Ja Oh ger ge Recess ea 79 Cleistogamy in the Malvaceae, Paul A. Fryxell .........-22....ccc-2ecceneecennnceennees 83 Swallenia, a new name for the California genus Ectosperma (Gramineae),
Thomas R. Soderstrom and Henry F. Decker ............22222222.2000000000000---- 88 Some cordilleran plant species new for the Sierra Nevada of California,
i IVUIGHOTAANG (Se AN DOMLDETS x0 eee ee eee 2 ee 93 Natural and artificial hybrids of Besseya and Synthyris (Scrophularia-
ceae), A. R. Kruckeberg and F. L. Hed glin .....0....2.0.222cccceceecceeeeee eee 109 Documented chromosome numbers of plants........00...22000.220--e2e-eee ee 116, 266 Artificial intergeneric hybrids of Helianthus and Viguiera, Charles B.
15 OSGI P=R | Ran See iy BPC are 2 eee Or Ze RPP eM IR See ca ee tT aR 118
Chromosome numbers in the Compositae. VII. Additional species from the southwestern United States and Mexico, A. M. Powell and B. L.
BEIRUT get ane Pa ee Oe ach eta ec kT ses hese pes toe eee se ae ae 128 lantscollection in Nepal, D:D: BHGUt cc... ccc eee eek ee et 145 Cytological studies in the genus Ficus. III. Chromosome numbers in sixty-
ENV ONSDECICSNI 1.0.) SCONAIE 2 oe. acs. cccstee teeta Been oe ae 153
Chromosome counts in the section Simiolus of the genus Mimulus (Scrophulariaceae). VI. New Numbers in M. guttatus, M. tigrinus, and M. glabratus, M. M. Mia, B. B. Mukherjee, and R. K. Vickery,
By tage ee a Bo Sr ON a ene eh 208d ce ren esl ewan cscs. eee eels 156 Extended dormancy of chaparral shrubs during severe drought, R. A. ENOTD CAIN EL Ae VE OON CY ace 80 avon oct scat soos a0ha Bel Maazel 161
Cytological observations on some genera of the Agavaceae, Marion S. Oi) ge ae ey SRE ar Sg ee, NON ts 8 an cue PA oe 163
Cytotaxonomy and distributional ecology of western North American vio-
lets; Jems Clausen. 053 oe ee eee er ED ee 173 Nomenclatural problems in the Acacia cornigera complex, Velua E. Rudd 198 Notes on the leaf epidermis and chromosome number of Swallenia (Gra-
mineae) @Dennts Andersons 2: | ee eee eee ee ee 201 The pollen grain morphology of Collomia as a taxonomic tool, Alfred R.
Loeolich, LL 255 ee ee eee era) oe ote Re ee 205 A™new species of pine irom Mexico; son Larsen 5 ee 217, Two new species related to Clarkia unguiculata, Frank C. Vase ................ 219 Taxonomic notes on the Chrysothamnus viscidiflorus complex (Astereae,
Compositae),, ordan- ©: Ard CrsOn = rst, eee ee 222 David Douglas and the digger pine: Some Questions, James R. Griffin...... 227), Bark photosynthesis in ocotillo, H. A. Mooney and B. R. Strain ................ 230
The genus Xerocomus Quelet in northern California, Harry D. Thiers..... 237 Survival of transplanted Cupressus and Pinus after thirteen years in Men-
docino County, California, Calvin McMillan ..0..22...22....22.cc22cccecceeeenees 250 A peculiar case of hemlock mistletoe parasitic on larch, Job Kuajt ............ 254 Lyonothamnoxylon from the lower Pliocene of western Nevada, Virginia
VE TPO SC: tote. Becht u Paces Pade Sa eet os Nines eae ieee eet 251 The Hordeum jubatum — caespitosum — brachyantherum complex in
Alaska, WoW. Maichell anda CO Wilione. 22 ee 269 The genus Eschscholzia in the South Coast Ranges of California, Wallace
| AGS Dy 2X) aaa BE AER OO OE De PR PPee od tet Mninty Seceey ete ear ge ee iy og iar rit 281
WiGlex: cos oe ee ey ec ee ge a ee 296
Borie |
ew VOLUME 17, NUMBER 1 JANUARY, 1963
Contents
PAGE JUGLANS HINDSII, THE CENTRAL CALIFORNIA BLACK WALNUT, NATIVE OR INTRODUCED? Harriette H. Thomsen ae 1
A CONTROLLED HYBRID BETWEEN SITANION HYSTRIX AND AGROPYRON TRACHYCAULUM, W.S. Boyle 10
CYTOTAXONOMIC OBSERVATIONS ON MENTZELIA, SECT. BaARTONIA (LOASACEAE), Henry J. Thompson 16
A REVISION OF THE GENUS THAXTEROGASTER, Rolf Singer and Alexander H. Smith 22
CYTOPHYLETIC ANALYSIS OF HYMENOXYS ANTHEMOIDES, Bernice M. Speese and J. T. Baldwin, Jr. 27
Reviews: Richard Evans Schultes, Native orchids of Trinidad and Tobago (Myron Kimnach); Albert N. Steward, La Rea Dennis, and Helen M. Gilkey,
Aquatic plants of the Pacific Northwest with vegeta- tive keys (Robert Ornduff); Ira L. Wiggins and John Hunter Thomas, A Flora of the Alaskan Arctic Slope (S. Galen Smith); Margaret McKenny, The savory wild mushroom. A Pacific Northwest guide (Isabelle Tavares) 30
NoTES AND NEws 32
A WEST AMERICAN JOURNAL OF BOTANY
PUBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY
MADRONO
A WEST AMERICAN JOURNAL OF BOTANY
Entered as second-class matter at the post office at Berkeley, California, January 29,
1954, under the Act of Congress of March 3, 1879. Established 1916. Subscription price
$6.00 per year. Published quarterly and issued from the office of Madrofio, Herbarium, Life Sciences Building, University of California, Berkeley 4, California.
BOARD OF EDITORS
HERBERT L. MAson, University of California, Berkeley, Chairman EpcAR ANDERSON, Missouri Botanical Garden, St. Louis LyMAN BENSON, Pomona College, Claremont, California
HERBERT F.. COPELAND, Sacramento College, Sacramento, California
Joun F. DaAvipson, University of Nebraska, Lincoln MItpreD E. MaArTurias, University of California, Los Angeles 24 MaArIon OWNBEY, State College of Washington, Pullman REED C. Rotiins, Gray Herbarium, Harvard University Ira L. Wicctns, Stanford University, Stanford, California
Secretary, Editorial Board—ANNETTA CARTER Department of Botany, University of California, Berkeley
Business Manager and Treasurer—Douglas M. Post Biology Department, San Francisco State College 1600 Holloway Avenue, San Francisco 27, California
CALIFORNIA BOTANICAL SOCIETY, INC.
President: Herbert L. Mason, Department of Botany, University of California, Berkeley. First Vice-President: Paul C. Silva, Department of Botany, University of California, Berkeley. Second Vice-President: Robert F. Hoover, California State Polytechnic College, San Luis Obispo. Recording Secretary: Mary L. Bowerman, Department of Botany, University of California, Berkeley. Corresponding Secretary, Margaret Bergseng, Department of Botany, University of California, Berkeley. Treasurer: Douglas M. Post, Biology Department, San Francisco State College, San
Francisco, California.
JUGLANS HINDSII, THE CENTRAL CALIFORNIA BLACK WALNUT, NATIVE OR INTRODUCED?
HARRIETTE H. THOMSEN
Many anthropologists have noted that a distinctive vegetation tends to appear on man’s habitation sites. Hrdlicka (1945) quoted Eyerdam as stating that the site of stone age villages in the Aleutians could gener- ally be recognized during spring and summer by the predominance of two perennial plants, Heracleum lanatum Michx. (wild rhubarb or cow parsnip), and Aconitum kamtschaticum Rchb. (monkshood). Lillard, Heizer and Fenenga (1939) state that nettles and thistles were often as- sociated with village sites in the Sacramento Valley. A more recent obser- vation was made by Elsasser (1960) that ‘““mule-ears” (Wyethia mollis Gray) sharply defined the extent of a California Indian habitation site in the Sierra foothills.
The association of certain plants and man has an obvious practical use in site reconnaisance; a two-year study made by Zeiner (1946) at the Angel Mounds in Indiana demonstrated that the location of buried walls and earthworks could be traced by the distribution of certain species of plants. Although of less immediate application, it would seem that con- sideration should also be given to the possibility that a useful chronology might be established on the basis of plant progression. In either case, such study would necessarily involve the differentiation of the plant cover into categories of indigenous and introduced flora.
The investigator into the relationship of vegetation to man’s occupa- tion in California would pretty surely turn to the publications of Willis Linn Jepson, whose ‘“‘A Flora of California,” “The Silva of California,” and “A Manual of the Flowering Plants of California,’ have been the primers for several generations of botanists. Jepson (1909, p. 365; 1910, p. 194) made the observation that the central California black walnut was to be found near ancient Indian village sites, although little signifi- cance was attached to his observation at the time.
An anthropological study of known Indian sites in the Bay Area coun- ties was initiated in an effort to ascertain if the black walnut is, in fact, a plant associated with pre-contact habitations of the California Indian. In the course of the investigation it became evident that on the basis of present-day evidence, definite proof is lacking as to whether or not the black walnut of central California is indigenous or introduced. The wide- spread use of black walnut as a rootstock for the commercial propagation of the English walnut (Juglans regia L.) has proliferated the occurrence of the black walnut to such an extent that no Indian association with a
given stand can be verified or disproved without a determination of its origin.
Maprono, Vol. 17, No. 1, pp. 1-32. January 30, 1963.
2 MADRONO [Vol. 17
The anthropological study is continuing; questions which should still be answered include: were the riparian locations that were so often occu- pied advantageous to both the trees and to the indigenous peoples, hence the shade and the fruits were utilized fortuitously? Since so good and portable a source of food must have been transported during the seasonal migrations of the Indians, is the presence of walnuts at non-riparian loca- tions the result of accidental seeding by man or by rodents? Or is their location in sites well above permanent water courses the result of natural distribution and an indigenous origin, the present isolated groves of trees representing relictual stands?
In order to attempt an answer to the possibility of natural origin as contrasted to introduction by Indians, the writer was led into a study of the fossil record, with the result that a case is made herein for the distri- bution of the central California black walnut, Juglans hindsii (Jeps.) Jeps., as an indigenous plant of central California.
Six to nine genera are recognized by taxonomists dealing with Jug- landaceae; the present investigation will be concerned with three of these: Carya, Pterocarya, and Juglans. Of the three sections in the genus Juglans (sections Rkysocaryon, Cardiocaryon and Dioscaryon), the discussion will deal primarily with section Rhysocaryon, since all of the western North American species of Juglans can be assigned to this section. Wolfe (1959, p. 13) considers that morphologically,
“There are two distinct groups of species in R/ysocaryon. One of these groups, including J. nigra and the Central and South American species tends to have leaflets which have a broad base and a broadly triangular (ovate) shape. ... The other group of species have teeth which are broad- ly triangular and dentate, or narrowly conical and serrate. The shape is narrowly triangular although a narrowly quadrate condition prevails in J. californica’. Wolfe’s work is mainly concerned with foliar character- istics, and minor attention is given to fossil fruits. An essential basis for differentiation between the extant species of Juglans which are either in- troduced or occur naturally in California lies in the fruits. Unlike its eastern cousin, the rugose J. nigra L., which has a strongly grooved black shell, or the important commercial so-called English walnut, J. regia L., the central California black walnut, J. Aindsi (Jeps.) Jeps., has a char- acteristically smooth or very lightly-grooved nut, light brown in color. In shape, the fruits of both J. migra and J. regia are characteristically longer and more pointed than in J. Aindsii. Specimens of J. Aindsu fruits taken in widely separated parts of the central California habitat show a consistent roundness, with a flattened stem end, reminiscent of the tiny Eocene J. clarnensis Scott. Foliar characters differ among the three, especially in the number of leaflets: J. Aindsiz has as many as nineteen leaflets and seldom fewer than fifteen; J. migra may have as many as twenty-seven leaflets, while J. vegza usually has seven, but sometimes five.
There are two centers of distribution for California black walnuts, one
1963 | THOMSEN: JUGLANS 3
in the southern part of the state and one in the north-central part, the trees occurring without connecting localities. The northern California black walnut was discovered along the lower Sacramento River area by Richard Brinsley Hinds of the Sulphur Expedition in 1837. By an his- torical accident, however, the black walnut of southern California was described first as J. californica by Watson (1875), who included within his concept J. rupestris Engl. var. major Torr., the walnut of Arizona and Texas. Jepson (1908) named the northern California trees as a variety of the southern ones, J. californica var. hinds, honoring their discoverer and saying that the trees of the two regions “‘differ somewhat.”
In his Flora, Jepson (1909) still contented himself with this varietal concept, averring that the northern trees were introduced from southern California by the trading of the native Indian tribes. In his Silva (1910), he continued the same concept but pointed out certain anomalies, namely that (1) Watson’s original description of J. californica did not cite a type and included as a synonym the walnut of Arizona and Texas, J. rupestris var. major, and (2) the southern, rather than the northern California walnuts must bear the name J. californica Wats. for several reasons: Wat- son had no northern specimen before him, his northern locality was vague, his description better fitted the southern form. It was not so easy, how- ever, for Jepson to ignore a further anomaly, the peculiar gap in distri- bution between southern and northern California (275 miles between stations in Ventura County and near Mount Diablo), and he concluded again that the introduction from the south by native tribes was the only plausible explanation for the existence of the northern trees. Later, how- ever, in his Manual (1923) he elevated the northern trees to the rank of a species, without further discussion, as J. hindsii (Jeps.) Jeps., a tacit admission that he must have come to the point of view that the original black walnuts of central California, present before the white man arrived, were indigenous and were not introduced from those of southern Califor- nia. His morphological treatment bears this out.
A review of the paleontological literature of the past half century may give perspective to the facts of distribution of California Juglans. Since the fossil beds are distributed in space as well as in time, the discussion of these deposits and their relation to Juglans can be more easily followed if they are examined by geographical groupings as well as by geological time periods. A glance at a relief map of the western United States dis- closes that territorially the area is broken up into clearly defined regions by the Coast, the Cascade and the Sierra Nevada mountain ranges; these ranges run roughly parallel longitudinally. Minor ranges, the Klamath to the northwest, and the Warner to the northeast, complete the California picture. One major transmontane range, the Tehachapi Mountains, shuts off northern from southern California, forming the southern rim of the great Central Valley, whose rivers converge from north, east and south towards San Francisco Bay.
4 MADRONO [Vol. 17
Some paleontological literature has been available since Jepson first became interested in the problem of Juglans and its distribution. Sud- worth (1908) remarked on the presence of fossil walnut remains in Cre- taceous and Tertiary formations, noting that “. ... in the northern Pa- cific coast region signs of ancient walnuts have been obtained from the Eocene formation, as well as from the gold-bearing gravel beds of the California Sierra’’. There follows a brief and very simplified discussion of the subsequent paleontological literature which bears on the subject.
The fossil floras of California, Oregon and Washington indicate tropi- cal conditions for the Eocene, although changes in physiography due to volcanic action and mountain uplift were making for localized climates. Scott (1954) described the fossil fruits and seeds from the Eocene Clarno formation of north-central Oregon. He discussed the similarity of the nuts of Carya and Juglans, designating the fruits of the Clarno species, Jug- lans clarnensis, and describing them as having features intermediate be- tween those of sections Cardiocaryon and Rhysocaryon. He says: “. . This is the first certain walnut described from rocks as old as Eocene. This occurrence of Juglans demonstrates that the fruits of the walnut and the hickory (Carya) were generically distinct during Eocene time’”’. It should be remarked here, however, that Wolfe (1959, p. 43) assigns J. clarnensis Scott to the Oligocene.
In the Oligocene, tropical fauna were still ranging up to Puget Sound, and the fossil flora of southern Alaska indicates a warm, temperate zone at that latitude (Smith, 1919). In other parts of North America, such as in North Dakota and Colorado, fossil beds containing Juglans material have been related to Oligocene time, and Wolfe (1959, p. 46) places J. kentensis, from a bed in northwest Oregon, in that time period.
MacGinitie (1937, p. 112) did an extensive study of the Weaverville beds, which he assigned to the Oligocene. The beds lie in the Klamath Range of northwestern California, clustered near the Trinity River, hemmed in between ocean and the rising mountains to the east. The dif- ficulties inherent in the identification of fossil material are well illustrated in the case of the Weaverville Juglans. Diller (1911) quotes Knowlton as having made determination of Juglans schimperi Lx. [reinterpreted by MacGinitie (1937) to Inga lancifolia MacG.]|, Juglans egregia Lx. and J. oregoniana Lx. from the Weaverville flora. MacGinitie reassigned both of the latter specimens to Juglans orientalis. He characterized his ex- panded interpretation of J. orientalis MacG. as having a much closer resemblance to various living species of Juglans in eastern Asia than to living forms in North America. LaMotte (1936) working in the upper Cedarville flora of the northern Great Basin (east of the Cascades and north of the Sierra Nevada), transferred Juglans oregoniana Lx. to Carya with what appears to be fairly cogent argument.
During an investigation of the Pliocene San Pablo beds, Condit (1938) identified fossil leaves from those beds as J. oregoniana Lx., defending
1963] THOMSEN: JUGLANS 5
his identification vigorously in opposition to LaMotte’s reassignment of this species to Carya. As far as the records of these four investigations show, the workers had leaves or leaflets only, and no fruits. But whether Juglans or Carya, the species described by MacGinitie as J. orientalis MacG. disappears from the North American scene, apparently unable to cope with the environmental changes of succeeding ages. No descendants are recognized in California today.
An obvious geographical unit in the western United States is that which centers in the Cascades. This mountain range, extending north-south from the Canadian border across Washington and Oregon, penetrates northern California for a brief distance, curving off toward the east and culminating at Mount Lassen, thus lying entirely north and west of the Sierra Nevada. This is an area in which extensive paleontological studies have been made. Chaney carried out investigations of the fossil floras of three Pliocene sites in the Cascades: Deschutes (1938), Dalles and Trout- dale (1944 a,b); he reported no Juglans from his studies of the Troutdale and Dalles flora, but described a species in another genus of the same family, Pterocarya oregoniana Chaney. He placed it in the East Asian Element, and listed P. stenoptera DC. as its nearest living equivalent species. The extant P. stenoptera, common in parts of China, is not gen- erically represented in North America today. Chaney’s study of the Deschutes flora (1938, p. 203) disclosed no evidence of either Juglans or Pterocarya.
Axelrod first studied the Sonoma flora in 1944 and later (1950a) he issued a report on “A Sonoma Florule from Napa, California”’, in which he identified Pterocarya oregoniana Chaney. In his digest of geological history which is included in Munz and Keck’s ‘“‘California Flora” (1959, p. 6) Axelrod stated that Pterocarya, together with a few other East American and East Asian species persisted in the mild coastal strip from central California northward, becoming extinct in the early part of the Pleistocene.
The wide area east of the Cascades is rather better documented than is the area to the west. This eastern region, tenanted by plant immigrants from the holarctic forest, provided a favorable environment for the de- velopment of a significant Miocene flora. LaMotte (1936, p. 90) believed that the general outlines of North America in upper Miocene time may be considered to have been much the same as today, the physiography and the inferred climate providing conditions favorable to an extremely uniform fossil flora. He reports the finding of numerous leaflets (1936, p. 116), with at least two specimens of leaves showing leaflets in place, but identifies the entire group as Carya egregia Lx., feeling that the char- acteristics resembling hickory outweighed those resembling Juglans. What seems important here is not that the botanists disagree on nomen- clature, but rather that the evidence for a dominant broad-leafed decid- uous forest of oak-hickory-beech association thrived east of the Cascades
6 MADRONO [Volal7
in Miocene time, an environment that would have been favorable to the persistence of the many species of Juglans found in Washington, Oregon and Idaho in preceding and contemporary fossil beds. Berry’s work (1927) on the flora of the Esmeralda beds convinced him that all evi- dence pointed to the presence of a permanent body of water in western Nevada in the Miocene.
Wolfe (1959, pp. 46, 47, 48)! describes two fossil Juglans from the Miocene: J. hesperia from northwestern Oregon and J. fryi from the northeastern part of the state. Both of these fossil forms had resemblances to J. kentensis referred to in the discussion of Oligocene specimens, but also exhibited what he considered to be significant morphological differ- ences. Chaney (1927, p. 74) identified materials from eastern Oregon as J. oregoniana Lx., and MacGinitie (1933, p. 50) identified material from southern Oregon as the same species, but Chaney considered the nearest living equivalent to be J. californica Wats., whereas MacGinitie consid- ered his specimen to be closer to J. migra L.
Two more references to Miocene Juglans appear in the literature. The first is J. nevadensis Berry (1928), exact provenience unknown, but found in the desert east of Truckee. Berry believed it to have a possible affinity to J. regia or J. sieboldiana Maxim of Japan, basing such evaluation on the lack of corrugation in the shell surface. The second is from Axelrod (1950b) who in his restudy of some of the Middle Pliocene Mount Eden flora, amended his description of J. beaumontiu Axelrod to include two species: a) J. beaumont Axelrod emend. with an affinity to J. rupestris Engelm. of the southwest United States and b) J. nevadensis Berry, de- scribed as having a smooth outer shell-coat with minor surface irregu- larities and greatest affinity to “J. californica of southern California”’ (p. 102).
Dorf (1936) investigated the Weiser beds of southwestern Idaho. This site, east of the rising Cascades, is placed in the Upper Miocene or Lower Pliocene time. Dorf imputed a more xeric environment with a rainfall of probably less than twenty-five inches. He described J. hesperia Knowl- ton which he related to J. nigra L., pointing out the similarity of the in- ferred Weiser climate to the climate of today in the eastern United States.
It would appear, therefore, that already in Miocene time, some Juglans species were established on the east side of the Sierra Nevada. Chaney (1938) concluded that the northern Sierra was sufficiently high at the beginning of the Miocene to have eliminated most of the genera requiring summer rainfall. Isolated there by the rising range and cut off from ocean moisture, the conditions became increasingly xeric, and the genus faced the inevitable choices which changing environments present to the plant world: adapt to change, migrate, or become extinct.
We come now to a consideration of the third geographical unit, the
1Juglans hesperia Knowlton has been transferred to Salix; Juglans hesperia Wolfe and J. fryi Wolfe (as well as J. kentensis Wolfe) are unpublished.
1963 | THOMSEN: JUGLANS 7
great Central Valley which lies south of the Cascades, east of the Coast Range, and west of the Sierra Nevada, closed off at the south by the Tehachapi Mountains. A number of fossil localities have been studied in the Coast Range and in the western foothills of the Sierras (see distribu- tion map, Chaney, 1944c). The Chalk Bluffs, La Porte and Oakdale beds have shown no material identified as Juglans, although the Chalk Bluffs flora did include specimens identified by MacGinitie (1941, p. 101) as Carya sessilis.
With Condit’s studies of the Remington Hill and Table Mountain floras (1944a, b), we reach the southern limit of fossil Juglans reported within the Central Valley. The Remington Hill and Table Mountain floras occur in, to quote a favorite phrase of the early paleontologists, “‘the auriferous gravels of the Sierras”. The Sierras of the Pliocene were yet of moderate height and the Table Mountain beds, located in a region drained by the Tertiary equivalents of the Merced and Stanislaus Rivers, may have been laid down at an elevation not more than five hundred feet above the level of a sea which lapped at the foothills—a sea which was a hundred miles or so farther inland than it is today. Condit reports that the Remington Hill beds are at an altitude of 3840 feet today, which would imp'y that they were laid down at a greater elevation than the beds of Table Mountain.
Condit (1944a, p. 42) described a fruit from the Remington Hill beds which he named Juglans pseudomorpha and thought might be related to J. nigra, but the association in the fossil flora suggested that it was inter- mediate between the typical eastern black walnut and the living Cali- fornia black walnuts, “J. californica of Southern California as well as J. hindsii of the inner north coast range bordering on the Sacramento Valley” (p. 28). Wolfe (1959, p.49) was of the opinion that Condit’s specimen is undoubtedly a Juglans, but that it is too badly crushed to permit comparisons.
Condit’s study of the Table Mountain flora (1944b) included a re- naming of Lesquereux’s Rhus typhinoides to Carya tvphinoides. He dis- cussed the similarities of Carya and Juglans at length, and considered the possibility that on ecological grounds C. typhinoides should be as- signed to Juglans rather than to Carya.
Wolfe (1959, p. 47) reassigned the Table Mountain leaflets to Juglans, giving them the name J. tuolumnensis. He was of the opinion that this fossil might have been derived through J. kesperia Wolfe, mentioned in the discussion of Miocene flora of northwestern Oregon, but he saw a more obvious relationship with J. californica Wats.
Fossil records for the Pleistocene in western North America are so sparse as to be almost non-existent. One allusion in the literature is made by Axelrod (1944, p. 118), wherein he refers to “..... the occurrence of a characteristic fruit of /uglans californica Watson in the Pleistocene of the San Joaquin Valley (H. L. Mason, oral communication, July,
8 MADRONO LNVoleaty/
1940)”. Verification of Pleistocene remains, perhaps through fossil walnut pollens from the San Joaquin Delta peats, would greatly reinforce our understanding of pre-Holocene distribution.
To summarize the fossil record:
(1) Members of the Juglandaceae, such as Juglans orientalis MacG. and Pterocarva oregoniana Chaney, which flourished in the western parts of California in the Oligocene and Pliocene, respectively, appear to have become extinct not later than early Pleistocene; as far as is known the only living equivalents are now found in Asia.
(2) The rise of the Cascades and of the Sierra Nevada in the Miocene isolated species of Juglans into westward and eastward slope environ- ments; across Oregon and Idaho and down the eastward side of the Sier- ras there appears to have been a migration of Juglans tolerating more xeric conditions, the plants developing a low, bushy conformity with foliage and fruit showing tendencies toward desert-shrub.
(3) By the Pliocene there were on the west side of the Sierra Nevada several species of Juglans that had made the adaptation from the sub- tropical environments of the Eocene and were flourishing in the cooler, summer-dry climate of the Pliocene at elevations now comparable to the lower Sierra Nevada. They occurred in the vicinity of the Tertiary rivers, probably not too high above the then great inland sea which reached to the foothills of the Sierra. It does not seem to be merely a coincidence that one finds today isolated stands of great trees at an altitude of around 1500 feet, on both sides of the Central Valley.
(4) As the Pleistocene ice advanced and the temperature lowered, the waters receded and the outline of the Central Valley was indicated. Three courses were open to the flora: it could stand steadfast and become ex- tinct, it could adapt itself to the environmental change, or it could mi- grate toward the south or toward the coast. It has been generally accepted that portions of the California flora migrated southward, advancing and retreating with the deterioration of the Tertiary climate. This assumption should be generally valid. Specifically, however, the record would indicate that it was not necessarily true of every genus. In the matter of Juglans, it would seem that there may have been a separation as early as Miocene time, with the result that J. californica Wats. was derived perhaps through J. nevadensis Berry, J. rupestris Engelm., J. beaumontiu Axelrod emend., or other similar small-leaved, small-fruited species, which tolerated the more xeric environment of the southern parts of the United States, where- as the ancestors of J. hindsiui (Jeps.) Jeps., may well have advanced coast- ward before the cold, following down the streams of the Central Valley, never approaching the coast beyond the edge of the fog-drip belt, retain- ing its mesic characters of lofty height, lush foliage, and smooth, faintly- grooved flattened nut.
It can be conciuded, therefore, that the climate and physiography were such as to make possible, or even probable, the independent evolution of
1963 | THOMSEN: JUGLANS 9
the southern California black walnut, J. californica Wats. and the north- ern species, J. Aindsii (Jeps.) Jeps., neither species being derived from the other. As a corollary, if the theory of separate and indigenous species is borne out by competent botanical investigation, the “anomaly” of dis- tribution raised so frequently in the past, does not exist; the presence of J. hindsii in north-central California assumes an aspect of rightness and logic, and requires no Indian trade to account for its narrow distribution in the Central Valley drainage basin and its focus in the San Francisco Bay area. Berkeley, California.
LITERATURE CITED
AXE LRoD, D. I. 1944. The Mulholland Flora, Carnegie Inst. Publ. 553:103-146. . 1950a. A Sonoma Florule from Napa, California. Carnegie Inst. Publ. 590: 23-71. ———-—. 1950b. Further studies of the Mount Eden flora, Southern California. Carnegie Inst. Publ. 590:75-117. Berry, E.W. 1927. The flora of the Esmeralda formation. Proc. U.S. Nat. Mus.
72 23,.1-15, — . 1928. A petrified walnut from the Miocene of Nevada. Jour. Wash. Acad. 18:6, 158-160.
CHANEY, R.W. 1927. Geology and paleontology of the Crooked River Basin, with special reference to the Bridge Creek flora. Carnegie Inst. Publ. 346:45-138. ——, 1938. The Deschutes flora of eastern Oregon. Carnegie Inst. Publ. 476:
185-216. ——-——. 1944a. The Dalles flora. Carnegie Inst. Publ. 553:285-321. 1944b. The Troutdale flora. Carnegie Inst. Publ. 553:323-351. 1944c. Pliocene floras of California and Oregon. Carnegie Inst. Publ. 533:frontispiece.
Conpit, C. 1938. The San Pablo flora of west-central California. Carnegie Inst. Publ. 476:217-268.
1944a. The Remington Hill flora. Carnegie Inst. Publ. 553:21-56.
—. 1944b. The Table Mountain flora. Carnegie Inst. Publ. 553:57-90. DiterR, J. S. 1911. The auriferous gravels of the Trinity River basin, California. Contr. to Economic Geology, Dept. of Int. U.S. Geol. Sur. Bull. 470:11-29. Dorr, E. 1936. A late Tertiary flora from southwestern Idaho. Carnegie Inst. Publ.
476:73-124.
ExsAsser, A.B. 1960. The archaeology of the Sierra Nevada in California and Nevada. Univ. Calif. Arch. Surv. 51:1-93.
Hrouicka, A. 1945. The Aleutian and Commander Islands and their inhabitants. The Smithsonian Inst., published by ‘ The Wistar Inst. of Anatomy and Biol- ogy, Phila. 43.
Jepson, W. L. 1908. The distribution of Juglans californica Wats. Bull. S. Calif. Acad. Sci. 7:23-24.
—————. 1909. Juglans, im Flora of California 1(2) :365.
1910. The silva of California. Univ. Calif. Memoirs 2:192-195. 1923. Juglans, 7x A manual of the flowering plants of California, p. 279. Assoc. Students Store, Univ. Calif.
Know ton, F. H. (as quoted by J. S. Diller). 1911. U.S. Geol. Sur. Bull. 470:23-24.
La Morte, R. S. 1936. The upper Cedarville flora of northwestern Nevada and adjacent California. Carnegie Inst. Publ. 455:57-142.
Litiarp, J. B., R. F. Herzer and F. Fenenca. 1939. An introduction to the arche- ology of Central California. Sacramento Jun. Coll. Bull. 2:65.
10 MADRONO [Vol. 17
MacGrniTtrz, H. D. 1933. The Trout Creek flora of southeastern Oregon. Carnegie Inst. Publ. 416:21-68.
1937. The flora of the Weaverville beds of Trinity County, California. Carnegie Inst. Publ. 465:85-151.
. 1941. A Middle Eocene flora from the Central Sierra Nevada. Carnegie Inst. Publ. 534:1-178.
Muwz, P. A., in collaboration with David D. Keck. 1959. A California flora. Univ. of Calif. Press, Berkeley and Los Angeles.
Scott, R. A. 1954. Fossil fruits and seeds from the Eocene Clarno formation of Oregon. Palaeontographica, Band 96, Abt. B: 66-97.
SMITH, J. P. 1919. Climatic relations of the Tertiary and Quaternary faunas of the California region. Proc. Calif. Acad. 4th Ser. IX:4, 123-173.
SupworTH, G. B. 1908. Forest Trees of the Pacific Slope. U.S. Dept. Agr.
Watson, S. 1875. Juglans californica Wats., in Revision of the genus Ceanothus, and description of new plants. Proc. Am. Acad. 10:333-350.
Wo re, J.A. 1959. Tertiary Juglandaceae of western North America. Submitted in partial satisfaction of the requirements for degree of Master of Arts in Pale- ontology, Univ. Calif. 1-96.
ZEINER, H. M. 1946. Botanical Survey of the Angel Mounds Site, Evansville, Indi- ana. Am. Jour. Bot. 33:83-90.
A CONTROLLED HYBRID BETWEEN SITANION HYSTRIX AND AGROPYRON TRACHYCAULUM
W.S. BoYLe
Students of evolution have become increasingly aware of the extensive hybridization that exists between genera and species in the grass family, particularly the tribe Hordeae. Probably no other family in the plant kingdom is destined to undergo such a fundamental revision of concepts of genetic relationships among genera as is occurring gradually in the Gramineae.
The present paper reports the meiotic chromosome behavior, fertility, and comparative morphology of a controlled hybrid between Sztanion hystrix (Nutt.) J. G. Smith and Agropyron trachycaulum (Link) Malte.
MATERIALS AND METHODS
Specimens of A. trachycaulum growing in fields near Logan, Utah, and those of Sitanion hystrix from Mantua, Utah, were transplanted to a field nursery in 1954. The crosses were made the following year.
Forty florets involving several spikes of S. Aystrix were hand-emascu- lated in June. Mature culms of Agropyron trachycaulum were placed in bottles of water and the bottles taped to stakes driven in the ground beside the culms of Sztanion hystrix. Each culm of S. Aystrix, with its adjacent pollinators, was then covered with Kraft paper sacks.
Two seeds were harvested in early August and planted later that same month in the greenhouse. The plants grew vigorously and a few spikes were produced the following year. One plant proved simply to be a selfed
1963 | BOYLE: HYBRID GRASS al
S. hystrix, but the other was indeed the hybrid. This was divided and repotted.
From 1959-1960 the plants grew well in the greenhouse but the pro- duction of flowering culms was sporadic. It was concluded that the high summer temperatures in the greenhouse interfered with normal anthesis; therefore the five clones were transplanted to the field nursery. In the field, they grew very well and flowered normally (fig. 2). The importance of field observations was emphasized by the contrast between plants grown in the greenhouse and the same plants grown in the field nursery. In addi- tion to much greater height, the plants grown in the field possessed inflor- escences larger in all respects than those of plants grown in the greenhouse.
B
Fic. 1. Meiosis in the hybrid: A, metaphase I, pollen mother cell, 12 II, 1 IV (x 1100); B, telophase II, pollen mother cell, lagging chromosomes (x 1315).
Numerous spikes were removed in the boot stage and fixed in New- comer’s solution (1953). Observations and photographs were taken from acetocarmine smears in temporary mounts.
MEIOTIC CHROMOSOME BEHAVIOR
METAPHASE I. Chromosome associations in the 204 pollen mother cells that were interpreted are summarized in Table 1. Although 24 different types of chromosome association were observed, approximately 75 per cent fell into the following categories: 14 II (23.5 per cent); 12 II, 11V (Zoro per cent): 13 1102 1 (l4./ percent); 11 11, 1 IV, 2 1 (iz7 per cent). Figure 1A is representative. Bivalents were present in all cells and averaged 12.2 per cell. Univalents averaged 1.13 per cell with a range of 0-6. Over half the cells possessed one or more quadrivalents, with an
12 MADRONO [Viole17
average of 0.6 per cell and a range of 0-3. Three trivalents and one hexa- valent were observed.
TELOPHASE I, II. Approximately a third of the 847 cells examined at telophase I contained one or more lagging chromosomes (Table 2). They averaged 0.6 per cell and had a range of 0-6. Nearly half of the 1756 cells examined at telophase II contained one or more lagging chromo- somes, with an average of 0.78 and a range of O-7 (fig. 1B).
Fic. 2. The hybrid in the field (x 146). Fic. 3. Spikes: A, Sitanion hystrix; B, the hybrid; C, Agropyron trachycaulum (* %).
Tetraps. Micronuclei averaged 0.37 per cell in the 2361 pollen grains examined at the tetrad stage. This is equivalent to 1.43 micronuclei per pollen mother cell. Approximately 70 per cent were without micronuclei. The maximum number of micronuclei in any one pollen grain was 3.
PoLueNn. The pollen was almost entirely abortive. Of the 9064 mature pollen grains examined, only 3 appeared to be fertile.
FERTILITY
In a careful search of approximately 2500 florets, not a single seed was found. The hybrid is completely sterile.
COMPARATIVE MORPHOLOGY The parental species are both highly variable, as the long synonomy lists suggest. However, obvious general differences exist between the two parents with respect to the morphology of the inflorescence. Except in size relationships, the hybrid is approximately intermediate between the two parents (figs. 3, 4).
1963] BOYLE: HYBRID GRASS 13
TABLE 1. CHROMOSOME ASSOCIATION AT METAPHASE I.
I II II IV VI Number cells
14 47 12 1 47
2 13 30
2 11 1 26
4 i did 10 Z 11
4 IK@) 1 ‘| i) 1 5
Z 9 2 3
3 11 1 Z
1 13 1 1
2 12 1
6 11 1
5 10 1 1 Ww 2 1
5 10 1 1
4 9 1 1
2 7 3 1 11 1 ]
4 11 1 11 1 1
5 10 1 1 13 1
1 12 1 1
Average 1.13 ay 0.19 0.60 0.004 204 per cell Total No.
In Agropyron trachycaulum the spikelets occur singly at the nodes, the rachis does not disarticulate, the glumes are very broad, and both glumes and lemmas are awnless. In Sitanion hystrix the spikelets usually occur in pairs, the rachis readily disarticulates at maturity, the glumes are very narrow, and both glumes and lemmas terminate in long, frequently twisted awns.
In the hybrid the spikelets occur singly at the nodes, the rachis tardily disarticulates at maturity, the glumes are moderately wide, and both glumes and lemmas have fairly short (1'4 cm.), diverging awns. The hybrid is a vigorous, tall bunchgrass apparently possessing considerable hybrid vigor.
DISCUSSION
The average number of univalents per pollen mother cell approximates the average number of micronuclei per pollen mother cell. The univalents would be expected to lag at first and second telophase and doubtless are the source of the micronuclei. The average number of laggards reported per cell at telophase I and II, however, was much lower than expected. A plausible explanation for this discrepancy could be our overly conserva-
14 MADRONO [Vol. 17
tive tendency in identifying laggards. Only chromosomes in the center of the plates were stipulated as laggards (fig. 1B). Very likely other chromosomes that approached the poles nevertheless failed to be included in the new nucleus and remained outside to form micronuclei.
The absence of quadrivalents in cytological studies of the parents tends to suggest that both are allotetraploids, and that the pairing in the hybrid is allosyndetic. Therefore these two species may have more or less homol- ogous genomes. On the other hand, Wagenaar (1959) has convincingly demonstrated autosyndesis among Hordeum chromosomes in a hybrid between H. jubatum L. and Secale cereale L. No quadrivalents were ob- served in the Hordeum parent. Dewey (1961) has conclusively demon- strated autosyndesis among chromosomes of Agropyron repens (L.) Beauv. and A. desertorum (Fisch.) Schult. in a hybrid between them. Stebbins and Pun (1953) similarly demonstrated that chromosomes of A. intermedium (Host.) Beauv. paired autosyndetically in a hybrid be- tween that species and S. cereale.
TABLE 2. FREQUENCY OF LAGGING CHROMOSOMES AT TELOPHASE I AND TELOPHASE II.
Average number of Percent with one Number
laggards per cell or mcre laggards of cells Telophase I 0.60 34.1 847 Telophase II 0.78 48.5 1/56
Since the writer cannot distinguish between the parental chromosomes in the hybrid, the precise nature of the pairing obviously cannot be in- ferred with confidence. The high frequency of quadrivalents, however, permits some speculation on homologies between these species. Over 50 per cent of the metaphase I plates possessed one or more quadrivalents. It is most unlikely that reciprocal translocation is responsible for any substantial percentage of these quadrivalents since they are absent in approximately half the cells and occur in variable numbers of the cells that do contain them. Even if it is conceded that the bivalent pairing may be exclusively autosyndetic, which is by no means established, the high number of quadrivalents suggests that important homologies exist be- tween Agropyron trachycaulum and Sitanion hystrix. The degree of ho- mology cannot be accurately assessed at present, but these two species unquestionably are much more closely related than their present taxo- nomic status indicates.
Stebbins et al. (1946) reported a controlled hybrid between Sztanion hystrix and a species of Agropyron (“San Benito”) believed to be most closely related to A. parishii Scribn. & Smith but originally identified as A, trachycaulum. The bivalent frequency (12.7 per cell average) closely approximates that of the hybrid reported in this paper. The frequency of univalents, laggards, and micronuclei in Stebbins’ hybrid, however, was much higher and the quadrivalent frequency was much lower than those
1963 | BOYLE: HYBRID GRASS 15
A B C
Fic. 4. Spikelets: A, Sitanion hystrix; B, the hybrid; C, Agropyron trachycau- lum (X 2).
found in the hybrid of this study. In the Stebbins’ contribution the authors suggested that A. saundersi (Vasey) Hitchc. is probably an F, hybrid between A. trachycaulum and Sitanion hystrix. The present study lends some, but not complete, support to this proposition. The type speci- men of Agropyron saundersi has considerably longer awns and shorter spikes than the controlled hybrid produced in the present study. Fur- thermore, in the type specimen of A. saundersti the spikelets are frequent- ly paired, in contrast to the single spikelets of the hybrid of the present study. Professor Arthur H. Holmgren has made the very plausible sug- gestion (oral commun.) that Sitanion longifolium J. G. Smith has served as one parent of Agropyron saundersiu. This species (sometimes referred to Sitanion hystrix) has large, coarse spikes with long, straight awns, and could well account for the differences observed between A. saundersti and the hybrid of the present study. Sitanion longifolium is known to be pres- ent in the general region where the type of Agropyron saundersii was collected.
SUMMARY
A controlled hybrid between Sitanion hystrix and Agropyron trachy- caulum is reported. Although 24 different types of chromosome associa- tions were observed at metaphase I in 204 interpreted pollen mother cells, 75 per cent fell into the following four categories: 14 II; 12 II, 1 IV; 13 II, 2 1; 11 I, 1 IV, 2 I. Bivalents were present in all cells and aver- aged 12.2 per cell. Over half of the pollen mother cells contained one or more quadrivalents. Univalents averaged 1.13 per cell. Lagging chromo- somes averaged 0.60 per cell at telophase I and 0.78 at telophase IT.
16 MADRONO [Vol. 17
Micronuclei averaged 1.43 per pollen mother cell. The hybrid is com- pletely sterile.
Although it was not possible to distinguish between autosyndetic and allosyndetic pairing, the high frequency of quadrivalents in the hybrid suggests that important homologies exist between the parental species.
The hybrid is morphologically intermediate between the two parents except for size relationships. Some support is given to the suggestion that Agropyvron saundersu had a similar origin.
ACKNOWLEDGMENT
The author wishes to express his appreciation to Professor Arthur H. Holmgren for many helpful discussions and to Robert Webster and Mrs. William Allderdice, student assistants.
Utah State University Logan, Utah
LITERATURE CITED
Dewey, D.R. 1961. Hybrids between Agropyron repens and Agropyron desertorum. Jour. Hered. 52:13-21.
Newcomer, E. H. 1953. A new cytological and histological fixing fluid. Science TS 2VG1:
STEBBINS, G. L., and Func Tine Pun. 1953. Artificial and natural hybrids in the Gramineae, tribe Hordeae. VI. Chromosome pairing in Secale cereale « Agro- pyron intermedium and the problem of genome homologies in the Triticineae. Genetics 38:600-608.
STEBBINS, G. L., J. I. VALENCIA, and R. M. Varencia. 1946. Artificial and natural hybrids in the Gramineae tribe Hordeae I. Elymus, Sitanion and Agropyron. Am. J. Bot. 33:338-350.
WaceEnaar, E. B. 1959. Intergeneric hybrids between Hordeum jubatum and Secale cereale. Jour. Hered. 50:195-202.
CYTOTAXONOMIC OBSERVATIONS ON MENTZELIA, SECT. BARTONIA (LOASACEAE)
HENRY J. THompPson!
A previous investigation (Thompson & Lewis, 1955) stated that infor- mation about chromosome numbers in Mentzelia would be of great value in the formation of evolutionary and taxonomic concepts in the genus. Section Trachyphytum, represented by twenty populations of ten species, was shown to be a polyploid complex on the base x9, with diploids (n=9), tetraploids (n=18), hexaploids (n—27), and octaploids (n=36). On the other hand, polyploids were not found in section Bartonza, al- though only two species were examined cytologically: Mentzelia multi-
1 This work received support from a National Science Foundation research grant
(G18720). The manuscript received the benefit of comments from professors Harlan Lewis and Peter H. Raven.
1963 | THOMPSON: MENTZELIA Ly
flora with n=9 and M. laevicaulis with n=11. The present paper reports chromosome numbers of n=9, 10, and 11, from forty-eight populations representing eleven of the approximately twenty species of section Bar- tonia, and establishes that aneuploidy, rather than polyploidy, is a major feature of evolution in this section.
Section Bartonia is clearly distinct from all other sections of Mentzelza. In section Bartonia the seeds are flat and circular in outline, with the margin extended into a wing. The flowers are relatively large and the five petals grade into the stamens through petaloid staminodia or at least broad-filament stamens. The plants are rosette-perennials and flower in late spring or summer. The flowers open and shed pollen in the late after- noon or evening and are pollinated by bees or hawk moths. Most of the species occur in the Rocky Mountain area of the United States, but one, M. albescens, also occurs in Argentina and Chile. This delimitation of section Bartonia agrees with that of Urban and Gilg (1900) and also with the description of the section by Darlington (1934). Darlington, however, without discussion included M. torreyi and M.reflexa in section Bartonia, perhaps inadvertently, because she did not modify her description of the section to accommodate them. They are clearly more similar to species of other sections.
CHROMOSOME NUMBERS IN BARTONIA
The nomenclature under which the following chromosome numbers are reported largely follows the most recent monograph of the genus (Darlington, 1934). The characteristics of the species as understood in this study are given briefly so that the chromosome numbers are clearly associated with groups of natural populations. All observations of chrom- osomes were made in squashed microsporocytes, using procedures previ- ously described (Thompson, 1960). Voucher specimens are on file in the herbarium of the University of California, Los Angeles (LA).
MENTZELIA LAEVICAULIS (Dougl. ex Hook.) Torr. & Gray n=11 Thompson 3089, 3197, 5 miles northwest of Gardiner, Park County, Montana. Ne 1690, 35 miles northeast of Garrison, Millard County, Utah. 3195, 42 miles southwest of Ely, Nye County, Nevada. 3194, Montgomery Pass, Mineral County, Nevada.
Mentzelia laevicaulis is widespread west of the Rocky Mountains, crossing that range into Montana and Wyoming. It is very distinct, with flowers that have five yellow petals 4-6 cm. long. There are no stami- nodia, but the five outer stamens have slightly broadened filaments about 2 mm. wide. The bracts at the base of the capsule are linear to linear- lanceolate and entire or with up to four short, pinnate lobes. Over one hundred individuals from ten widely scattered populations (see also Thompson & Lewis, 1955) have been examined cytologically and all had eleven pairs of chromosomes with two of the pairs conspicuously larger than the other nine.
18 MADRONO [Vol. 17
MENTZELIA DECAPETALA (Pursh) Urban & Gilg n=11 Thompson 3198, 5 miles northwest of Gardiner, Park County, Montana. sy 3206, 1 mile northwest of Florence, Fremont County, Colorado. Ernst 746, east of Hennessey, Kingfisher County, Oklahoma. Thomtson 3098, 5 miles south of Raton, Colfax County, New Mexico. ie 2084, 25 miles southeast of Lubbock, Crosby County, Texas. Mentzelia decapetala is widespread east of the Rocky Mountains from Idaho and Alberta south to Texas. In habit the plants are much like M. laevicaulis, but the flowers have ten white petals (five petals and five staminodia ) 5—8 cm. long. The bracts are pectinate and borne on the sides of the capsule. None of the chromosomes of M. decapetala is as large as the two largest chromosomes in the M. laevicaulis genome. The chromo- some number of MW. decapetala was reported as n=11 by Hamel (1938), but no voucher specimen was cited. Mentzelia decapetala and M. laevi- caulis hybridize readily where they grow together near Gardiner, Mon- tana (Thompson 3086, 3087, and 3199). The hybrids are common, easily recognized, produce very little good pollen, and set no viable seeds.
MENZELIA NuDA (Pursh) Torr. & Gray n=10 Thompson 1800, 2 miles north of Big Springs, Deuel County, Nebraska. os 3097, 15 miles south of Fountain, Pueblo County, Colorado. mt 1676, 6 miles east of Walsenberg, Huerfano County, Colorado. a 2085, 25 miles southeast of Lubbock, Crosby County, Texas.
Mentzelia nuda occurs on the plains east of the Rocky Mountains from Montana to Texas. The plants are 5-10 dm. tall, with the older plants producing five or six strict stems from the caudex. The pubescence is dense and the leaves are the most scabrous in the section. The white petals and petaloid staminodia, about ten in number, grade into the stamens through numerous narrow staminodia. The bracts at the base of the cap- sule are laciniate.
The identity of MM. nuda has been a matter of some confusion. Pursh (1814, p. 328) based his description on plants grown from seed gathered by Nuttall in 1811 along the Great Bend of the Missouri River. Accord- ing to Pennell (1936, p. 14) this locality is between the White River and the present city of Pierre, South Dakota. Pursh (loc. cit.) says, ““This species has smaller flowers, and the leaves are not so glaucous as the fore- going |M. decapetala|; in every other respect the above description is applicable to the present one... ”’. Considering these remarks and the locality, the name can apply only to the species considered here. I con- sider M. stricta (Osterh.) Stevens ex Jeffs and Little a synonym. The Nuttall specimen cited by Darlington (1934, p. 163) is not the type of M. nuda; Pursh apparently never saw it. Most plants cited by Darlington (1934, p. 162-3) as M. nuda are referable to M. multiflora.
MENTZELIA STRICTISSIMA (Woot. & Standl.) Darl. n=10 Thompson 2081, 4 miles west of Fort Sumner, De Baca County, New Mexico. 2087, 26 miles southwest of Odessa, Ector County, Texas.
This species occurs in eastern New Mexico and western Texas, partic-
1963 | THOMPSON: MENTZELIA 19
ularly in the drainage of the Pecos River. It is most similar to M. nuda, differing only in its smaller flowers (petals less than 25 mm. long) and more entire bracts. Mentzelia strictissima is not known to occur sympa- trically with M. nuda, and these two taxa may be only geographical races.
MENTZELIA RUSBYI Woot. n—10 Thompson 1670, 2 miles north of Eagle Nest, Colfax County, New Mexico. i 3103, 18 miles southwest of Las Vegas, San Miguel County, New Mexico. 1665, 6 miles east of Flagstaff, Coconino County, Arizona.
The plants of the collections cited here are erect, not branched below, and the inflorescence is compact. The rosette leaves are linear, about 15 cm. long, 8 mm. wide, and dentate to nearly entire. The petals are 15-20 mm. long, white and often with a tinge of apricot at the apex. The bracts at the base of the capsule are pinnately lobed. Harrington (1954) has considered M. rusbyi and M. nuda to be only varietally distinct |Z. nuda (Pursh) Torr. & Gray var. rusbyi (Woot.) Harr.].
<6
MENTZELIA CHRYSANTHA Engelm. n=10 Thompson 3096, 1 mile north of Florence, Fremont County, Colorado. Mentzelia chrysantha was described from plants collected near Canyon City, Colorado, from the same vicinity as the collection cited here. The plants of this species have numerous lower branches, which are decum- bent, and a compact inflorescence. The petals are golden-yellow and the leaves are lanceolate and shallowly lobed. The bracts at the base of the cylindrical capsules are entire and the seeds are narrowly winged.
MENTZELIA SPECIOSA Osterh. n=10 Thompson 3093, 5 miles south of Castle Rock, Douglas County, Colorado.
The collection here referred to MW. speciosa fits the original description very well. The petals are golden-yellow, much like those of M. chrysan- tha, but the inflorescence is open, the stems reddish, not white, the leaves more deeply lobed, the seeds more broadly winged, and the lower branches are fewer and not decumbent. The name M. speciosa has not been used since Darlington (1934) placed it in synonomy under M. multiflora (Nutt.) Gray; however, the type of M. multiflora is from Santa Fe, New Mexico, and has straw-white petals. Furthermore, recent collections from the Santa Fe area that match the type of M. multiflora have nine pairs of chromosomes. Mentzelia speciosa is apparently restricted to the east slope of the Rocky Mountains of Colorado.
MENTZELIA DENSA Greene n—10 Thompson 1684, Cotopaxi, Arkansas River, Fremont County, Colorado.
The plants of the population studied are 3—4 dm. tall with many branches from the caudex. The petals are golden-yellow, 16—18 mm. long, and grade through staminodia into the stamens. The capsules are cylin- dric, tapered at the base, 10-13 mm. long, and subtended by a linear, entire bract. The leaves are 4—8 cm. long, pinnate, usually with six lobes,
20 MADRONO [Vot. 17
these 4 mm. long and 2 mm. wide, with the midrib 2 mm. wide. These plants agree very well with Greene’s (1896) original description. Dar- lington (1934) cited only three specimens under MW. densa, all from the Grand Junction area of western Colorado, although Greene (1896) stated, ‘Common in the Canon of the Arkansas in southern Colorado, and else- where among the foothills”’.
MENTZELIA LACINIATA (Rydb.) Darl. n=10 Thompson 1687, 11 miles west of Pagosa Springs, Archuleta County, Colorado. The plants of this population have golden-yellow petals and are gen- erally similar to M. densa but differ in having more deeply lobed, lacinate leaves and narrowly winged seeds. Mentzelia laciniata has been reported from southwestern Colorado and northwestern New Mexico.
MENTZELIA MULTIFLORA (Nutt.) Gray n=—9
Thompson 3095, 7 miles north of Penrose, Fremont County, Colorado. - 1675, 3 miles west of La Veta Pass Summit, Costilla County, Colo. i 3099, 8 miles south of Raton, Colfax County, New Mexico. " 3100, 18 miles southwest of Las Vegas, San Miguel County, New
Mexico.
ae 1672, 6 miles west of Red River, Taos County, New Mexico. s 1669, 7 miles east of Taos, Taos County, New Mexico. = 1668, 25 miles southwest of Taos, Taos County, New Mexico. i 1667, 38 miles west of Albuquerque, Bernalillo County, New Mexico. ce 2076, 15 miles northwest of Datil, Catron County, New Mexico. 2074, 4 miles east of Quemado, Catron County, New Mexico. 3107, 6 miles east of Pie Town, Catron County, New Mexico. be 2091, 5 miles west of Las Cruces, Dona Ana County, New Mexico. eB 2089, 6 miles east of Allamore, Hudspeth County, Texas. S 2058, 7 miles east of Pecos, Ward County, Texas. : 3212, 1 mile west of Sanders, Apache County, Arizona. 1066, Highway 66 at Painted Desert, Apache County, Arizona. y 3222, 3 miles west of Winkleman, Pinal County, Arizona. 2094, 21 miles northeast of Benson, Cochise County, Arizona. 1664, 10 miles southwest of Prescott, Yavapai County, Arizona. - 1662, 2 miles northeast of Congress Jct., Yavapai County, Arizona. = 2067, Salt River at Highway 60, Gila County, Arizona. 2066, 10 miles west of Globe, Gila County, Arizona.
Raven 14818, 13 miles north of Puerto Penasco, Sonora, Mexico.
Plants of the kind referred here occur in most of the western states and in northern Mexico and, as delimited in Darlington’s monograph (1934), they form the most variable complex in the section. All the plants in this group have petals 15-25 mm. long, yellow to nearly white, that grade through broad staminodia into the stamens. The capsules are cylin- dric to urceolate, 15-25 mm. long, usually with a linear, entire bract at the base and with seeds that are broadly winged. The leaves are pinnately lobed to dentate. Mentzelia multiflora (Nutt.) Gray obviously applies to this group of plants. The type was collected by Gambel near Santa Fe, New Mexico, and was described as having petals 34 inch long, straw- white in color. My collections from this region in New Mexico that agree with the original description have nine pairs of chromosomes. Mentzelia
1963 | THOMPSON: MENTZELIA 21
nuda (Pursh) Torr. & Gray is sometimes incorrectly applied to this group of plants.
Four of the collections cited here, Thompson 1664, 1662, 2067, and 2066, differ from the others in having the darkest yellow petals, the most urceolate capsules, and more entire leaves. Plants of this kind grow in the mountains of central Arizona between the southern desert and the Mogol- lon Rim.
MENZELIA MULTICAULIS (Osterh.) Darl. n=—I11 Thompson 3201, Dinosaur National Monument, Uinta County, Utah. a 3203, 8 miles west of Rio Blanco, Rio Blanco County, Colorado. x 3205, 1 mile north of Walcott, Eagle County, Colorado.
Mentzelia multicaulis occurs in northwestern Colorado and northeast- ern Utah. The flowers have five yellow petals, but the next inner whorl may be either five staminodia or five stamens with filaments nearly as broad as the petals. A distinctive characteristic of M. multicaulis is that either the petals, the staminodia, or both are very broad, 9 mm. long and 6 mm. wide. In this species the capsules are urceolate, 8 mm. long and subtended by a linear bract. The seeds have very narrow wings. The lower leaves are 4—6 cm. long, pinnatifid, the four to six lobes are nearly 1 cm. long and 2 mm. wide and the midrib is 3 mm. wide. The upper leaves are entire and linear. All of the plants observed were more than one year old and had several stems 1—3 dm. long from the old caudex. Mentzelia multi- caulis is most similar to M. humilis (Gray) Darl., but the latter differs in having the petals and staminodia short and narrow and seeds with broad wings.
Chromosome numbers have been reported for two other species of sec- tion Bartonia. Mentzelia albescens (Gill.) Griseb., the only species of section Bartonia in South America, was reported as n=11 by Covas and Schnack (1946) from material collected at Lujan, Provincia de Mendoza, Argentina. Hamel (1938) reported n=9 from root-tip sections of plants identifid by him as M. humilis. Hamel did not cite a voucher specimen and the identity of his material is unknown. The seeds for his culture were obtained from the New York Botanical Garden.
SUMMARY AND CONCLUSION
Two sections of Mentzelia, sections Bartonia and Trachyphytum, are morphologically more similar to each other than to any of the other sec- tions. However, their validity as distinct phylads within the genus is indi- cated not only by consistent morphological differences but also by con- trasting patterns of chromosomal change in the course of their evolution. Section Trachyphytum has only one basic chromosome number (x=9) although it comprises diploid, tetraploid, hexaploid, and octaploid species that form one large polyploid complex. Section Bartonia, on the other hand, has no known polyploid species, but three basic chromosome num- bers are present in the group; one highly variable species has n=9, seven
De MADRONO [Vol. 17
species have n= 10, and four species have n=11. The prevalence of poly- ploidy in one section and aneuploidy in the other undoubtedly reflects a fundamental difference in genetic systems and a corresponding phylo- genetic separation of the two sections.
Department of Botany, University of California, Los Angeles
LITERATURE CITED
Covas, G., and B. ScHNAcK. 1946. Numero de cromosomas en antofitas de la Region de Cuyo (Republica Argentina). Rev. Arg. Agron. 13:153-166.
DARLINGTON, JOSEPHINE. 1934. A monograph of the genus Mentzelia. Ann. Missouri Bot. Gard. 21:103-226.
GREENE, E. L. 1896. New or noteworthy species. Pittonia 3:99.
Hamet, J. L. 1938. Etude de la mitose somatique et numération chromosomique chez quelques Loasacées. Rev. Cyt. Cytophys. Vég. 3:153.
Harrincton, H. D. 1954. Manual of the plants of Colorado. Denver, Sage Books.
PENNELL, F.W. 1936. Travels and scientific collections of Thomas Nuttall. Bar- tonia 18:1-51.
PursH, F. 1814. Flora Americae Septentrionalis. Vol. I.
Tuompson, H. J. 1960. A genetic approach to the taxonomy of Mentzelia lindleyi and M. crocea (Loasaceae). Brittonia 12:81-93.
Tuompson, H.J., and H. Lewis. 1955. Chromosome numbers in Mentzelia (Loasa- ceae). Madrono 13:102-107.
Urpan, I., and E. Gitc. 1900. Monographia Loasacearum. Nova Acta Abh. Kaiserl. Leop.-Carol. Deutsch. Akad. Naturf. 76:22-97.
A REVISION OF THE GENUS THAXTEROGASTER SINGER RotF SINGER AND ALEXANDER H. SmitH!
Since the original publication of the genus Thaxterogaster (Singer, 1951) and the subsequent monograph (Singer & Smith, 1958), further studies have been made as time and available material permitted, and these have led to the discovery of additional species. Species are here grouped into sections, since certain distinct trends are now evident. Those species treated in detail in our monograph (doc. cit.) are not redescribed here, but detailed accounts of new or transferred taxa are treated critically.
We wish to acknowledge assistance from the National Science Founda- tion, which made possible the studies at the Royal Botanic Gardens, Kew, England. We also express our thanks to Dr. G. Taylor, Director of the latter institution, for the privilege of studying the collections there, and to Dr. Clark Rogerson, Curator, New York Botanical Garden, New York, for the opportunity to study the collections of Hymenogaster at that institution.
1 Papers from the Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina, and the University Herbarium and Depart- ment of Botany (Paper no. 1142), the University of Michigan, Ann Arbor, Michi- gan, U.S.A.
1963 | SINGER & SMITH: THAXTEROGASTER 23
KrEy TO SECTIONS OF THAXTEROGASTER
A. Clamp connections present on hyphae of the gastrocarp. B. Spores small (rarely up to 13 uw long), either smooth or with an adhering ver-
ruculose-rugulose exosporial ornamentation . . Sec. 4. Micros porogaster
B. Spores larger and always distinctly ornamented. C. Spores angular when immature, subglobose . . . . Sect. 2. Blestogaster C. Immature spores not angular... . . . . . Sect. 3. Thaxterogaster
A. Clamp connections absent. D. Spores with an apical beak and the wrinkled outer layer of the wall loosen-
ing to a considerable extent . . . . . . Sect. 1. Scabrogaster D. Spores lacking a distinct apical beak anil the exosporial ornamentation adher- ing to the inner wall eee a ee ect. SA poy porasce,
Sect. 1. Scabrogaster Singer & Smith, sect. nov.
Sporis subrostrato-mucronatis, ornamentatione subrugulosa separabili. Typus. Thaxterogaster scabrosum (Cooke & Massee) Smith & Singer.
1. Thaxterogaster scabrosum (Cooke & Massee) Smith & Singer, comb. nov. Secotium scabrosum Cooke & Massee, Grevillea 20:35. 1891.
Peridium hemispherical, depressed, dingy-olive or grayish, minutely scabrid. Gleba lacunose, septa gill-like, waved and folded, dark reddish brown. Stipe very short, almost obsolete.
Spores 15-18 (20) & 7.5—10 », dark rusty brown in KOH, near yellow- ish tawny in Melzer’s reagent; with a slightly wrinkled outer layer which tends to loosen variously, the plage area not differentiated; apex tapered to a blunt-pointed beak-like projection but this consisting entirely of wall material (no pore present).
No other microscopical details obtainable from type.
It is clear from the type, the original description, and Smith’s observa- tions on the spores that this species is a Thaxterogaster connecting that genus to the group of species in Hymenogaster sensu lato with spores hav- ing an outer wrinkled wall and an apical beak. This group in Hymeno- gaster makes up the largest element in that genus of diverse spore types.
Type studied. “On the ground. Domain, Melbourne (Baron Muel-
en > (Ke).
Sect. 2. Blestogaster Singer & Smith, sect. nov.
Sporis subglobosis, juventute plus minusve angulosis.
Typus. Thaxterogaster brevisporum Singer.
2. THAXTEROGASTER BREVISPORUM Singer, Persoonia 1:386. 1960.
We emphasize the angularity of the spores because this feature may be of some importance in connecting the species to members of Hymeno- gaster sensu lato.
Sect. 3. THAXTEROGASTER.
Spores ellipsoid, over 13 » long, never angular, not with a distinct apical beak, the outer layer of spore wall adherent to inner layer; hyphae hav- ing clamp connections at the septa.
24 MADRONO [Vol. 17
Key TO SPECIES OF SECTION THAXTEROGASTER
A. Peridium lacking a thick gelatinous epicutis. B. Peridium pallid, becoming brownish . . . . . . . 6. T. magellanicum B. Peridium violaceous. C. Spores mostly 15-18 » long; stipe poorly developed; growing under Notho-
fagus pumilo .. 1 + © = 6 Gene ae ee ee (on Euiolacemn C. Spores mostly 13-15 » long; stipe well developed; growing under Nothofagus dombeyi and Saxegothaea .......... . .. 4.7. dombeyti
A. Peridium with a thick gelatinous epicutis. D. Peridium white ftp i ad 3 SL GR 2A SN ie lencoce pial D. Peridium olive yellow to bister . . . ... .: =. =. =. «. 5. LT. pingue
3. THAXTEROGASTER VIOLACEUM Singer, Mycologia 43:216. 1951, and BrittonvalOe 20/1958.
4. THAXTEROGASTER DOMBEYI Singer, Persoonia 1:385. 1960.
5. THAXTEROGASTER PINGUE (Zeller) Singer & Smith, Brittonia 10: 211. 1958. Secotium pingue Zeller, Mycologia 33:211. 1941.
6. THAXTEROGASTER MAGELLANICUM Singer, Mycologia 43:219. 1951, and Brittonia 10:208. 1958.
7. THAXTEROGASTER LEUCOCEPHALUM (Massee) Singer & Smith, Brit- tonia 10:210. 1958. Secotium leucocephalum Massee, Grevillea 19:95. 1891.
The following data are taken from the holotype at Kew: Peridium duplex, the outer layer a thick gelatinous layer of appressed hyaline nar- row hyphae with clamp connections. Inner layer of floccose broader (5-12 ») hyphae. Tramal plates of interwoven hyphae yellowish in KOH (but not reviving well and apparently not gelatinous). Details of hyme- nium not obtained. Spores 12-16 & 7-8 mw, warty rugulose and rusty brown in KOH (as in Cortinarius).
These data on the holotype verify the characters of the species as pub- lished previously by us (loc. cit.).
Sect. 4. Microsporogaster Singer & Smith, sect. nov.
A sectione Thaxterogastero sporis minoribus, interdum sublevibus differt. Typus. Thaxterogaster subalbidum Smith.
Key TO SPECIES OF SECTION MICROSPOROGASTER
A. Spores almost smooth; stipe-columella solid . . .. . . 8.7. subalbidum A. Spores verruculose-rugulose; sterile base chambered, columella hollow (at least when dried) fe Ss gee Shas DAMEN ee aba cap as Om wae TIE
8. Thaxterogaster subalbidum Smith, sp. nov.
Sporis sublevibus; stipite columellaque solidus.
Typus. Thaxter, Fungi Hypogeus No. 4, March 5, 1906, Punta Arenas, Chile, South America (FH).
Gastrocarp 1-2 cm. broad (estimated on basis of dried material), ir- regularly globose to convex, the surface silky and white, drying whitish; eleba dingy cinnamon brown, of minute chambers nearly filled with
1963 | SINGER & SMITH: THAXTEROGASTER 23
spores; stipe-columella well developed, solid, percurrent, the surface pal- lid, the interior slightly darker, extending well below the lower margin of the peridium, the estimated length 1—-1.5 cm., width 5-7 mm., the peridium either connected to the stipe-columella or free in places to expose slightly the gleba.
Spores 10-13 6-8 uy, ellipsoid with a short-pointed, inconspicuous sterigmal appendage, the wall 1.3—2 » thick (as seen on fractured spores), the surface very minutely warty, wrinkled to punctate-roughened (almost smooth as seen under high dry), pale cinnamon in KOH; no apical dif- ferentiation.
Basidia 28-33 8-10 p, clavate, 4-spored, hyaline in KOH. Cystidia none. Basidioles resembling immature basidia. Tramal plates with an indistinct subhymenium of narrow branched elements intergrading with the central area which is of interwoven filaments, some with somewhat enlarged cells, hyaline in KOH. Epicutis of peridium a thick layer of appressed, narrow (2-3 »), hyaline, smooth, thin-walled hyphae; this layer grading imperceptibly with the context which has broader (4—7 pz) hyphae, some oleiferous hyphae, and is slightly yellowish in KOH; clamp connections present.
The nearly smooth spores separate Thaxterogaster subalbidum from the other known Thaxterogasters. In the material in the Zeller collec- tions (NY), a duplicate of the packet of specimens cited above is on the same sheet as the type of Hymenogaster fragilis. The description of the latter, however, obviously does not apply to ‘“‘Fungi Hypogeus No. 4,” which we here designate as the type of 7. subalbidum. Material of Hy- menogaster fragilis sent to Zeller, apparently by Thaxter, and in the Zeller collections, is marked with red crayon as an indication that it is the holotype of H. fragilis, but in the original description of that species the location of the “type” is given as the Farlow Herbarium. We do not know whether Zeller saw all the material or not. The holotype of T. subaibidum is that portion of “Fungi Hypogeus No. 4” deposited in the Farlow Herbarium.
9. Thaxterogaster fragile (Zeller & Dodge apud Dodge & Zeller) Smith & Singer comb. nov. Hymenogaster fragilis Zeller & Dodge apud Dodge & Zeller, Ann. Missouri Bot. Gard. 21:646. 1934.
Gastrocarp 1—2 cm. diam., subglobose to pear-shaped, the surface whit- ish or gray, drying pallid to pale cinnamon-buff; gleba chambered, fragile, the cavities fairly large for a fungus of that size and very irregular in shape, bright cinnamon as dried; stipe-columella consisting of a cham- bered sterile base, 6 mm. broad (reminding one of the marginate bulb in a Cortinarius), this extended into a hollow thin-walled percurrent colu- mella which when dried may be obliterated entirely except for the cavity; gleba attached to the length of the columella (if latter is still evident) ; peridium even, silky, extending to the stipe-columella (but possibly slight- ly free at times).
26 MADRONO [Vol. 17
Spores 10-12 7-8.5 », ovate-pointed and with a short, pointed sterig- mal appendage, dull cinnamon brown in KOH, thick-walled, warty- rugulose, the outer wrinkled layer not separating appreciably from the inner one; apical beak solid (lacking a pore).
Peridium with a surface layer of indefinite thickness of appressed fila- ments 3-6 « diam., hyaline in KOH, smooth, thin-walled and with medal- lion clamps at cross walls. This layer intergrading with the context which is composed of broader hyphae and some laticifers (?). Details of hyme- nium and tramal plates not evident.
Punta Arenas, Chile, March 1906. R. Thaxier. Word “type” under- lined in red on the packet in the Zeller collections (NY).
The description of Zeller and Dodge obviously applies to this material, which is distinct from the March 5, 1906, collection (i.e., Thaxterogaster subalbidum) by the brighter cinnamon, more fragile gleba, ovate-pointed to almost broadly fusoid conspicuously roughened spores, chambered sterile base and hollow columella.
Parks 2182, filed in the Zeller collections (NY) as Hymenogaster fragilis, is not the same as either of the Thaxter collections. The spores and the microscopic characters of the peridium are those of the type of Hymenogaster gilkeyae, but all the details of the hymenium and tramal plates had collapsed beyond the point of reviving. No clamps were pres- ent as far as could be observed.
Sect. 5. Aporpogaster Singer & Smith, sect. nov.
Hyphis defibulatis, sporis haud vel vix rostrato-apiculatus; strato externo sporarum haud separabili.
Typus. Thaxterogaster conicum (Hesler) Singer & Smith.
KEY TO SPECIES OF SECTION APORPOGASTER A. Gastrocarp globose, depressed; spores 7-llu broad . . . 10.7. porphyreum A. Gastrocarp typically conic and elongate; spores mostly 11-13 uw broad. 11. T. conicum
10. THAXTEROGASTER PORPHYREUM (Cunningham) Singer, Lilloa 26: 105. 1953; see also Brittonia 10:212. 1958. Secotium porphyreum Cun- ningham, Proc. Linn. Soc. N.S. W. 49:114. 1924.
11. THAXTEROGASTER CONICUM (Hesler) Singer & Smith, Brittonia 10: 214. 1958. Secotium conicum Hesler, Jour. Elisha Mitchell Soc. 49:153. 1933.
Facultad de Ciencias Exactas y Naturales Buenos Aires, Argentina
Department of Botany University of Michigan Ann Arbor, Michigan
LITERATURE CITED SINGER, R. 1951. Thaxterogaster—a new link between gasteromycetes and Agaricales. Mycologia 43:215-288. SINGER, R., and A. H. SmitH. 1958. Studies on secotiaceous fungi I: a monograph of the genus Thaxterogaster. Brittonia 10:201-216.
1963 | SPEESE & BALDWIN: HYMENOXYS ZY.
CYTOPHYLETIC ANALYSIS OF HYMENOXYS ANTHEMOIDES BERNICE M. SpEESE and J. T. BALDWIN, JR.
This paper reports the chromosomal number for Hymenoxys anthe- moides (Juss.) Cass., type species of the genus, and relates this datum to the chromosomal situation as known for Hymenoxys. The analysis is cyto- phyletic, in the original meaning of the word (Baldwin, 1939).
Parker (1950) published certain new combinations in Hymenoxys Cass. that she later (1951, in manuscript) used in a monograph of this genus. Earlier, she had asked us to survey the chromosomes of Hymen- oxys and had supplied us with thirty-five collections of seeds representa- tive of both subgenera, of fourteen of the twenty-four species recognized by her, and of five varieties.
Fic. 1. Specimens of Hymenoxys anthemoides grown from Argentine seed (Bald- win 15580). Six inch scale shown.
28 MADRONO [Vol. 17
Speese and Baldwin (1952) published the results of their studies of Hymenoxys: twelve species had 2n numbers of 30; H. acaulis (Pursh) Parker had 2n numbers of 60 in three varieties and of 30 in var. ivesiana Greene; H. odorata DC. had a 2n number of 22. Parker (1960) used chromosomal evidence and stem and leaf characters as bases for raising H. acaulis var. ivesiana to specific rank: H. ivestiana (Greene) K. F. Parker.
Jackson (1957) reported an n number of 15 for H. argentea (A. Gray) Parker and thus substantiated our count of 2n of 30 for the species. Tur- ner, Beaman, and Rock (1961) published n numbers of 15 for two species of Hymenoxys from Nuevo Leon, Mexico: A. insignis (A. Gray) CKIL., a species of tall biennials not investigated by us and most closely related to H. grandiflora (Torr. & Gray) Parker for which we had found that 2n=30; and H. odorata (Rock 264, TEX), annuals, for which we had reported a 2n number of 22 from New Mexico (Parker and McClintock 7009, US) and Arizona (K. Parker 7459, US). For plants of H. odorata from almost the same Arizona station (P. Raven 11731, UC), Raven and Kyhos (1961) also determined an n number of 11: they thus corroborate our count for the species. In addition to the count of n—=15 for the Nuevo Leon material of H. odorata (Rock 264) reported in the 1961 paper, Turner obtained a similar count (fide Johnston) for a second Nuevo Leon collection (M. C. Johnston 5860).
ee 0, MahOgeedpagagy
Z 2,
Fic. 2. Chromosomes of Hymenoxys anthemoides (the plants grown from Argen- tine seed): left, mitotic metaphase, 2n=30; right, metaphase I of meiosis, n=15. Chromosomes were drawn X 2200 and reduced by one-third in reproduction.
Dr. Parker (letter of August 31, 1961) wrote us that all five specimens cited above are typical H. odorata. It is clear that the chromosomal sit- uation in this species needs further study. The chromosome counts re- ported in Turner, Beaman, and Rock (1961) are from pollen-mother- cell smears of buds fixed in the field during the summer of 1959. Our ex- perience has been that preparations from material so fixed are often dif- ficult to interpret, and especially so if the weather were hot at the time of fixation.
Parker (1951) stated that H. odorata, a widely distributed annual in the Midwest and Southwest, is closely allied both to the Mexican H.
1963 | SPEESE & BALDWIN: HYMENOXYS 29
chrysanthemoides DC. and to the South American H. anthemoides ( Juss.) Cass., type species for the genus, but that these three annuals are quite distinct. Two other species are in South America: one is annual; the other, annual or biennial. Speese and Baldwin (1952) wrote: “If it should be discovered that the South American species—and especially the type species—fall into a chromosome series with H. odorata |2n=22: an annual] or into any series different from that evidenced by the ma- jority of Hymenoxys representatives as interpreted by Parker |2n=30 or 60: mostly perennials], reason would then be at hand to suspect the validity of Parker’s treatment.”
Parker, awaiting additional chromosomal data on Hymenoxys, has de- layed publication of her monograph, and both she and we have made a number of attempts to obtain viable seeds of South American species. Finally, on January 15, 1960, Dr. Arturo Burkart most kindly collected fruiting specimens of H. anthemoides and sent them to Dr. Parker, who placed a voucher specimen in the United States National Herbarium, Smithsonian Institution, and sent us plants with mature achenes. We grew seedlings (Baldwin 15580, US, fig. 1) and examined their chromo- somes: H. anthemoides has a 2n number of 30, an n-number of 15 (fig. 2).
In summary, the basic chromosome number of sixteen of twenty-five species (twenty of thirty-one taxa) accepted by Parker (1951, 1960) for Hymenoxys and including the type species is 15; the plants are either di- ploid or tetraploid. The basic number for H. odorata is 11. (We assume the report of n—15 for this species to be incorrect.) These numbers indi- cate phyletic trends. From chromosomal evidence alone, one would con- clude that H. odorata is wrongly placed in this species. The chromosomal data for the other species, however, lend support from one more discipline to Dr. Parker’s interpretation of Hymenoxys.
College of William and Mary, Williamsburg, Virginia.
LITERATURE CITED
BaLtpwin, J. T., JR. 1939. Certain cytophyletic relations of Crassulaceae. Chron. Bot. 5:415-417.
Jackson, R. C. 1957. Documented chromosome numbers of plants. Madrofo 14: 111-112.
PARKER, Kittie F. 1950. New combinations in Hymenoxys. Madrono 10:159. 1951. Monograph of the genus Hymenoxys Cass. Unpublished manuscript, pp. 1-112. 1960. Two species of Hymenoxys (Compositae) new for Arizona. Leafl. West. Bot. 9:92-93. Raven, P. H., and D. W. Kyuos. 1961. Chromosome numbers in Compositae: II. Helenieae. Am. Jour. Bot. 48:842-850. SPEESE, BERNICE M., and J. T. Batpwin, Jr. 1952. Chromosomes of Hymenoxys. Am. Jour. Bot. 39:685-688. Turner, B. L., J. H. BEAMan, and H. E. L. Rock. 1961. Chromosome numbers in the Compositae. V. Mexican and Guatemalan species. Rhodora 63:121-129.
30 MADRONO [Vol. 17
REVIEWS
Native Orchids of Trinidad and Tobago. By RicHarp Evans SCHULTES. 275 pp., 25 photographs, 74 drawings, 1 map. Pergamon Press, 1960. $15.
Until recently the determination of orchids has often proved a difficult task. The literature pertaining to the 20,000 or so species is naturally voluminous, much of it is widely scattered and difficult of access, and in consequence most identifications have to be carried out by specialists at such large orchid herbaria and libraries as those at Harvard University. Within the last decade, however, a number of floristic treat- ments of the Orchidaceae have appeared, including those for the United States, Mexico, Guatemala, Venezuela, Brazil, Peru, and Malaya; now the general botanist and amateur student are much better equipped to work with this fascinating family.
Dr. Schultes, Curator of the Botanical Museum of Harvard University, has pre- pared an excellent account of the orchids of Trinidad and its companion island, Tobago. The subject of this meticulously written and well-illustrated treatise is re- stricted to an area of less than 2,000 square miles, but the book will be of more than local interest because a large number of the species are found elsewhere in Latin America. Trinidad and Tobago are not true islands in the geologic sense, being non- volcanic in origin. Trinidad is only seven miles from the coast of Venezuela and endemism is low—7 per cent; as is well-known, on islands never connected with the mainland endemism is often very high (e.g., 75 per cent in Hawaii). The flora of Trinidad is also distinctly South American, rather than Caribbean, and this applies as well to the orchids.
Schultes recognizes 181 species of Orchidaceae, two less than are found in the Gramineae, the largest family on the islands. Included in each generic description are characters found in the entire geographic range, whereas the species descriptions apply only to plants from the two islands. Descriptions and citations of specimens are very complete, and there are also presented such useful data as the flowering times, the etymological origin of the generic names, and the citation of notes pub- lished elsewhere.
Most of the photographs are undistinguished, but are adequate for purposes of identification. Unfortunately, the one of Pleurothallis diffusa appears again on an- other page mislabeled Stelis muscifera. Quite outstanding are the drawings, many of which are reprinted from other recent floristic works on orchids. By such well- known artists as Gordon Dillon and G. Dunsterville, they add much to the value of the book.
This work can be especially recommended as an introduction to a representative selection of orchid taxa, many of which are widespread in the Americas——Mvyron KimnacH, Botanical Garden, University of California, Berkeley.
Aquatic Plants of the Pacific Northwest with Vegetative Keys. By (late) ALBERT N. Steward, La Rea Dennis, and HeLten M. Girxey. Illustrated. Oregon State College Studies in Botany, No. 11. Oregon State College Press, Corvallis. v + 184 pp. 1960. $2.50.
This manual was intended to cover “all true aquatics and also species whose life cycle includes some stage that requires either the saturation of the substrate with water or the presence of an ambient aqueous medium” occurring in Oregon, Wash- ington, British Columbia, and Alaska. The authors have not succeeded in doing this, but they have produced a manual with some merits which will be useful once its limitations are recognized.
The text covers hepatics, mosses, vascular cryptogams, and angiosperms. Species found both in moist and dry areas were to be excluded. Main emphasis in the keys is on vegetative characters, although reproductive structures are used either secon- darily, or when vegetative traits fail. Because of the generous ecological qualifications given above, one cannot argue with the inclusion of Alnus oregona, Urtica lyallii, Physocarpus capitatus, Oenothera hookeri, or Gaultheria shallon in this manual,
1963 | REVIEWS Spl
although they would not be considered aquatic plants by many botanists. The incorporation of such doubtful species should not be troublesome to users of the manual. It is wiser for a work covering plants of a specific habitat to be over- generous than to be parsimonious. In contrast, however, a number of genera with a legitimate claim to inclusion have been omitted. Among these are Elatine, Lindernia, Camassia, Habenaria, Porterella, Limnanthes, Limosella, Navarettia, Tillaea, Berula, Sphenosciadium, and Lilaeopsis. Some of these genera are widespread and are likely to be picked up by persons collecting in the wet lands of the Pacific Northwest. Furthermore, in some genera such as Hydrocotyle, Cicuta, Eryngium, and Lobelia, not all species growing in moist areas are given, which increases the risk of mis- identification in these genera. I have not found any genera or species represented in Alaska, British Columbia, or Washington but not in Oregon in this manual, suggesting that it will have limited use outside the state.
On the credit side, particularly noteworthy are the vegetative keys, especially to groups such as the Cyperaceae, which frequently do not flower for long periods of time and are difficult to identify when they do. Many of the vegetative distinc- tions between taxa will be of interest since these are often passed over in floras in faver of differences in the reproductive organs alone.
This manual cannot be recommended for use in identification unless it is used in conjunction with the other manuals available for Oregon and Washington. It has some serious drawbacks which could easily be eliminated in a second edition by rewording the title, the preface, and by expanding the coverage to include those important groups which have been omitted—RoBERT ORNDUFF, Department of Biology, Reed College, Portland.
A Flora of the Alaskan Arctic Slope. By Ira L. WicctIns and JOHN HUNTER THOMAS. vii + 425 pp. 1962. Univ. Toronto Press, Canada. $9.50.
This book is a comprehensive manual of vascular plants, including complete keys, descriptions of all taxa, and distribution maps. It is based largely on recent exten- sive collections by the authors and others. The field, herbarium, and library work was supported largely by the Office of Naval Research and the Arctic Institute of North America. The taxonomic treatment generally follows Hulten’s and Anderson’s floras of Alacka, but it includes some original taxonomic interpretations, takes into account recent taxonomic papers, and cites many more collection localities, including range extensions to the American Arctic for numerous species.
The authors have designed their book for use as a handbook for field identifica- tion as well as a technical reference work for professional taxonomists. The omission of synonymy and illustrations is unfortunate, as is the failure to cite distribution records from Hulten’s “Flora of Alaska” and from collections other than those listed by the authors. These deficiencies are more than balanced, however, by such features as complete keys and descriptions, helpful comments on taxonomic problems in many taxa, a glossary, a short bibliography, and detailed distribution maps ac- companied by a valuable gazetteer.
The habitat notes and the sections on the environment contribute toward the integration of taxonomy and ecology. The essay on “minor habitats” (7.e., those dis- turbed by man) emphasizes dramatically “the serious nature of comparatively minor disruptions of the normal balanced environmental and ecological conditions” in the Arctic. The vastness and remoteness of the Arctic tend to make people in general overlook this fact, but it is all too evident to the visitor to the extensively disturbed Point Barrow region. On the other hand, the American Arctic presents unsurpassed and largely unexplored opportunities for detailed study of the effects of ecological disturbance in a region where man has as yet effected only very local changes. The new flora of the Arctic Slope will be a valuable tool in the hands of biologists fol- lowing this as well as many other lines of investigation—S. GaLEN SmitTH, Depart- ment of Botany and Plant Pathology, Iowa State University, Ames.
32 MADRONO [Vol. 17
The Savory Wild Mushroom. A Pacific Northwest Guide. By Marcaret Mc- Kenny (with the collaboration of D. E. Stuntz). xiv + 133 pp., 1 text-figure, 33 black-and-white and 48 color photographs. University of Washington Press, Seattle. 1962. $3.95. Paperback.
One of the most welcome aspects of this attractive handbook of mushrooms is the inclusion of a brief, up-to-date chapter on mushroom poisons, written by Dr. Varro E. Tyler, Jr., Professor of Pharmacognosy, University of Washington. For this reason, this book should be of particular interest to physicians; it provides a con- venient source of information on types of poisons, symptoms, and treatment, as well as descriptions and illustrations of the most dangerous of the poisonous mushrooms. Although subtitled “A Pacific Northwest Guide,” this handbook should be of interest to any collector, because it includes a large number of more or less cosmopolitan species.
Following a brief introduction to the nature of mushrooms and instructions for their collection, the species are organized into three groups—edible mushrooms, poisonous mushrooms, and nonpoisonous mushrooms to be avoided. Most of the common edible and poisonous species are included. Enough precautions about col- lecting, determining, and eating are mentioned to make it possible to recommend this book, with no serious reservations, to the amateur collector as a reliable handbook. For the gourmet, there is a chapter of tantalizing recipes for cooking mushrooms.— ISABELLE TAVARES, Department of Botany, University of California, Berkeley.
NOTES AND NEWS
The year 1962 marked the passing of two noted botanists, John James Thornber, who died in Tucson,. Arizona, on November 22 at the age of ninety, and Joseph F. C. Rock, who died in Honolulu, Hawaii, on December 5 at the age of seventy-nine. Dr. Thornber, who joined the staff of the University of Arizona in 1901, was a bota- nist of international repute and was Dean of the College of Agriculture at the Uni- versity of Arizona from 1922 to 1928. His intimate knowledge of the Arizona flora has not been surpassed. Dr. Rock was Professor of Botany at the University of Hawaii from 1911 to 1919. He spent many years exploring in remote parts of south- western China. In addition to his contributions as a plant explorer, he was a noted linguist, cartographer, anthropologist, and ornithologist. Many of his plant discoveries were introduced into cultivation through the University of California Botanical Garden. His beautifully prepared specimens are deposited in herbaria throughout the world.
The Segundo Congreso Mexicano de Botanica, organized by the Sociedad Botdnica de México, will be held in San Luis Potosi from September 17 to 20, 1963. Queries concerning the Congress may be addressed to Biol. Fernando Medellin, Apartado Postal 458, San Luis Potosi, S.L.P., México. The highly successful first Congress was held in Mexico City in October, 1959. Among the 172 participants were 26 from the United States.
NOTICE TO CONTRIBUTORS. With my retirement from the University of California in June, 1963, I am also retiring from the Editorial Board of Madrono, a post that I have held for twenty-nine years. In anticipation of
this time, I am asking all contributors to send new manuscripts to Dr. John H. Thomas, Division of Systematic Biology, Stanford University, California. Dr. Thomas will take over the duties of the editorship beginning with the July, 1963, issue of Madroho.—HErBeErtT L. Mason, Chairman, Editorial Board.
INFORMATION FOR CONTRIBUTORS
Manuscripts submitted for publication should not exceed an estimated 20 pages when printed unless the author agree to bear the cost of the ad- ditional pages at the rate of $20 per page. Illustrative materials (includ- ing “typographically difficult” matter) in excess of 30 per cent for papers up to 10 pages and 20 per cent for longer papers are chargeable to the author. Subject to the approval of the Editorial Board, manuscripts may be published ahead of schedule, as additional pages to an issue, provided the author assume the complete cost of publication.
Shorter items, such as range extensions and other biological notes, will be published in condensed form with a suitable title under the general heading, ‘“Notes and News.”
Institutional abbreviations in specimen citations should follow Lanjouw and Stafleu’s list (Index Herbariorum. Part 1. The Herbaria of the World. Utrecht. Second Edition, 1954).
Articles may be submitted to any member of the Editorial Board.
Membership in the California Botanical Society is normally considered a requisite for publication in MADRONO.
MADRONO
A WEST AMERICAN JOURNAL OF BOTANY
A quarterly journal devoted to the publication of botanical re- search, observation, and history. Back volumes may be obtained from the Secretary at the following rates:
Volume I, 1916-1929 . . . . . $5.00 Volume IT, 1930-1934... . . . . » 5:00 Volume IIT, 1935-1936 Mae ee Bee ag 0.0) Volume IV, 1937-1938. . . . . 5.00 Volume V, 1939-1940 . . . . . 5,00 Volume VI, 1941-1942 ge genet hy ESTO Volume VII, 1943-1944 . . . . 5.00 Volume VIII, 1945-1946 . . . . 5.00 Volume IX, 1947-1948 oo Me al a 1GKOO Volume X, 1949-1950 . . . . . = 7.00 Volume XI, 1951-1952 rhe rae 2 3)4 0,0, Volume XII, 1953-1954 . . . . 8.00 Volume XIII, 1955-1956 . . . . 8.00 Volume XIV, 1957-1958 . . . . 8.00 Volume XV, 1959-1960 et a OOO Volume XVI, 1961-1962 . . . . 12.00
Single numbers:
Volumes: andll 2.0% Soe 50 Volumes II through XIV... .. 1.00 Volumes V" 5 3 “2s | heap ae 1.25 Volume XVI. (2 «vale eee 1.50
The subscription price of MaproNo is $6.00 per year. If your institution does not now subscribe to MADRONO, we would be grateful if you would make the necessary request. Since there is no extra charge for institutional subscriptions, an individual may hold membership in the California Botanical Society on the basis of his institution’s subscription.
Address all orders to:
Corresponding Secretary California Botanical Society Department of Botany University of California, Berkeley 4, California
ADRONO
VOLUME 17, NUMBER 2 APRIL, 1963
Contents
PAGE
QUATERNARY CLOSED-CONE PINE FLORA FROM TRAVER- TINE NEAR LITTLE Sur, CALIFORNIA, Jean H. Langen- heim and J. Wyatt Durham 33
CHROMOSOME NUMBERS OF SOME PHYTOGEOGRAPHICAL- LY INTERESTING CHILEAN Piants, D. M. Moore 52
CHROMOSOME COUNTS IN SECTION ERYTHRANTHE OF THE GENUS MIMULUS (SCROPHULARIACEAE). II., Robert K. Vickery, Jr., Barid B. Mukherjee and Delbert Wiens 53
AN ANALYSIS OF VARIATION IN VIOLA NEPHROPHYLLA, Norman H. Russell and Frank S. Crosswhite 56
Review: H. H. Allan, Flora of New Zealand (Robert Ornduff) 66
NOTES AND NEws: CHROMOSOME NUMBERS IN CROSSO- soMA, Peter H. Raven and Marion S. Cave 68
A WEST AMERICAN JOURNAL OF BOTANY
BLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY
MADRONO
A WEST AMERICAN JOURNAL OF BOTANY
Entered as second-class matter at the post office at Berkeley, California, January 29,
1954, under the Act of Congress of March 3, 1879. Established 1916. Subscription price
$6.00 per year. Published quarterly and issued from the office of Madrofio, Herbarium, Life Sciences Building, University of California, Berkeley 4, California.
BOARD OF EDITORS
HERBERT L. Mason, University of California, Berkeley, Chairman EpcAR ANDERSON, Missouri Botanical Garden, St. Louis LyMAN BENSON, Pomona College, Claremont, California
HERBERT F. COPELAND, Sacramento College, Sacramento, California
Joun F. Davinson, University of Nebraska, Lincoln MitpreD E. MArtuias, University of California, Los Angeles 24 MARION OWNBEY, State College of Washington, Pullman REED C. Rotiins, Gray Herbarium, Harvard University Ira L. Wiccrns, Stanford University, Stanford, California
Secretary, Editorial Board—ANNETTA CARTER Department of Botany, University of California, Berkeley
Business Manager and Treasurer—Douglas M. Post Biology Department, San Francisco State College 1600 Holloway Avenue, San Francisco 27, California
CALIFORNIA BOTANICAL SOCIETY, INC.
President: Herbert L. Mason, Department of Botany, University of California, Berkeley. First Vice-President: Paul C. Silva, Department of Botany, University of California, Berkeley. Second Vice-President: Robert F. Hoover, California State Polytechnic College, San Luis Obispo. Recording Secretary: Mary L. Bowerman, Department of Botany, University of California, Berkeley. Corresponding Secretary, Margaret Bergseng, Department of Botany, University of California, Berkeley. Treasurer: Douglas M. Post, Biology Department, San Francisco State College, San
Francisco, California.
Sn
1963 | LANGENHEIM: LITTLE SUR FLORA 33
QUATERNARY CLOSED-CONE PINE FLORA FROM TRAVERTINE NEAR LITTLE SUR, CALIFORNIA
JEAN H. LANGENHEIM AND J. Wyatt DURHAM
Interest in the relation of plants to deposition of calcareous charged water as travertine, tufa or sinter has existed at least since the time of Vergil and Pliny. Most of the studies have been concerned with currently active springs, and discussion has tended to center around the role of bacteria and algae, primarily, in producing deposition or at least in in- creasing precipitation of calcium carbonate from calcium bicarbonate (Agardh, 1827; Cohn, 1862; Weed, 1888; Meunier, 1899; Jones, 1914; Kellerman and Smith, 1914; Emig, 1917; et al.). In some cases other plant structures such as leaves, twigs, wood, etc., have been noted. The Little Sur travertine terraces in Monterey County, California, are inter- esting not only in terms of their formation and related geologic history, but also because the plant fossils give additional evidence regarding the distribution of the closed-cone pine forest along the central coast of Cali- fornia during the Pleistocene.
The travertine outcrop was discovered by Durham during a class field trip in October, 1958, and collections were made at that time as well as during a similar trip the following year. Additional collections and recon- naissance studies were made by both authors later in the second year. Durham is responsible for the geologic observations and discussion herein. Thanks are due to Robert Simmonds for running the amino acid tests on the gastropod shells and to Allyn B. Smith for identifying the shells. Likewise appreciation is expressed to H. L. Mason for criticizing the manuscript and to Jack Wolfe for checking the determination of fossils as well as for criticism of the manuscript. Most of the plant fossil speci- mens are located in the Paleobotanical Collections of the Department of Botany, University of Illinois. Duplicate specimens are deposited in The Museum of Paleontology, University of California at Berkeley. Voucher specimens for description of the modern vegetation are in the Herbarium of the Department of Botany, University of Illinois.
FORMATION OF TRAVERTINE
The term “travertine” has been used for the Little Sur deposits because it seems to be the most inclusive term for deposits of this type. Emig (1917) has pointed out that the use of terms associated with this type of deposit has been varied and confused. He uses the term “travertine” for deposits of white, gray or brown concretionary calcium carbonate with cavernous and irregularly banded structure, soft and chalk-like to hard and crystalline, often containing leaves, twigs and mosses. Emig likewise indicates that ‘“tufa” is a “more ancient term” for travertine which was used by Vergil and Pliny in the same sense that travertine is now used in
MaproNo, Vol. 17, No. 2, pp. 33-68. April 29, 1963.
34 MADRONO [Vol. 17
A ag
VEL perp |
\ ENS AOE! src : me : ; bass <a 2 i/
NWSW, @ Aas oe Ree ON: / \ \Wi7 lV Bod : eae 5 7m y \ ~ \ 1) 2 a rz ey ‘ avy : side
Fic. 1. Sketch of travertine terraces as seen from south. California State High- way 1 in foreground.
Italy. He considers tufa as a cellular variety of calcite that has been de- posited from calcareous springs around nuclei of algae, mosses, leaves, twigs and other plant structures. Another term which has indiscriminate- ly been interchanged with travertine by some is “calcareous sinter” (Weed, 1888). In its original German sense sinter means dross from iron and implies an initial process of heating before deposition takes place. Thus Emig thinks that deposits formed in the immediate vicinity of hot springs may be correctly called “‘sinter.”” Twenhofel (1950) indicates that calcareous deposits around springs are known as travertine, but distin- guishes tufa and sinter as porous and travertine as compact and banded.
The material from the Little Sur occurrence varies from a very soft chalky, cavernous rock to hard, finely crystalline concretions and bands. It varies in color from white to creamy to ochreous, and at times is stained with ferrous iron. Because of the proximity of the Serra Hill fault (Trask, 1926), it seems probable that the terraces were produced as a result of hot springs activity along this fault. Hot springs are also known to occur at the present time about twenty-one miles south of the area
1963] LANGENHEIM: LITTLE SUR FLORA 315)
along the Sur Thrust. Preservation of the leaves is often relatively poor, probably because of the effect of their having fallen into a hot spring. as well as the manner in which the calcium carbonate was precipitated. Fur- thermore, it is difficult to uncover the fossils in their entirety because they do not occur along a bedding plane and because frequently the margins of the leaves are curled under.
LITTLE SUR TRAVERTINE TERRACES
The terraces are located in the Santa Lucia Mountains, Monterey County, California. They are exposed along State Highway 1, nearly one mile north of the mouth of Little Sur River (lat. 36° 20.7’ N., long. 121° 53.45’ W.).
There are four prominent terraces on the east side of the road and on the south side of a prominent ravine (fig. 1), and there is seemingly at least one below the road on the cliffs extending to the ocean. There is con- siderable marble (Sur Series) cropping out of the hill slopes to the east of the presumed location of the Serra Hill Thrust fault (Trask, 1926) and also on the north side of the ravine. According to Trask’s map, the bedrock underlying the terrace deposits should be sediments of the Creta- ceous ‘‘Chico Group.” The south side of the present ravine on the north edge of the terraces has been cut into travertine which is 40—50 feet thick in places. It seems probable that the hot springs forming these deposits were in an old swale associated with the Serra Hill fault, and were located on an old surface prior to the development of the details of the present topography; the cascading pools were responsible for the sequence of terraces now observed. For convenience, the terraces have been numbered in sequence from the highway up the slope to the highest recognized, with the forefront of Terrace 1 being exposed in part in the road cuts. The apparent terrace below the road was observed from a distance only and is not included in the following descriptions.
Terrace 1 extends approximately 120 feet from north to south; Ter- race 3, which is smaller, extends 200 feet from north to south; Terrace 2 is a little over 150 feet in the same direction as is Terrace I. The frontal deposits of Terraces 1 and 2 are each at least 40—50 feet thick.
The thickest accumulation of travertine is near the south margin of the adjacent ravine, seemingly indicating that the center of accumulation was either at that point or near the axis of the present ravine. The de- posits thin to the south, but Terraces 3 and 4 extend from the ravine on the north to the gully to the south where their southern extent is trun- cated by the present topography. Terraces 1 and 2 do not extend that far south.
The lower part of the deposit consists of occasional pure layers and tongues of travertine intercalated with alluvial gravels. In the roadcut exposures and outcrops at the north end of Terrace 1, the upper 15—20 feet of sediments are much purer travertine than the lower part. The lower part of this interval has beds with concentrated masses of Cupressus
36 MADRONO [Vol. 17
branchlets as well as a few pine cones. Just below the Cupressus band there is a zone with abundant leaves of plants such as Ceanothus, Garrya, Rubus, Ribes, etc. Occasional masses of abundant Equisetum or leaves of the above plants are scattered in the exposures along the roadcut.
Much of the surface of the undisturbed exposures above the highway is covered with a dense caliche that probably conceals the fossil content. The best fossil collecting was in the roadcut and in the exposures at the north end of Terrace 1 on the south side of the ravine. On the upper ter- races the most abundant fossils observed were sedge-like leaves with scattered Garrya leaves. On the forefront of Terrace 2 clumps of sedges were noted in growth position, apparently indicating water flowing down- hill through a series of small pools.
The time of formation of the travertine terraces can not be determined directly at the present time. Trask (1926) considered the faulting in the area to have occurred at the end of the Pliocene, but this might indicate mid-Pleistocene in present-day terms. Topographically, the terraces ap- pear to be situated on the upper part of the “steeply inclined slopes below” of Trask (1926, p. 176). However, the present sea cliffs in this area are obviously being eroded into this older topography and both the north and south edges of the terraces have been truncated by the erosion producing the present topography. Thus the terraces must be younger than Trask’s “steeply inclined slopes below,” and older than the pres- ent cycle of erosion. The ravine on the north is more than 50 feet deep where it cuts through the terraces and probably required considerable time for its formation.
Two gastropods which occur commonly in the travertine were identi- fied as Plespericola pinicola Berry and Helminthoglypta dupetithouarst (Deshayes) by Allyn G. Smith of the California Academy of Sciences who stated that he can not distinguish the fossil specimens from those living in the area today; consequently he would think that the terraces were Sub-Recent in age. However, Mr. Robert T. Simmonds (unpublished Ph.D. thesis, Univ. of Illinois) has been studying gastropods of the gen- eral Polygyra group from the Recent and the Pleistocene to see if it is possible to establish ages by means of amino acid deterioration in the shells. Mr. Simmonds ran chromatogram tests for amino acids on the shells of fossil specimens of Helminthoglypta dupetithouarsi from the travertine and on present-day specimens from the terraces. On the basis of his amino acid dating technique, Simmonds states that the fossil shells in the travertine definitely predate historic times and in all prob- ability are more than 10,000 years old. A possible uncertainty in accept- ing the amino acid date might be the effect of heat (from the warm water of the hot springs) on deterioration of the amino acids. It is thought, however, that by the time the water had reached the surface and the travertine had been precipitated, the temperature must have been low- ered to a point where it would not seriously affect the amino acids. The age of 10,000 or more years would appear to be substantiated by the in-
1963] LANGENHEIM: LITTLE SUR FLORA OW
Ss
Trinidad Head
DISTRIBUTION MAP OF QUATERNARY AND RECENT
CLOSED-CONE PINE FORESTS
Drakes Bay &
Mussel Rock---
“San Bruno Point Ano Nuevo
panncug NN oe ES
“San Simeon 47 Cambria
Morro Bay CY pacho Hills
// Purisima Ridge Carpinteria
Fic. 2. Map showing distribution of Pleistocene and recent closed-cone pine forests in California. Slanted lines indicate present-day distribution and black circles indi- cate fossil records of closed-cone pine forests.
ferences from the topographic relationships of the terrace deposits.
The plant fossils do not furnish conclusive evidence as to the age of the deposits as during the Pleistocene the coastal region of California did not have changes in flora comparable to those elsewhere apparent during the change from glacial to interglacial epochs. Mason (1934) indicates that northern plants range farther south but the general aspect of the flora was the same. A lag in plant response to climatic fluctuations has made the presence of northern species useless as a means of detailed cor- relation at the present time. It is only at the periphery of the distribution of the flora that differences can be noted which may be used for this purpose.
38 MADRONO [Vol. 17
Fossit PLANTS FROM THE TRAVERTINE
Most of the fossils preserved in the travertine are impressions of leaves, with a few casts of stems and cones. In most of the other Pleistocene deposits recording the history of the closed-cone pines in California, ie. Carpinteria, Santa Cruz Island and Tomales Bay, there have been many more fruits or seeds preserved than leaves. Several samples of the traver- tine were tested for pollen and spores, but no recognizable remains were observed.
Of the species represented by the travertine fossils, the following are considered to be forest plants: Pinus radiata, P. muricata (?), Pseudot- suga menziesu (?), Cupressus cf. goveniana, Quercus agrifolia, Myrica californica, Ribes cf. sanguineum var. glutinosum, Rubus cf. parvti- florus (?), R. cf. vitifolius (2), Ceanothus cf. griseus, Garrya elliptica; and the following hydric or streamside plants: Equisetum cf. hiemale var. californicum, Carex spp., Juncus spp., algae.
Descriptions of the fossils and discussions of the modern occurrence of the species represented follow.
PINUS RADIATA Don and P. MuricaATA Don (fig. 3, g and /). Casts of three partially preserved cones were found in the lower beds of Terrace 1. Preservation is poor, or at least the cones appear to have been weathered previous to being incorporated into the travertine. One specimen has only the lowermost scales preserved; however, the umbos are rounded to a knob which is a distinctive character of P. radiata. Another specimen has what appears to be one side of the cone preserved. It is 8 cm. long and ca. 5 cm. wide. The umbos on the lower scales are rounded, but they are also of a form which might be interpreted as that of P. muricata. The third specimen is so poorly preserved that it can only be stated that there is indication of a pine cone. However, below the piece of cone there ap- pears to be a fascicle of three needles. Pinus muricata has two needles in a fascicle whereas P. radiata may have either three or two needles. The needles appear to be narrower than the average for extant P. radiata. As a result of some operation that caused the margins of the needles to become inrolled toward the midvein, the needles are circular in cross sec- tion and their width has been reduced. This inrolling may have occurred if they dropped into hot spring water. There also appear to be other scat- tered fragments of needles which may be turned on edge.
Fossil cones and wood of P. radiata have been found in the Pliocene Merced Sandstone (Glen, 1959), the Pleistocene deposits at Mussel Rock, at Carpinteria, and at Tomales Bay. Pinus radiata is the most abundant species in the Tomales deposit where it is represented by wood, cones, and leaves. The only previously known fossil occurrence for P. muricata is the Tomales Formation where it was far less abundant than P. radiata.
At present P. radiata is a coastal endemic of central western California, occurring from Pescadero, San Mateo County south to San Simeon, Mon- terey County. A variety also occurs on the coastal islands off of southern
1963 | LANGENHEIM: LITTLE SUR FLORA 39
California. Pinus muricata ranges discontinuously through about ten de- grees of latitude, from Trinidad Head in Humboldt County to La Purisi- ma Ridge in Santa Barbara County, thence southward through insular California to Guadalupe Island. On Huckleberry Hill near Monterey both P. radiata and P. muricata occur together but P. muricata tends to occur on shallow soils.
PSEUDOTSUGA MENZIESII (Mirb.) Franco(?). Another cast which is apparently a cone has a maximum length of 2.5 cm. and diameter of ca. 2 cm. It suggests either Picea or Pseudotsuga. There is no indication of the bract subtending the cone scale which is the usual means of identi- fying living Pseudotsuga menziesii. This has been true of all of the other Pleistocene records of the species (Mason, 1927, 1940; Potbury, 1932; Chaney and Mason, 1934); however, in Carpinteria and Tomales de- posits, the portion of the bract lying underneath the cone scale and pro- tected thereby, is preserved. This serves as a means of separating fossil cones of Pseudotsuga from those of Picea. In the Tomales flora certain fossils were identified as Picea because the cones bore a greater number of scales in proportion to their size than do those of Pseudotsuga and also because needles and twigs were found. Preservation of the Little Sur material is too poor to make positive determination possible. Mason (1934) states, however, that Tomales is the southern limit of the known range of Picea in the Pleistocene.
Pseudotsuga perhaps shared dominance with Pinus radiata and P. muri- cata of the forests about Tomales Bay in the Pleistocene as indicated by the hundreds of parts of cones. Pseudotsuga has also been reported from the Willow Creek flora on Santa Cruz Island. It has been recorded in the Pliocene of California (Dorf, 1930) where it is most often associated with Sequoia.
The modern range of Pseudotsuga menziesii along the coast is from Salmon Creek in the Santa Lucia Mountains northward to the northern end of Vancouver Island, British Columbia.
CUPRESSUS cf. GOVENIANA Gord. (fig. 3, b). Concentrated masses of branchlets of Cupressus occur in the lower part of Terrace 1, in the same interval where the pine cones were found. The branchlets are ca. 1—1.5 mm. wide; the leaves likew.se are approximately 1—-1.5 mm. long, oppo- site, appressed, acute but tending to be more or less blunt. Two cupres- soid fructifications were found.
Species of Cupressus are common in the California Pleistocene, being reported from La Brea, Carpinteria, Santa Cruz Island and Tomales de- posits. Mason (1927, 1934), however, points out the difficulty in specific identification of fossil material because of the close relationships among the various modern species. Determinations must be based largely upon ecological considerations suggested by the association of species and in many cases such a determination can not be accepted as final.
40 MADRONO [Vol. 17
Cupressus macrocarpa is restricted to granodiorites on the headlands of Monterey Bay where it is an associate of Pinus radiata. Cupressus go- veniana occurs around Monterey on the Monterey Shales, and is asso- ciated in the highlands with both Pinus radiata and P. muricata. In Men- docino County Cupressus pygmaea occurs in the vicinity of Fort Bragg, where it is an associate with Pinus muricata, Pseudotsuga menziesi, Picea sitchensis, Tsuga heterophylla and Sequoia sempervirens. The only other possibility for these fossils is Cupressus sargentit which has the widest range of the species of Cupressus in California. It is not strictly coastal although it is reported from some localities near the sea. It ranges from the southern Santa Lucia Mountains north to Mendocino County. De- spite its wider geographic distribution, it tends to be limited to serpentine outcrops. Thus from the location, general association and the absence of granodiorites in the vicinity of the terraces, it appears that the branch- lets in the travertine are more likely to be referable to C. goveniana than the other species.
QUERCUS AGRIFOLIA Nee (fig. 3, e; fig. 4, e). Upon initial study, the abundant specimens of Garrya elliptica were thought to be variable forms of Quercus agrifolia. However, close investigation of the venation pat- terns showed that a few specimens only were probably Quercus. These specimens occurred in the same beds with those assigned to Garrya. No fructifications were found.
Quercus agrifolia has been reported in the Pleistocene both from the asphalt beds of Rancho La Brea (Frost, 1927) and Carpinteria as well as the Tomales Formation (Mason, 1934). The Carpinteria material is more nearly like the shrub form of the species, i.e. those occurring on sterile shoots or crown sprouts that come up after fire. Neither acorns nor cups were found in the Carpinteria deposits, although they are common at Tomales Bay and elsewhere in the Pleistocene.
The modern Q. agrifolia typically is a coastal oak ranging through the coastal mountains of central and southern California. In the isolated rem- nants of the original closed cone-pine forests scattered from La Purisima Ridge northward to Sonoma County coast, Q. agrifolia is the dominant
EXPLANATION OF FIGURE 3.
Fic. 3. Fossils from Little Sur travertine deposits. a, Rubus cf. parviflorus Nutt. Portion of folded leaf. & 1. Univ. I'linois Pa‘eobot. Coll., Hypotype L-1. 6, Cupressus cf. goveniana Cord. Branchlet and possible frutification. * 1. Univ. Illinois Paleobot. Coll., Hypotype L-3.c. Rubus cf. parviflorus Nutt. Portion of leaf. x 1. Univ. Illinois Paleobot. Coll., Hypotype L-2. d, Ribes cf. sanguineum var. glutinosum (Benth.) Loud. Portion of leaf. * 1. Univ. Illinois Paleobot. Coll., Hypotype L-4. e, Quercus agrifolia Nee. Portion of leaf. * 1. Univ. Illinois Paleobot. Coll., Hypotype L-6. f, Ribes cf. sanguineum var. glutinosum (Benth.) Loud. Portion of leaf. & 1. Univ. Illinois Paleobot. Coll., Hypotype L-5. g. Pinus muricata Don? Portion of cone. X 1. Univ. Illinois Paleobot. Coll., Hypotype L-7. h, Pinus radiata Don. Portion of cone. x 1. Univ. Illinois Paleobot. Coll., Hypotype L-8.
1963] LANGENHEIM: LITTLE SUR FLORA 41
g
Fic. 3. Fossils from Little Sur travertine deposits.
42 MADRONO [Vol. 17
understory tree in the forest. From the basis of general association, it seems probable that these fossil leaves are referable to QO. agrifolia.
MyrIcA CALIFORNICA C. & S. A single leaf impression is suggestive of Myrica californica. It is ca. 4 cm. long and 1 cm. wide, and the dentition is not preserved because of strong inrolling of the margin of the leaf. Although preservation is not sufficient for unquestioned identification, the venation seems to indicate this species.
M yrica californica is represented in the Pleistocene deposits of Tomales Bay by an abundance of fruits and a few leaf impressions from indurated nodules. It is also known from the Santa Cruz Island and Carpinteria floras. No records are available from the more arid Rancho La Brea deposits.
The extant Myrica californica is strictly a coastal plant, attaining its best development only on the coastal side of the mountains in valleys im- mediately exposed to the sea. It ranges from the Puget Sound region southward through the coastal area of Oregon and northern California, thence becoming rarer southward where it extends to the Santa Monica Mountains of southern California. Mason (1934) suggests that the south- ernmost records of present distribution are probably relicts of Pleisto- cene distribution. The distribution of Myrica californica ends abruptly in the region of what probably was the southernmost extension of the cool humid Pleistocene climate.
RIBES cf. SANGINEUM var. GLUTINOSUM (Benth.) Loud. (fig. 3, d and f). Ribes leaves occur frequently with Rubus in beds of Terrace 1. As in the case of the other leaves, none are preserved in entirety, with masses of fragments occurring on top of each other. Almost all of the fragments are bent and frequently the edges turned under. The leaves are rounded in general outline with palmate divergence of three major veins indicating three lobes. They average 2—5 cm. wide. These characters tend to indicate the Ribes sangineum group, although the venation pattern fits the variety glutinosum better than the typical form. The variety now occurs in open places or brushy sites or in the closed-cone pine forest from Santa Bar- bara to Humboldt County. There is no previous Pleistocene record.
Rusus cf. PARVIFLORUS Nutt. and Rubus cf. viTIFoLIus C. & S. (fig. 3, a and c). In the lower beds of Terrace 1 Rubus leaves occur commonly. They tend to be found with masses of Ribes leaves and usually apart from the large concentrations of Ceanothus and Garrya which occur in the same sequence of beds. As usual only fragments of the leaflets are pre- served, with the specimens being bent and folded so that it is difficult to determine the nature of the whole leaflet in most specimens. Some of the specimens are generally ovate in shape with others being palmately lobed. The size likewise varies greatly with the amount of the fragment pre- served. Venation characters help indicate the size, however, especially by indicating the possibility of lobing. In many fragments the venation is
1963 | LANGENHEIM: LITTLE SUR FLORA 43
pinnate toward the terminal portion of the leaflet, but where the lower portion of the leaflet is preserved there is palmate divergence even though the lobe may not be intact.
The characters of some of the fragments suggest Rubus vitifolius as leaflets of this form may be either ovate or palmately lobed. However, other specimens, in terms of larger size and nature of the lobing appear to fit Rubus parviflorus better. Thus there appears to be little question of the presence of Rubus, but the amount of a leaflet preserved raises a question as to specific determination.
The only known Pleistocene record of Rubus vitifolius is from Tomales Bay (Mason, 1934) although Mason assumed that it was present in most of the coastal region of California. At present Rubus vitifolius is one of the most abundant shrubs in the valley and low hill country of California, occurring chiefly along streams and springy flats.
Rubus parviflorus fruitlets occur in the Pleistocene beds of Tomales Bay and from the San Bruno deposits. At the present time it is common along canyon streams and in open woods from southern California to Alaska. A clearly marked variety, called velutinus, occurs at the present time in the closed-cone pine and redwood forests near the coast from Santa Barbara to Mendocino County. It frequents streambanks and moist swales not too exposed to the sun.
CEANOTHUS cf. GRISEUS (Trel.) McMinn (fig. 4, b and c). Ceanothus leaves are abundantly represented in Terrace 1 in the beds which likewise frequently contain Garrya and sedge-like leaves. The leaves appear to have fallen in great quantities into the pool at one period, as the speci- mens are frequently laid one on top of the other. The specimens are frag- mentary with the edges of the leaves universally curled under. This is due not only to the possible curling effect of hot water, but also to the natural tendency of leaves of some species of Ceanothus to be revolute. Because the edges are curled, it is impossible to determine the margin. The shape generally is oblong-ovate to broadly elliptical. The portion preserved varies in length up to 4 cm., indicating that some were even longer. The greatest width is 2.5 cm. with 1.5 cm. being common. Some show three prominent veins, although a few have only one. The lateral “‘reticulate” venation pattern is also prominent. These characters fit within the range described for C. thyrsiflorus or C. griseus. At one time C. griseus was con- sidered a local variant of C. thyrsiflorus (Jepson, 1925), although recent- ly it has been raised to specific status (McMinn, 1942; Munz and Keck, 1958).
Ceanothus thyrsiflorus is native to the cool coastal region from Santa Barbara County, California, north to Douglas County, Oregon, being especially abundant in the redwood belt of central and northern Califor- nia. On the Monterey Peninsula it is replaced by C. griseus which occurs with the closed-cone pines and on the open slopes in the chaparral. Both of these species hybridize with other species of Ceanothus, providing ad-
44 MADRONO [Vol. 17
ditional sources of variation. The general characers preserved in these fossils place them within either the group of C. thyrsiflorus or that of C. griseus. The habitat and the associated community however suggest C. griseus.
Ceanothus thyrsiflorus has been reported from the California Pleisto- cene beds of Santa Cruz Island, Carpinteria, San Bruno and Tomales Bay. Mason (1934) states that the one leaf impression from the Tomales flora looks much like the gviseus variant, but restricts his identification to C. thyrsiflorus.
GARRYA ELLIPTICA Dougl. (fig. 4, d and f). Fragments of leaves that appear to be Garrya occur most abundantly in beds of Terrace 1, just below the zone containing the masses of Cupressus branchlets. However, they occur occasionally in deposits of the upper terraces where sedge-like leaves are the predominant observed fossils. The fact that only Garrya was found with the sedges, however, may be more a function of the expo- sures being covered with caliche and thus either masking the other fossils or rendering access to fossils very difficult in the upper terraces. On the other hand, the leaves of Garrya are more coriaceous than most of the others found and thus may have been preserved better under the possibly more unfavorable conditions in the upper hot pools.
The portions of the leaves preserved vary from 3 to 6 cm. in length and they indicate that the entire leaves must have been oblong or elliptic. Few leaf margins are intact, the edges usually being curved under. The general character of the leaf is within the range of variability occurring in Garrya elliptica, although one could easily confuse leaves of this group with those of the live oaks (particularly Quercus agrifolia).
Garrvya elliptica is a coastal species ranging through west-central Cali- fornia from Monterey north to Humboldt County. It is frequent in the understory of the closed-cone pine forests at Monterey and may serve as an indication of considerable fog and a rainfall of at least twenty to forty inches.
The genus Gavrya has been described from Pliocene beds of Bennett Valley and Coalinga, California (Dorf, 1930). In the Pleistocene the only known records are based on specimens closely referable to the modern G. elliptica. The Pleistocene specimens from Carpinteria, and Santa Cruz Island are both south of the present range.
EQUISETUM Cf. HIEMALE var. CALIFORNICUM Milde (fig. 4,a). Portions of stem occur thoughout the beds from Terrace 1 and are particularly abundant in the exposure along the roadcut. They are preserved as in- ternal casts, showing only the nodes and no sheaths; the stems average 8-10 mm. in diameter. The internodal ridges are well preserved; the aver- age number is approximately 30.
The specimens could be referable either to Equisetum laevigatum or E. hiemale var. californicum as far as diameter (2—8 mm. or 5-15 mm. respectively) is concerned. However, the fossils are larger and more
1963 | LANGENHEIM: LITTLE SUR FLORA 45
Fic. 4. Fossils from Little Sur travertine deposits. a, Equisetum cf. hiemale var. californicum Milde. Portion of stem. X 1. Univ. Illinois Paleobot. Coll., Hypotype L-9. b. Ceanothus cf. griseus (Trel.) McMinn. Portion of leaf. * 1. Univ. Illinois Paleobot. Coll., Hypotype L-10. c, Ceanothus cf. griseus (Trel.) McMinn. Portion of leaf. x 1. Univ. Illinois Paleobot. Coll., Hypotype L-11. d, Garrya elliptica Doug]. Portion of leaf. & 1. Univ. Illinois Paleobot. Coll., Hypotype L-12. e, Quercus agri- folia Nee. Portion of leaf. & 1. Univ. Illinois Paleobot. Coll., Hypotype L-14. f, Gar- rya elliptica Dougl. Portion of leaf. 1. Univ. Illinois Paleobot. Coll., Hypotype L-13.
robust in general than E. laevigatum. The two distinct rows of silica tubercles also tend to indicate E. hiemale var. californicum.
Potbury (1932) reported small portions of Equisetum stems with nodes and portions of internodes from the San Bruno deposits. There was no previous report of Hquisetum in the California Pleistocene but Potbury predicted that it would be recorded.
Equisetum hiemale var. californicum is found living alongside streams and in marshy places from southern California to Humboldt County and thence north to Alaska.
CAREX spp. and JuNcus spp. Abundant sedge-like leaves and stems occur throughout the beds of the various terraces. There are the usual signs of curling or inrolling of the leaves. They vary in width from 2 mm. to 18 mm., the most common range being 2.5—5 mm. They give the gen- eral appearance of leaves of sedges with very prominent vertical stria-
46 MADRONO [Vol. 17
tions. A few flat leaves approximately 15-18 mm. wide might indicate Typha latifolia; however, lack of horizontal striations makes this iden- tification seem unlikely. The stems are circular and hence would not be sedges which have a triangular cross section. Along the streams and springs in the area at present both sedges and rushes are common. Grasses, however, can not be excluded as a possibility, but sedges seem more probable in a hot springs environment. Along the edge of Ter- race 2 masses of sedge appeared to be in growth position, bent in such a manner as to indicate water flowing downhill. On this terrace almost any portion of travertine sampled for fossils produced masses of these Carex and/or Juncus leaves.
Carex is known from both the San Bruno and Tomales Pleistocene deposits in California. In both cases, perigynia, which are a much better means of identification than leaves, were preserved. There is no record of Juncus from the California Pleistocene, but there is no reason why it should not be expected.
AtGaAE. In several beds tangled masses of fibrous filaments occur which give the appearance of algae. A sample was tested for the presence of spores but no recognizable forms were recovered. Thus determination of these obvious algal remains can not be made.
Certainly the presence of algae is to be expected in travertine deposits. Davis (1897) states that algae are responsible for the peculiarities in structure of travertine, i.e. the formation of filaments, ropes, beads, tufts, etc. Otherwise he believes that the calcium carbonate would be laid down “like coats of whitewash.” The algal inhabitants of hot springs are all members of one or two families, closely related to one another, as condi- tions are generally not suitable for most algae. Emig (1917) states that in the Arbuckle Mountains of Oklahoma there is a natural cycle of plants that accompanies the appearance and development of travertine. During the earliest states of development the unicellular green algae (i.e. Protococcus) and blue-green algae (i.e. Oscillatoria and Lyngbya) are present, followed by filamentous green algae (i.e. Vaucheria, Oedo- gonium, Rivularia, Cladophora, et al.) that grow in felt-like masses. Also mosses that aggregate in dense tufts are present.
VEGETATIONAL RELATIONSHIPS
An assemblage of the recognized plants in the fossil assemblage indi- cates a closed-cone pine forest with streams or springs along or through the forest. The forest apparently included Pinus radiata, possibly P. muri- cata, and Cupressus goveniana. There is likewise a possibility of Pseudo- tsuga menziesi. The understory trees include an abundance of Garrya elliptica, Ceanothus griseus, lesser amounts of Quercus agrifolia and pos- sibly Myrica californica. Understory shrubs include Rubus parviflorus, Rubus vitifolius and Ribes sanguineum. Springs are necessary to the for- mation of travertine, and thus the presence of Equisetum as well as sedges, rushes, and algae was to be anticipated.
1963 | LANGENHEIM: LITTLE SUR FLORA 47
It appears probable that in this area there was a closed-cone pine forest along a canyon or swale with travertine being deposited along a stream— hot spring area. Perhaps as the hot springs became more active, imme- diately adjacent trees and shrubs were killed, possibly accounting for the abundance of needles of Cupressus and leaves of Ceanothus and Garrya. Likewise as activity became even more extensive, only sedges and rushes could persist, hence their prevalence in certain portions of the terraces.
The present vegetation surrounding the area of the terraces gives no evidence of closed-cone pine forest. It is predominantly a coastal sage pattern, which generally is characteristic of the drier and rockier slopes of the outer California Coast Ranges. The major dominant in the Little Sur area is Artemisia californica with Baccharis pilularis very abundant. Salvia mellifera, Rhus diversiloba, and Pteridium aquilinum also are common. Eriogonum spp., Eriodictyon crassifolium, Diplacus aurantia- cus, Dendromecon rigida, Agastache spp. are scattered although locally common.
Temporary streams occupy most of the nearby ravines and gullies, al- though a few apparently have a year-round source of water. An occasional wind-trained redwood and Photinia arbutifolia occur in such sites. About a quarter of a mile south of the terraces there is a semi-permanent stream with Salix scouleriana, Amelanchier alnifolia, Cornus californica and Rkamnus spp. forming a dense shrubby overstory. The understory consists of Equisetum telmateia and Stachys chamissonis, with Rubus vitifolius climbing over the other plants. Species of Carex and Juncus are also present.
Coastal sage is the predominant vegetation pattern northward from the terraces to approximately Bixby Creek (about 3 miles from Little Sur River) where species of Ceanothus and Arctostaphylos form a dense chaparral cover. From thence north to Carmel, coastal sage and chaparral alternate, with the latter occupying the more mesic sites such as north- facing slopes. Pinus radiata and Cupressus goveniana do not appear until near Carmel. The forest occupying the ridge between Carmel Bay and Monterey Bay is probably the best developed closed-cone pine forest in central and southern California. This forest occupies the entire Monterey Peninsula to the exclusion of other forest types, but extends only a few miles into the interior, and only a short distance to the north as well as to the south. Whether this forest was more extensive in the early history of white man is not known, but there is record of lumbering operations. Also pine roots have been found in the soil to the north and Mason (1934) states that it is probable on the basis of typically associated species that the forest once extended northeastward to the Salinas River.
To the immediate south of the area, coastal sage is replaced by grass- land with scattered oaks (primarily Quercus agrifolia), thence into a local redwood forest in the Big Sur area. Southward the only known groves of Pinus radiata occur near San Simeon and Cambria. The forests are similar in aspect to those of the Monterey region but not as rich in species.
48 MADRONO [Vol. 17
Occasional individuals of P. muricata and the kinds of shrubs suggest more favorable soil conditions than at Monterey. In the Pecho Hills, pine is largely replaced by a dense live oak assemblage. Neither pines nor Ceanothus occur in an assemblage at Purisima Ridge that is other- wise like the Monterey forest. Pinus radiata var. binata likewise occurs on Guadalupe Island.
Despite the fact that obviously only a small portion of the forest flora was preserved in the travertine, the pattern of the flora fits that of the modern Monterey closed-cone pine forest. The entire forest there is dom- inated by Pinus radiata; the most conspicuous understory tree is Quercus agrifolia with scattered trees of Myrica californica and Umbellularia cali- fornica. The most characteristic feature of the forest is the extensive chaparral-like ground cover made up of species of Ceanothus and Arcto- staphylos associated with a number of other shrubs such as Adenostoma, Lonicera, Ribes, Rosa, Rubus, Symphoricar pos, et al. At slightly higher elevations Cupressus goveniana occurs, whereas Pinus muricata is found in areas of greatest summer fog and heaviest winter rainfall. Mason (1934) points out that local variations in ecological aspect of the forest flora seem to be related to depth and character of the soil as well as atmos- pheric moisture. Thin, rock soil layers contain dense brush and in some cases take on an aspect of typical pine barrens. The largest trees and most open forest occur when soil is deep and ground water abundant. From the possible presence of P. muricata as well as Cupressus, it appears that the forest at the time of travertine deposition might have been at a site with considerable fog. Myrica likewise indicates this. The preponder- ance of Garrva elliptica and relatively few preserved specimens of Quer- cus agrifolia are noteworthy. It is likewise interesting that there is an abundance of Ceanothus but no record of Arctostaphylos, which one would expect to be preserved more eas:ly than the Ceanothus. The abun- dance of Ribes and Rubus are to be expected in a more mesic aspect of the forest such as one finds along streams or shaded canyons, and prob- ably on deeper soils.
VEGETATIONAL HISTORY
The present distribution of the closed-cone pine forest is highly dis- continuous (fig. 2). The northernmost outpost is Trinidad Head in north- ern California, where it is almost obscured by redwood forest. There then is a gap 100 miles south to Inglenook where it occurs as typical coastal forest for about 100 miles to Fort Ross. After a 50-mile break it recurs on Inverness Ridge across from the fossil localities at Tomales Bay. Some remnants also occur on San Geronimo Ridge in Marin County. There is another 75 mile gap to Point Ano Nuevo where another small grove occurs, then another 40 miles to the south end of Monterey and Carmel bays where there is an extensive forest. It is another 60 m.lies south to San Simeon and Cambria before the forest type is encountered again, and thence southward to the isolated patches occurring at Pecho Hil's and
1963 ] LANGENHEIM: LITTLE SUR FLORA 49
La Purisima Ridge. There is another break on the mainland of 500 miles to Point San Quintin, Baja California, Mexico. Remnants of closed-cone pine forests also occur on Cedros Island 150 miles south of Point San Quintin, and on Guadalupe Island 200 miles off the coast of Baja Cali- fornia. They also occur on Santa Rosa and Santa Cruz islands of the Channel group.
These modern forests are not homogeneous floristically or ecological- ly. Evidence from fossil deposits from such widely separated areas as Tomales Bay, Carpinteria and Santa Cruz Island indicates that this for- est assemblage is a relict of a past flora and that it was more homogeneous throughout its range in the past than today (Mason, 1934). Mason also points out that the present discontinuity is due to significant geological events, and that for a relatively long interval of time the forest type occupied an extensive area, of which its modern occurrences are but fragments. In all localities in the Pliocene and Pleistocene where fossil records of closed-cone pines are known, the site of deposition has been along one of the major fault lines in the California Coast Ranges. Like- wise in all cases the areas occur along fault zones near a block that has been positive throughout the Pliocene at least. Most conspicuous are the blocks severed from the mainland by the San Andreas fault. In the Pleis- tocene deposits some of the plant materials have come from lands up- lifted since the Pliocene but presence of sediments from older landmasses has always been demonstrable.
Mason likewise states that the closed-cone forests today are confined to lands that were islands throughout much of the Tertiary and particu- larly during the Pliocene. Homogeneity of species content and absence of other contemporary assemblages in the Pleistocene support this hypoth- esis. This insular relationship is further supported by the high degree of endemism among the forest species: 58 per cent of the woody plants are endemic to the California floristic province with 29 per cent of the above restricted to closed-cone pine forests. This is remarkable considering the very small area of the California province occupied by closed-cone pines. Because most of the endemic species have closely related species else- where in the Coast Ranges, it appears probable that there was insular isolation with water barriers. Also the relation of the modern flora to that of the islands off of southern California is important. Several species such as Vaccinium ovatum and Arbutus menziesiu have their southward exten- sion on islands or occur on the southern mainland only in localities that appear to have been Tertiary islands. Pinus radiata grows in the general region of Monterey Bay, Morro Bay and on Guadalupe Island (200 miles off the coast of Baja California), but does not grow on the mainland in between. The sporadic occurrence of Cupressus from Guadalupe Island to coastal Mendocino County also is another similar distribution. Cupres- sus is known to have occurred in every closed cone-pine forest of which there is fossil record. In many areas it now is extinct, e.g. Santa Barbara region and Santa Cruz Island. Likewise Pinus radiata has disappeared
50 MADRONO [Vol. 17
from both of these localities, and both P. radiata and Cupressus are now absent from Tomales Bay although they were common there during the Pleistocene.
Mason further indicates that the major differences in aspect that pre- vail in the forest throughout its range today are due primarily to the mingling of the original species with those of migrating populations and consequent selection of floras due to changing climatic conditions in Pleis- tocene and Recent time. The California insular floras are considered by Mason to represent the original vegetation of coastal California prior to their invasion by continental floras. The discontinuous closed-cone pine forests along the California coast are the last remnants of these insular floras. The fossil occurrences described in this investigation give addi- tional documentation of the former more extensive distribution of this vegetation type.
SUMMARY
An assemblage of plant fossils from travertine terraces in Monterey County, California, provides additional documentation of a former more extensive distribution of closed-cone pine forests during the Quaternary. The travertine terraces are considered to be 10,000 or more years old as evidenced by amino-acid dating of gastropod shells in the travertine and substantiated by the topographic relationships of the deposits. The plant fossils are relatively poorly preserved impressions of leaves, with a few casts of stems and cones. Despite the fragmentary record, the forest ap- parently included Pinus radiata, Cupressus goveniana and possibly Pinus muricata and Pseudotsuga menziesu. An abundance of Garrya elliptica, Ceanothus griseus, lesser amounts of Quercus agrifolia and possibly My- rica californica occurred as understory trees. Rubus parviflorus, Rubus vitifolius and Ribes sanguineum var. glutinosum comprise the recorded understory trees. Equisetum hiemale var. californicum as well as sedges, rushes and algae also occurred commonly. Thus it appears probable that there was a closed-cone pine forest along a canyon or swale with traver- tine being deposited along a stream—hot spring area. The present vege- tation surrounding the area of the terraces, being predominantly a coastal sage pattern, gives little evidence of closed-cone pine forest. The pattern of the fossil flora, however, fits closely that of the modern closed-cone pine forest on the Monterey peninsula. The location of the travertine terraces also generally supports Mason’s (1934) conclusion that fossil records of closed-cone pines have occurred near fault blocks that were positive during the Pliocene. This insular relationship helps to account for the highly discontinuous and endemic nature of the floras.
Biological Laboratories, Harvard University, Cambridge, Massachusetts
Department of Paleontology, University of California, Berkeley
1963] LANGENHEIM: LITTLE SUR FLORA on
LITERATURE CITED
AGARDH, C. 1827. Aufzahlung einiger in den 6streichischen Landern gefundenen neuen Gattungen und Arten von Algen, nebst ihrer Diagnostik und beigefugten Be- merkungen. Flora 10:625-640.
CHANEY, R.W. and H. L. Mason. 1930. A Pleistocene flora from Santa Cruz Island, California. Carnegie Inst. Publ. 415:1-24.
. 1933. A Pleistocene flora from the asphalt deposits of Carpinteria, Cali- fornia. Carnegie Inst. Publ. 415:45-79.
Coun, F. 1862. Einen Vortrag uber die Algen des Karlsbader Sprudels und deren Antheil an der Bildung des Sprudelsinters. Flora 45:538—-540.
Davis, B. M. 1897. The vegetation of the hot springs of Yellowstone Park. Science, NS 6:145-157.
Dorr, E. 1930. Pliocene floras of California. Carnegie Inst. Publ. 412:1-108. Emic, W.H. 1917. Travertine deposits of Oklahoma. Okla. Geol. Surv. Bull. 29:1-76. Frost, F. H. 1927. The Pleistocene flora of Rancho La Brea. Univ. Calif. Publ. Bot. 14:73-98. GLEN, WILLIAM. 1959. Pliocene and lower Pleistocene of the western part of the San Francisco peninsula. Univ. Calif. Publ. Geol. Sci. 36(2) :147-198. Jepson, W.L. 1925. A manual of flowering plants of California. Univ. Calif. Press. Berkeley. Jones, J. C. 1914. The tufa deposits of the Salton Sink. Carnegie Inst. Publ. 193: 79-84. KELLERMAN, K. F. and N. R. SmitH. 1914. Bacterial precipitation of calcium car- bonate. Jour. Wash. Acad. 4:400-402. Mason, H. L. 1927. Fossil records of some west American conifers. Carnegie Inst. Publ. 346:139-158. . 1934. Pleistocene flora of the Tomales formation. Carnegie Inst. Publ. 415:81-179. . 1940. A Pleistocene record of Pseudotsuga macrocarpa [Carpinteria, Cali- fornia]. Madrono 5:233-235. . 1942. Distributional history and fossil record of Ceanothus, in Ceanothus by M. van Rensselaer and H. E. McMinn. Santa Barbara Bot. Gard. . 1957. A Flora of the Marshes of California. Univ. Calif. Press. Berkeley and Los Angeles. McMinn, H. E. 1942. A systematic study of the genus Ceanothus, in Caenothus by M. van Rensselaer and H. E. McMinn. Santa Barbara Bot. Gard.
Meunier, S. 1899. Observations relativ du depot de certain travertins calcares. Compt. Rend. Acad. Paris 129:659-666.
Muwz, P.A. and D. D. Keck. 1959. A California flora. Univ. Calif. Press. Berkeley and Los Angeles.
PotBury, SUSAN S. 1932. A Pleistocene flora from San Bruno, San Mateo County, California. Carnegie Inst. Publ. 415:25-44.
SIMMONDS, R. T. 1962. Wisconsin Geochronological determinations based upon amine decay rates. Ph.D. dissertation, Dept. Geol., Univ. Illinois, Urbana.
Trask, P. 1926. Geology of Point Sur Quadrangle, California. Univ. Calif. Publ., Bull. Dept. Geol. Sci. 16:119-186. Pl. 16.
TWENHOFEL, W. H. 1950. Principles of sedimentation. McGraw-Hill. New York.
WEED, W. H. 1888. Formation of travertine and siliceous sinter by the vegetation of hot springs. U.S. Geol. Surv. Ann. Rept. 9:613-676.
5Z MADRONO [Vol. 17
CHROMOSOME NUMBERS OF SOME PHYTOGEOGRAPHICALLY INTERESTING CHILEAN PLANTS
D. M. Moore
Within the flora of Chile there are a number of floristic elements of which perhaps three may be singled out as having particular phytogeo- graphical interest. Two, in North Central and Southern Chile, show ties across the equator with, respectively, Southwestern North America and parts of the North Temperate region and a third, in South Chile, is linked over Antarctica with Australia and New Zealand. Taxonomists are cur- rently investigating in detail several groups of species within these ele- ments because the inter-continental affinities are, in several cases, at the specific level and they are especially relevant to problems of plant evo- lution and distribution.
TABLE 1. CHROMOSOME NUMBERS OF SOME CHILEAN PLANTS.
SPECIES NUMBER LocaLiTy COLLECTION GOODENIACEAE Selliera radicans Cav. n=—8* W. side of Tumbes Pen- Moore 286 insula, Prov. Concepcion, UCB, US, RSA Chile COMPOSITAE Franseria chamissonis n=18* ca.1.5km.N.of Lirquen, Moore 291 subsp. bipinnatisecta Prov. Concepcion, Chile UCB, US, RSA (Less.) Wiggins & Stockwell
Adenocaulon chilense Less. 2n=46** ca. 200 m., Perez Rosales Moore 327 Pass, Prov. Llanquihue, UCB, US Chile
Adenocaulon bicolor Hook. 2n=46** ca. 400m., near Corvallis, Chambers 1535 Benton County, Oregon, LA
U.S.A. Adenocaulon bicolor Hook. 2n=46** Shinano-Oiwake, Mt. I. Fukuda var. adhaerescens Asama, Nagano Prefec- in October, 1960 Makino ture, Honshu,
Japan, ca. 1000m.
* Meiosis in pollen mother cells. ** Mitosis in root tips.
The three Chilean species for which data are recorded in this note are all comparable in having the same chromosome numbers as have been determined for them or their close relatives in other continents. Thus, as in the counts for Chilean specimens presented in Table 1, Selliera vadicans Cav. has eight bivalents at meiosis in plants from New Zealand (Hair and Beuzenberg, 1960) and Franseria chamissonis Less. subsp. bipinnatisecta (Less.) Wiggins and Stockwell forms eighteen bivalents
1963 | VICKERY, MUKHERJEE & WIENS: MIMULUS D0
in Californian material (Wiggins and Stockwell, 1937). Representatives of three disjunct taxa of the genus Adenocaulon have all been counted for the first time (Table 1) and all share a somatic complement of forty-six chromosomes. Further studies of relationships within this genus are in progress.
I wish to thank Mr. Donald Kyhos for the preparation showing meiosis in pollen mother cells of Franseria. Dr. Kenton L. Chambers, Oregon State University, Corvallis, and Mr. I. Fukuda, Tokyo Women’s Chris- tian College, Japan, very kindly collected the seed of Adenocaulon. My collecting trip to Chile was financed from National Science Foundation Grant No. G-13518 made to Dr. Harlan Lewis.
Department of Botany, The University, Leicester, England.
LITERATURE CITED
Hair, J. B. and E. J. BEUZENBERG. 1960. Contributions to a chromosome atlas of the New Zealand Flora. 4. Miscellaneous families. N. Z. Jour. Sci. 3:432-440. Wiccins, I. L. and P. STOCKWELL. 1937. The maritime Franseria of the Pacific Coast. Madrono 4:119-120.
CHROMOSOME COUNTS IN SECTION ERYTHRANTHE OF THE GENUS MIMULUS (SCROPHULARIACEAE). II’
RoBERT K. VICKERY, JR., BARID B. MUKHERJEE AND DELBERT WIENS
This investigation formed part of our long range biosystematic study of section Erythranthe (Vickery, 1956; Vickery, Mukherjee, and Wiens, 1958). It had two purposes. The first was to determine the chromosome numbers of two rare species, Mimulus eastwoodiae Rydberg and M. nel- son Grant. The second was to analyze the genome homologies of the more common species of the section.
For the cytological portion of the investigation, buds expected to con- tain the desired stages of meiosis were placed for 24 hours in a freshly prepared fixative consisting of 3 parts absolute ethanol to 1 part glacial acetic acid saturated with ferric acetate. The anthers were then dissected from the buds, squashed, and lightly stained in aceto-carmine. For each determination, the chromosomes of ten or more cells were carefully studied and counted under a phase contrast microscope. Many of the configurations were recorded with the aid of a camera lucida.
Both M. eastwoodiae and M. nelsoni were found to have n=8 chrom- osomes (fig. 1 and table 1) as do the three more common species of the section, M. cardinalis Douglas, M. lewisti Pursh, and M. verbenaceous
1 This work was supported by grants from the National Science Foundation and the University of Utah Research Fund.
54 MADRONO LVol. 17
(Greene (Brozek, 1932; Vickery, Mukherjee, and Wiens, 1958). Mimulus eastwoodiae (fig. 2) is known from only six or seven localities in southern Utah and northern Arizona. The populations occur in shaded seeps in the desert at elevations below 4,500 feet. Generally, suitable seeps are on perpendicular or overhanging sandstone walls. In contrast, M. nelsoni (fig. 2) grows by small mountain streams in the cloud forest, 8,000 feet or above, in the central portion of the Sierra Madre Occidentale of Mex-
5848 6211 3423
— 20 10 O 10 micra
Fic. 1. Camera lucida drawings of meiotic chromosomes of M. eastwoodiae (5848), M. nelsonii (6211), and the Fy hybrid (3423) of M. cardinalis (5311) « M. verbe- naceous (5924). All pollen mother cells are at or near metaphase I.
TABLE 1. CHROMOSOME CouUNTS IN MIMULUS, SECTION ERYTHRANTHE.
A. CULTURES OF NATIVE SPECIES.
n=8 M. eastwoodiae Arches National Monument, Grand County, Utah, altitude 4,200 feet, Vickery 339 (5848). n=8 M. nelsonii Crest of Sierra Madre on Durango-Mazatlan Road, Durango, Mexico, altitude 9,000 feet, Wiens 2642 (6211).
B. INTERSPECIFIC AND INTERPOPULATION HyprIDs.”
n—=8 Fy (3424) of M. verbenaceous (5924) * M. cardinalis (5316).
n=8 Fy, (6031) of M. lewisii (5875), Wasatch Mountains form) x M. lewisii (5032), Sierra Nevada form).
—8 Fy, (3423) of M. cardinalis (5311) * M. verbenaceous (5924).
—8 Fy, (3425) of M. cardinalis (5031) X M. lewisii (5875, Wasatch Moun- tains form).
n=8 Fo (3502) of M. cardinalis (5077) & M. lewisii (5051, Sierra Nevada form).
2 The chromosomes showed regular pairing and/or regular segregations in all cases.
1963 | VICKERY, MUKHERJEE & WIENS: MIMULUS ap)
6211 fe
| C
Fic. 2. Photographs of the two rare species of Mimulus studied in this investiga- tion, M. eastwoodiae Rydberg (5848, UT 28,932) and M. nelsoniz Grant (6211, UT 54,858).
<
mM.
ico. The full range of this unusual and strikingly beautiful species is not known. We collected it near the crest of the Durango-Mazatlan road, which is only the second locality reported (Grant, 1924).
For the cytogenetic part of the investigation, each interspecific com- bination was made reciprocally and was repeated using different cultures of the parental species. From one to ten flowers were carefully hand- pollinated for each combination. The putative hybrid seeds were sown and the resulting seedlings carefully checked to verify their hybrid nature. The hybrids that were investigated cytologically included combinations involving the three major species M. cardinalis, M. verbenaceous, and M. lewis (table 1). They included the two distinctive forms of the latter species. In all cases the hybrids showed regular bivalent chromosome asso- ciations at metaphase I and normal segregations of the chromosomes in the later stages of meiosis.
The results of the present investigation taken in conjunction with the previous study (Vickery, Mukherjee, and Wiens, 1958) suggest that the common chromosome number of section Erythranthe is n=8. The regular bivalent association and normal segregation of the chromosomes of the hybrids involving four of the main taxa suggest the presence of essen- tially homologous genomes in these forms and possibly throughout the
56 MADRONO [Vol. 17
section. Apparently, evolution in section Erythranthe is proceeding prin- cipally by the accumulation of diverse genes in the various populations and species rather than by the accumulation of chromosomal differences.
Department of Genetics and Cytology University of Utah, Salt Lake City, Utah
Department of Obstetrics and Gynecology Columbia University, New York City, New York
Department of Biology University of Colorado, Boulder, Colorado
LITERATURE CITED
Brozek, A. 1932. Mendelian analysis of the “red-orange-yellow” groups of flower colours in Mimulus cardinalis Hort. Preslia 11:1-10.
GranT, A. L. 1924. A monograph of the genus Mimulus. Ann. Missouri Bot. Gard. 11:99-388.
PENNELL, F.W. 1951. Scrophulariaceae zm Illustrated Flora of the Pacific States by Leroy Abrams. Stanford Univ. Press. Vol. III, pp. 688-731.
VickErY, R.K., Jr. 1956. Data on intersectional hybridizations in the genus Mim- ulus (Scrophulariaceae). Proc. Utah Acad. 33:65-71.
VicKERY, R.K., Jr., B. B. MUKHERJEE, AND D. WIENS. 1958. Chromosome counts in section Erythranthe of the genus Mimulus (Scrophulariaceae). Madronfo 14: 150-153.
AN ANALYSIS OF VARIATION IN VIOLA NEPHROPHYLLA NorMAN H. RUSSELL AND FRANK S. CROSSWHITE!
The genus Viola is well known to be one of the more taxonomically “difficult” of the temperate angiosperms. Though in North America the “species” were sorted out in what appeared at the time to be a satisfac- tory manner (Brainerd, 1920), subsequent studies have shown that their limits are anything but clear. In particular regions it is possible to distin- guish separate forms easily; in others there is so much morphological and ecological variability that distinct forms or even morphological types are very difficult to describe. Polyploidy, introgression, genetic drift in isolated populations, and other hypotheses have been used to explain this situation.
More important than the explanation of this morphological and physio- logical variation is the accurate and objective description of it. A method for the more objective comparison of units (individuals and aggregations of individuals) has been suggested by the senior author elsewhere (Rus- sell, 1961, 1962) and is used in the present analysis and description. It consists of the preparation and correlation of multiple pair comparisons,
1 This research was supported by National Science Foundation grants G—4994 and G-12113.
1963 | RUSSELL AND CROSSWHITE: VIOLA oi,
ei SGP 3 aga) snl \ wo Ree oJ 50 Se é ty ee ae) Ree Vaan Sen ea | Se ile ; nf de Sia if ee a p oe Nie vl Nr : - i : We ee co ~— <A ws \ Hi oa ss ; . soo ae a poh! if V “= | c, 2 - @ 0 e \, “\ a - he Sooo a eS i Pd ay. SS a ane so eee _ Hoare ee ee e ‘ : L ee i ; e A ae a ° Ce; = od % OF ~ 40 a, teas , oe | “fee oo! ° oe % 5.08 ee - - i) 3 e fa e Ve e mae bd 2 oo i ' e e es zi i) r ; O ee ' e N eo ogee oF ' oo? ,e¢ ; of rae 5 we re? bl er e HH \ - e, \ é be \ ek aia ee ; oe e .° ‘“ “ 30 Dd wee one pee . \e J %e = O. . | Bais \ Shae Dawe ON
Fic. 1. Geographical distribution of Viola nephrophylia Greene.
using an index similar to that described by Anderson (1936), and is described in greater detail below.
Becker (1925), in the generally accepted classification of the genus Viola, recognized eleven sections, the most complex of which is the sec- tion Plagiostigma. This section was further divided into several subsec- tions, the best known of which is the subsection Boreali-americanae, a group of perhaps twenty-six so-called “species,” found exclusively in North America. Of these, only a single one, Viola nephrophylla Greene (Pittonia 3:144—-45. 1896), occurs to any extent in the Rocky Mountains, the remainder being found almost exclusively in eastern and central North America. Viola nephrophylia has the largest range of any of the stemless blue violets. A map of its distribution (fig. 1) is based upon the examina- tion of specimens from about sixty herbaria and represents the subjec- tive decisions of the senior author on the basis of ten years of field and herbarium study of the genus Viola. Each dot, therefore, represents the place of collection of a herbarium specimen that, in his opinion, would key to V. nephrophylia. As is true of all such maps, it suffers from these sub- jective decisions and, therefore, must represent, perhaps, only an approxi- mation to reality. The species, very likely, also occurs in Mexico, but we have not seen specimens from there. In many parts of its range, particu- larly where it does not grow with other kinds of stemless blue violets, it is morphologically distinct. However, in many situations in eastern North America, it is difficult to distinguish from related violets.
The habitat of V. nephropAylla varies widely in the Rocky Mountains. Usually we found it growing in moist, grassy, grazed fields, frequently in the shade of willows. Other populations were found along the shaded, sandy edges of canyon streams. In the eastern United States it grows both in open, grazed, poorly drained meadows and along the rocky shores of lakes in glaciated country. Figure 2 indicates the general appearance of this violet during the spring flowering period.
MADRONO [Vol. 17
On oo
Fic. 2. Viola nephrophylla, spring appearance, < %rds.
COLLECTION METHODS
Fifteen samples were taken by the junior author during the growing season of 1960, wherever they could be found, in rather extensive travels through Arizona and Colorado. Their locations are shown in figure 3. The number of specimens taken at each location varied from sixteen to fifty, depending upon the size of the local population. Ordinarily fifty specimens were obtained, but in some instances this was not possible. Plants were collected no closer together than six feet, to lessen the pos- sibility of sampling two members of the same clone. They were usually measured or scored while fresh, but in some instances the plants were washed, pressed, and dried before examination. All the plants measured, and the measurement data, are deposited in the herbarium of Arizona State University.
1963 | RUSSELL AND CROSSWHITE: VIOLA 59
GUNNISON
(ees
(COL
GUNNISON
SAGUA
Fic. 3. Locations of population samples in Arizona and Colorado.
After a preliminary inspection of herbarium material, those character- istics showing the most conspicuous differences between individuals were chosen for analysis. The following characters were measured or scored as indicated:
1. Length of the lamina of the largest mature leaf.
2. Breadth of this lamina.
3. Distance from the apex of this lamina to one of the basal lobes.
4. The angle made by one-half of the apical margin of this leaf with
the horizontal.
60 MADRONO [Vol. 17
TABLE |. DISTRIBUTION (EXPRESSED AS PERCENTAGES) OF LAMINA LENGTH/BREADTH RATIOS, AND INDEX VALUES ASSIGNED FOR THE FIFTEEN POPULATIONS ANALYZED.
Hybrid Index Scores
O I 2 Crosswhite Coll. No. 50259 60-69 70579 80-89 90-99 100-109 -LI0-L1I9.—s 120-129 130-159 14504 2.00 18.00 30.00 970A f°) fo) fe) 965 12.50 56.25 31.25 971 ) Co) 35.00 1239 2.00 16.00 50.00 62 f°) 26.67 40.00 23.33 10.00 0) (e) fo) fc) 843 fe) f°) 16.00 48.00 24.00 8.00 O (o) 4.00 1064 ) 8.00 20.00 36.00 32.00 4.00 fo) ) fo) 1065 Oo f°) 12.00 60.00 18.00 8.00 2.00 (o) 0) 1397 0) 17.50 65.00 17.50 ) ) 0) ) fo) 1304 ) 4.00 36.00 44.00 16.00 0) ) fo) fo) 1062 ) Co) 26.67 53.33 13.33 f°) 0) 6.67 ) N29 0) 4.00 14.00 30.00 48.00 4.00 (o) f°) fe) 1087 ) 2.22 2.22 2.0.00 40.00 26.67 6.67 2.22 ) /054 ) ) 4.00 24.00 28.00 24.00 16.00 4.00 fe)
5. Number of teeth on one-half the margin.
6. Pubescence of the upper lamina surface (scored as 0 = glabrous, 1 = slightly pubescent, 2 = moderately pubescent, and 3 = heavily pubescent).
7. Pubescence of the lower lamina surface (scored as indicated for the upper surface).
8. Pubescence of the margin of the lamina (scored as 0 = glabrous, 1 = hairy over half or more of its extent).
9. Pubescence of the petiole (scored as 0 = glabrous, 1 = 10 or more hairs present).
These leaf characteristics are those used by the senior author in studies on other stemless blue violets (Russell, 1955, 1956a, 1956b). It might have been desirable to measure other plant structures also, but either no conspicuous differences were noted in these or they were not present in all the samples. During the summer, when the majority of the samples were taken, no open flowers are produced, and, at times, no cleistogamous flowers or fruit. Rhizome differences, though present, were small and dif- ficult to measure accurately.
PREPARATION OF THE INDEX
After compilation of the data (nine measurements or scores for each specimen), three ratios were calculated: lamina length/lamina breadth; lamina length/length from lamina apex to lobe; and lamina length/lami-
1963 | RUSSELL AND CROSSWHITE: VIOLA 61
TABLE 2. INDEX VALUES ASSIGNED FOR EACH OF THE FIVE CHARACTERS USED IN THIS STUDY.
Ranges in Characters
Character Value O Value | Value 2
Lamina length/ breadth ratio Omos NGOmed S10) |S)
Lamina length/length to lobe ratio SOT ISS) Ome SBOriIg
Lamina length/twice no. lamina teeth pOORmo2 Aa\O= S/S)
Apical angle of lamina ZO So- 40°-49° One coe
Total lamina* pubescence ‘Sr .s) SyrS) O22
*Pubescence was scored on an orbitrary scale for both leaf surfaces, the margin, and the petiole. The value of 8 represented the greatest hairiness found, and that of O complete glabrousness.
na teeth. The distributions for each of the characteristics and ratios were then plotted for each sample. An example of one such distribution is shown in Table 1, for the lamina length/breadth ratio (an approximate measurement of overall shape). In the event real intra- or intersample differences were noted, index values (0, 1, and 2) were assigned to the extremes and median conditions found. This is illustrated in Table 1, and in Table 2 the index value assignments for all the characteristics used in the subsequent analyses are given.
In Table 3 the collections are arranged in order of the value of the mean index, in an attempt to reveal the existence of altitudinal or lati- tudinal clines. Total sample variation did not fall into any such pattern, although the Arizona samples generally had low values. Only a morpho- logical “cline” can be shown, and we were unable to discover any geo- graphical or ecological factor to which this could be definitely related.
The samples at opposite ends of Table 3 show considerable difference, enough so that if only these two extremes were known they might be called different “species” under present nomenclatural practices in plant taxonomy. As an example, the index distributions of two extreme collec- tions have been plotted in figure 4, on a percentage basis. The differences
62 MADRONO [Vol. 17
TABLE 3. SUMMARY OF DIFFERENCE INDEX SCORES FOR FIFTEEN POPULATION SAMPLES.
Crosswhile No.
Coll.No. Elev. Latitude Spec. O | 2 3 4 5 6 7 8 9 10 Mean Location /450A 5800' 35° 50 hon (ey a2 9S of | 3.12 Oak Creek, Ariz. 970A 6000' 34°5' 16 4 5 2 4 | 3.63 Lakeside, Ariz. 965 5800' 31°50! 16 A Ae ee i 3.81 Portal, Ariz.
971 6000' 34°5' 40 1 7 1 6 9 | 4.48 Bog Creek, Ariz. 1239 9400' = 38°50! 50 OB Ba 7? ey ey! 4.76 Cement Creek,Colo. 62 8400' 38°30' 30 ome OVE OF cso { 4.80 Almont, Colo.
843 6000! 34°20! 25 2 eae 7 ile al \ 5.32 East Verde, Ariz. /064 9200' 39° 25 2a 5 6 4 2 5.44 Gothic, Colo. /065 9300' 39° 50 10) 10) 1/2 13s 2 5.90 Gothic, Colo.
1397 7900' 38°20! 40 6 7 12 7 8 6.10 Gunnison, Colo. /304 9200' 40° 25 2 8 6 4 4 | 6.12 Nederland, Colo. 1062 8400' 38°30' 15 2 A 4e ee lee: 6.13 Almont, Colo. M29 9000' 38°50! 50 fee Ae He Te eeg 6.54 Crested Butte,Colo. /087 9500' 39° 4! WS eR ee A 7.29 Gothic, Colo. 1054 9400' 39° 25 3 5 10 7 8.84 Gothic, Colo.
between these two curves were analyzed in the following manner (Russell, 1961, 1962):
1. Percentage of the distance from the mode of one aggregation to the extreme value of the scale assigned to the other aggregation.
2. Percentage of the distance on the total index scale that the range of values for the aggregation does not cover.
3. Percentage of the total index for the range discontinuity (plus val- ues ) between the pair of samples or the total overlap (minus values) between them.
4. Percentage of the total index for the distance between the modes of the two curves.
These four descriptive features of the curves are analyzed in such a way
as to show the greatest morphological separation of the two populations.
TABLE 4. MATRIX OF TOTAL DIFFERENCE INDICES FOR ALL SAMPLE COMPARISONS, TO BE READ HORIZONTALLY.
Crosswhite
Coll No” I4SQA 970A (905 97) 1239 “M162 2845" 004 1065, 51807 904 Oe Ve Oe Coa Meany /450A x -29 -29 «14 fe) 44 39 44 89 63 67 73 72 (WO 150 48.50 970A - 43 x -33 «13 O 50 42 50 100 7I 75 8l 8l 122 167 55.43 965 -43 - 83 x 13 ) 50 42 50 100 7 ie 8I 8 122 167 51-86 97/ - 1 =43 43 X -15 fh 14 -8 5! 15 26 S20 S278 eles 16.00 1239 14 = fan eaulie e229) Xx OF F=I5 fe) 50 14 25 31 3 78 122 20.50 62 10 -l2 =“|20-37 4-30 aXe 25 Elo (o) Oo -25 -19 -I9 33. 78 = -6.00 843 39 14 14-14-1525 X -50 =| -42 | -6 18 33 «O«77 3.07 1064 49 29 2913 O =i2 0 15 x 14-33 Oy 57 -8 50 100 14.93 1065 100 88 88 63 63 38 38 2 KS) 34 25) 50-14 43 27.57 1397 87 72 72 «44 43 25 15 -8 -33 x | -8 =i, 43 100 31.86 /304 55 37 38 O13 13-12 13-2 -50 -49 Xeuse5 -8 43 99 10.43 1062 67 50 50 25 25 0) 2s 2| —6(7 00150 -6r XP es2 28 = 85 8.93 M29 66 50 50 25 25 fe) O -22 -29 -14 -1I5 -12 Xx Oo 668 13.71 1087 110 = 100 100 «78 78 55 56 44-14 14 15 7 18 x 8 47.79
1054 190 190 190 168 167 «145 145 144 100 128 128 l21 118 66 X 142.86
1963 | RUSSELL AND CROSSWHITE: VIOLA 63
Two separate analyses must be made, one for each population as com- pared to the other. The results of the analyses of the curves shown in
figure 4 are: Crosswhite 965 ........ 64/27/-9/55 Crosswhite 1054............ Total Index—137 Crosswhite 1054 ........ 82/64/-9/55 Crosswhite 965..........-- Total Index—192
Average Difference Index—164.5
Pictorialized scatter diagrams (figs. 5 and 6) illustrate the population differences further.
40
30
% 90
O | 205 4 5 6 7 8 9 #10 INDEX VALUE
Fic. 4. Index distributions for two extreme populations; further explanation im text.
Total difference indices were computed for the other possible compari- sons and were smaller than the above, grading down to minus values. They are shown in Table 4. When all the index distributions were con- verted to percentages and summed, a curve was obtained (fig. 7) which does not seem to indicate the presence of two or more types or “‘species”’ in this complex, but instead indicates a series of variable populations cen- tering about an average overall morphological condition. A more pene- trating analysis, using techniques such as those suggested by Rogers and Tanimoto (1960) might indicate whether or not more than one morpho- logical type is present, and such an analysis is being planned. Other taxo- nomists have given specific rank to certain aberrant or differing types in the range of Viola nephrophylla; namely V. arizonica Greene, V. cognata Greene, V. prionosepala Greene, V. McCabeiana Baker, and V. Clauseni- ana Baker. We do not interpret the present descriptive data to support the recognition of different morphological types or “species.”
64 MADRONO [Vol. 17
65 35
c Spe
5 er ; 4 x if 55 © 25 44 4 q z 50 Z 20 ity = FSC 1054 45 ' + i? LAMINA LENGTH 40 aL ' Fig. 6 z 9) 2 35 Ww a o z 30|[ | 30 = aq * dtd 25 aD FSC 965 20 % 10 20 . 15 ro) 5 20 25 30. 35 40 45 o 12 3 4 5 6 7 8 9 0 LAMINA LENGTH INDEX VALUE Fig. 5 Fig. 7
Fics. 5-6. Pictorialized scatter diagrams of Viola nephrophylla populations: 5, Crosswhite 965; 6, Crosswhite 1054. Fic. 7. Distribution (converted to percentage values) of indices for all populations.
A sample collected in Oak Creek Canyon, Arizona, during September, 1960, deserves special mention (Crosswhite 1450A).The specimens differ (fig. 8) only in the relationship of lamina breadth to lamina length, the ratios exposing approximately equal numbers of plants with leaves wider than long and with leaves longer than wide. We have considered that the two differing regression lines may be explained either as the result of measurement errors, ‘‘juvenile’” and mature leaves having been measured in equal numbers, or as the result of a nearly equal distribution of the members of a pair of alleles differently affecting lamina growth in the area collected. We have no data to test the second hypothesis, but have re-examined and remeasured the leaves and have, we believe, eliminated the possibility of error in choice of leaves.
Th taxonomic disposition of such aggregations of plants as that which we now call Viola nephrophylia still, of course, remains subject to the caprices of taxonomists who, under the present international rules, may justify their nomenclatural decisions by reference to their intuitive judg- ments. In this exploratory study, considering only a few samples in a part of the range of V. nephrophylla, we have submitted what we feel is good evidence for the rejection of the synonyms listed earlier for the Rocky
1963 | RUSSELL AND CROSSWHITE: VIOLA 65
95 zi -9 85 75 - -0 -9-0 —O me | $0 T 65 -4 as - ~ 9 55 7 -9 4 to-0 i : 4% BB S © f a) 4 > fo! < a6 z 35 =) é é 25 e FSC 1l1450A
15 25 35 45 55 65 75 85 95
LAMINA LENGTH Fic. 8. Pictorialized scatter diagram for population sample, Crosswhite 1450A.
Mountain area. More elaborate studies are in progress on populations from the eastern and north-central part of the United States.
Arizona State University Department of Botany, Tempe, Arizona
LITERATURE CITED
ANDERSON, E. 1936. Hybridization in American tradescantias. Ann. Mo. Bot. Gard. 23:511-525. BECKER, W. 1925. Viola in Engler, A. and K. Prantl, Die Naturliche Pflanzenfamilien, ed. 2, vol. 21:363-376. BRAINERD, E. 1921. Violets of North America. Vt. Agr. Exp. Sta. Bull. 224. RoceErs, R. J., AND T. T. TAnrmorTo. 1960. A computer program for classifying plants. Science 132:1115-1118. RusseEL1, N. H. 1955. Local introgression between Viola cucullata Ait. and V. septen- triontalis Greene. Evolution 9:436—440. . 1956a. Techniques for species comparison in violets. Proc. Iowa Acad. Sci. 63:157-160. . 1956b. Regional variation patterns in the stemless white violets. Am. Midl. Nat. 56:491-503. . 1961. The development of an operational approach in plant taxonomy. Syst. Zool. 10:159-167. . 1962. A different approach to taxonomy. Assoc. Intern. Syst., Caen, France.
66 MADRONO [Vol]. 17
REVIEW
Flora of New Zealand. By H. H. Arran. Vol. I. Indigenous Tracheophyta: Psilopsida, Lycopsida, Filicopsida, Gymnospermae, Dicotyledones. Government Printer, Wellington. liv + 1085. (April) 1961. $14.70.
Although it was never published, the first flora of New Zealand was compiled nearly two centuries ago by Daniel Solander, a student of Linnaeus. Following it were works by notable botanists such as Allan Cunningham, Joseph Hooker, Thomas Kirk, and T. F. Cheeseman. This first volume of the newest “Flora of New Zealand” by H.H. Allan follows in the best tradition of New Zealand botany and is a distin- guished contribution to the world’s botanical literature.
The part issued includes all vascular plants indigenous to New Zealand except monocotyledons, presumably to be the subject of a later volume. Solander’s manu- script described 360 species of vascular plants; Allan deals with 1457 species and 280 varieties of native vascular cryptogams, gymnosperms, and dictoyledons. If a rough estimate of the number of monocotyledons were added, the total number of native taxa would considerably exceed 2000. Allan originally conceived a revision of the second edition of Cheeseman’s “Manual of the New Zealand Flora” (1925) which was out-of-date and had long been unobtainable, but it became clear that a complete reworking of the flora was necessary. Unfortunately, Allan did not live to see the publication of the work to which he had devoted so much of his time and energy. It was completed and guided through the press under the able and sympa- thetic direction of Lucy B. Moore. Most of the Flora was prepared by Allan, but sev- eral groups were treated by Miss Moore, including difficult genera such as Myosotis and all of Hebe except the whipcord hebes, and others were handled by M. B. Ashwin, who contributed sections on Euphrasia, the whipcord hebes, Parahebe, and Pygmea.
The preface to the work is followed by several pages of useful annals devoted to bibliographic citations and short notes on botanical literature relevant to New Zealand covering the period from Solander’s manuscript of 1769 to papers published in 1958. A short section discusses the New Zealand Botanical Region, taken to in- clude the Kermadec Islands, Three Kings Islands, Chatham Islands, the subantarctic islands (Antipodes, Aucklands, Campbell, Macquarie, and the Snares), and the three main islands of New Zealand. Next are pages explaining abbreviations used in the text, a list of authors of New Zealand taxa, and a synopsis of the classes and orders of the plant groups covered by the Flora. The arrangement of ferns follows a system by Holttum and that of the dicotyledons follows the first edition of Hutchinson’s “The Families of Flowering Plants.” The artificial keys to the families of dicotyle- dons and to the genera are praiseworthy for their simplicity; they seldom use more than single pairs of characters for making a choice. If satisfaction fails here, there are additional generic keys following each family listing in the main portion of the book.
New taxa are circumscribed in English in the text with their Latin diagnoses appended in a special section. Also included are a glossary covering technical termi- nology used in the descriptions and a series of drawings illustrating terms describing leaf morphology. The alphabetical list of Maori plant names will be of special use to visitors, since even professional botanists in New Zealand often refer to plants by names such as kowhai, ti, rimu, manuka, puriri, and pohutukawa. Some of these names have inspired genera such as Hoheria from houhere, Tupeia from tapia, Corokia from korokio, and additional specific names such as totara, maire, and taraire. A lengthy but useful group of supplementary notes precedes the index, the final section of the book. These notes make some typographical or factual correc- tions, additional comments, and amendments of parts of the text based on recent collections or newly-published monographs. They bring the work up to date through 1960 and correct virtually all errors in the text.
The original and critical nature of the Flora is reflected by the fact that three new genera, 29 new species, and 61 new varieties are described in it. The new genera
1963 | REVIEW 67
are: Kirkophytum, which includes two species formerly referred to Stilbocarpa (Araliaceae) ; Neopanax, to accommodate species formerly found in Nothopanax (Araliaceae) ; and Kirkianella as a monotypic genus for what has unhappily been called Crepis novae-zelandiae, a composite of uncertain affinities. Other genera have been dropped from the New Zealand flora, including Leucopogon, which Allan rele- gated to Cyathodes.
In this volume, 290 genera are recognized. Northern hemisphere botanists may be surprised to find familiar genera such as Myosotis (34 spp.), Ranunculus (43 spp.) and Epilobium (50 spp.) so amply represented. The largest genus is Hebe, with 79 species recognized—this even after the removal from it of peripheral entities such as Pygmea and Parahebe. Next in size are Celmisia (58 spp.), Epilobium (50 spp.), and Coprosma (45 spp.). Nearly two-thirds of the three dozen endemic genera are monotypic. The vascular flora as a whole is about 80 per cent endemic according to an estimate made some years ago by Cockayne. Each island in the botanical province has its own endemic flora, and a number of these species have extremely limited ranges, being known only from a few or single colonies. Of particular phyto- geographical interest are the Three Kings Islands, only 8 square kilometers in area, which support about a dozen endemic species and two endemic genera. One of these genera, Plectomirtha (the only representative of the Anacardiaceae in New Zealand), is known only as a single tree! At least two species, Tecomanthe speciosa (the only member of the Bignoniaceae in New Zealand) and Alectryon grandis (Sapindaceae) occur only as single individuals on these islands!
Introduced plants were not recorded in this volume, but in view of their large number (576 species reported by Cheeseman in 1925) their omission is understand- able. Since many of these introduced species belong to genera not otherwise found in New Zealand, there is little chance that they will be confused with indigenous taxa. Species such as Oxalis corniculata and Picris hieracioides were included because they were collected early in the 19th century only shortly after settlement of New Zealand began, but they are probably introductions. Others, such as Sonchus asper, presum- ably arrived with the Maori, since they were collected by the first Europeans to land in New Zealand.
In the text, the species are given ample descriptions with a full author citation, place of publication, type locality, and notes on range and habitat. Synonymy is not complete, but covers most names relevant to New Zealand material. Many regional floras serve only as guides to identification and little else, giving no hint as to the botanical problems present in the area covered. But, as pointed out in the preface to the Flora, “it is recognized that many species are inadequately known and a second, but not necessarily secondary, objective has been to indicate directions in which more investigations are needed,” which objective has been admirably fulfilled. Par- ticularly outstanding are the copious notes discussing various problematical specimens in herbaria, and field observations which will be of great help to botanists dealing with these groups in the future. One of the most valuable and novel features of the Flora is the inclusion of sections augmenting the generic treatments with comments on heteroblasty, sexual expression, polymorphy, hybridism, horticultural forms, taxa of uncertain position, taxonomic synopses, notes on special problems, synopses of growth forms, and general field observations and remarks reflecting Allan’s wide knowledge of plants in the field and the literature concerning them.
A recurrent theme in the Flora is the suggestion that hybridization is responsible for the variation within many plant groups. Although both men had published on the subject separately, in 1934 Leonard Cockayne and Allan published a list of nearly 500 wild species-hybrids in the New Zealand flora. This paper must have met con- siderable skepticism and opposition from botanists in other areas of the world, where interspecific sterility barriers are the sine qua non of orthodox biosystematy. Never- theless, Allan received support from subsequent workers in the dominion and else- where, and continued to maintain that hybridization is a potent force in the evolu- tion of the New Zealand flora. It is evident from numerous well-documented exam-
68 MADRONO [Vol. 17
ples that actual, rather than apparent, hybridization does occur and is largely respon- sible for the taxonomic complexity of these groups.
The Flora is illustrated with several excellent line drawings by Nancy Adams, who also designed the attractive dust jacket. To achieve a volume of handbook size, very thin paper was used; the nearly 1100 pages make a book only 2 cm. thick. The copious notes are printed in 6-point type, which seems too small to be read comfortably for very long. The printing and binding are very well done. The small size of the volume should not belie the riches it contains. RoBERT ORNDUFF, Depart- ment of Botany, Duke University, Durham, North Carolina.
NOTES AND NEWS
CHROMOSOME NUMBERS IN CROSSOSOMA. Since the relationships of the small fam- ily Crossosomataceae have been a subject of discussion, it is of interest to record the chromosome numbers of two species of the only genus. Crossosoma californicum Nutt. is confined to Santa Catalina, San Clemente, and Guadalupe islands off the coast of southern California and Baja California, whereas C. bigelovii Wats. is found about the margins of the Sonoran Desert in California, Arizona, Baja California, and So- nora. Crossosoma parviflorum Rob. & Fern. and C. glaucum Small, both described from Arizona, are probably not distinct from C. bigelovii at a specific level, and so the family probably consists of only two species. The chromosome number of C. cali- fornicum was determined from buds collected from Pebbly Beach Canyon near the
Fic. 1. Chromosomes of Crossosoma at meiotic metaphase I, a, C. californicum ; b, C. bigelovii. Both figures * 2600.
water purification plant, Santa Catalina Island, Los Angeles County, California (Taylor & Ornduff 4383, UC); from material propagated at Rancho Santa Ana Botanic Garden, taken from a collection (Wolf 1487, RSA; fig. 1a) made at the junc- tion of Pebbly Beach and Renton Mine roads, Santa Catalina Island; and from material of undetermined origin cultivated in the East Bay Regional Parks. All of these collections had a gametic chromosome number of n=12, with no meiotic irreg- ularities observed, as did a single collection of C. bigelovit from Morongo Valley, Riverside County, California (Davis 105, RSA; fig. 1b). The twelve pairs of rela- tively small chromosomes found in these plants are markedly different from the five very large pairs found in Paeonia (Ranunculaceae), with which Crossosoma has been allied. They are, however, more or less similar to the chromosomes found in a num- ber of other families of angiosperms. PETER H. RAvEN, Division of Systematic Biology, Stanford University, California, and Marion S. Cave, Department of Botany, Uni- versity of California, Berkeley.
INFORMATION FOR CONTRIBUTORS
Manuscripts submitted for publication should not exceed an estimated 20 pages when printed unless the author agree to bear the cost of the ad- ditional pages at the rate of $20 per page. Illustrative materials (includ- ing “typographically difficult” matter) in excess of 30 per cent for papers up to 10 pages and 20 per cent for longer papers are chargeable to the author. Subject to the approval of the Editorial Board, manuscripts may be published ahead of schedule, as additional pages to an issue, provided the author assume the complete cost of publication.
Shorter items, such as range extensions and other biological notes, will be published in condensed form with a suitable title under the general heading, ‘““Notes and News.”
Institutional abbreviations in specimen citations should follow Lanjouw and Stafleu’s list (Index Herbariorum. Part 1. The Herbaria of the World. Utrecht. Second Edition, 1954).
Articles may be submitted to any member of the Editorial Board.
Membership in the California Botanical Society is normally considered a requisite for publication in MaproNo.
MADRONO
A WEST AMERICAN JOURNAL OF BOTANY
A quarterly journal devoted to the publication of botanical re- search, observation, and history. Back volumes may be obtained from the Secretary at the following rates: ~
Volume I, 1916-1929 . . . . = . $5.00 Volume IT, 1930-1934 . . . . = . §.00 Volume IIT, 1935-1936 oi arn Mh ert eo) Volume IV, 1937-1938 2. % "ia 35208 Volume V, 1939-1940. 2 ... 4%. 5.00 Volume VI, 1941-1942 ote cate akee MORO Volume VII, 1943-1944 . . . . 5.00 Volume VITI, 1945-1946 . . . . 5.00 Volume IX, 1947-1948 Re a Onl) Volume X, 1949-1950... 0 .0) « 2 9 ag00 Volume XI, 1951-1952 Re ee ee te 0, Volume XII, 1953-1954 . . . . 8.00 Volume XIII, 1955-1956 . . . . 8.00 Volume XIV, 1957-1958 . . . . £8.00 Volume XV, 1959-1960 oe, Ri em EO Volume XVI, 1961-1962 2 ican a. e200
Single numbers:
Volumes and il 2)" 209 ae eee 50 Volumes II through XIV. . ..... 1.00 Volume Sev toc: eee ee 25 Volume’ XVI. 3. es eet eee #250
The subscription price of MapRoNo is $6.00 per year. If your institution does not now subscribe to MADRONo, we would be grateful if you would make the necessary request. Since there is no extra charge for institutional subscriptions, an individual may hold membership in the California Botanical Societv on the basis of his institution’s subscription.
Address all orders to:
Corresponding Secretary California Botanical Society Department of Botany University of California, Berkeley 4, California
x ee ee 6 Sea SS SS
Aes MADRONO
VOLUME 17, NUMBER 3 JULY, 1963
Contents PAGE
A NEw SPECIES OF ISOPYRUM ENDEMIC TO THE QUEEN CHARLOTTE ISLANDS OF BRITISH COLUMBIA AND ITS RELATION TO OTHER SPECIES IN THE GENUS, J. A. Calder and R. L. Taylor 69
CYTOPHYLETIC ANALYSIS OF HYMENOXYS ODORATA: A RECAPITULATION, B. L. Turner (th
A NOoTE ON TAXONOMIC CHARACTERS IN LOLIUM, Frank C. Vasek and J. Kirk Ferguson 79
CLEISTOGAMY IN THE MALVACEAE, Paul A. Fryxell 83
SWALLENIA, A NEW NAME FOR THE CALIFORNIA GENUS ECTOSPERMA (GRAMINEAE), Thomas R. Soderstrom and Henry F. Decker 88
Reviews: Irving W. Knobloch and Donovan S. Correll, Ferns and Fern Allies of Chihuahua, Mexico (Rolla M. Tryon); K. R. Sporne, The Morphology of Pteri- dophytes (Job Kuijt); C. Leo Hitchcock, Arthur Cronquist, Marion Ownbey, and J. W. Thompson, Vascular Plants of the Pacific Northwest (Robert Ornduff) 89
NoTES AND News: NEW DISTRIBUTIONS FOR FOUR GRASSES IN OREGON, Kenton L. Chambers and La Rea J. Dennis; CALypso BULBOSA IN THE SANTA CRUZ Mountains, Thomas A. Crandall; NoTES ON TETRA- COCCUS AND CHILOPSIS IN BAJA CALIFORNIA, MEXICO, Edmund C. Jaeger 91
A WEST AMERICAN JOURNAL OF BOTANY
‘PUBLISHED QUARTERLY BY THE CALIFORNIA BOTANICAL SOCIETY
MADRONO
A WEST AMERICAN JOURNAL OF BOTANY
Entered as second-class matter at the post office at Berkeley, California, January 29,
1954, under the Act of Congress of March 3, 1879. Established 1916. Subscription price
$6.00 per year. Published quarterly and issued from the office of Madrofo, Herbarium, Life Sciences Building, University of California, Berkeley 4, California.
BOARD OF EDITORS
EpcAR ANDERSON, Missouri Botanical Garden, St. Louis LymMAN BENSON, Pomona College, Claremont, California HERBERT F. COPELAND, Sacramento College, Sacramento, California Joun F. Davipson, University of Nebraska, Lincoln MivprepD E. MATHIAS, University of California, Los Angeles 24 ROBERT ORNDUFF, University of California, Berkeley MaArRIon OwNnBEY, State College of Washington, Pullman REED C. RoLiins, Gray Herbarium, Harvard University Ira L. Wiccins, Stanford University, Stanford, California
Editor—JoHN H. THoMAS Dudley Herbarium, Stanford University, Stanford, California
Business Manager and Treasurer—Douglas M. Post Biology Department, San Francisco State College 1600 Holloway Avenue, San Francisco 27, California
CALIFORNIA BOTANICAL SOCIETY, INC.
President: Herbert L. Mason, Department of Botany, University of California, Berkeley. First Vice-President: Paul C. Silva, Department of Botany, University of California, Berkeley. Second Vice-President: Robert F. Hoover, California State Polytechnic College, San Luis Obispo. Recording Secretary: Mary L. Bowerman, Department of Botany, University of California, Berkeley. Corresponding Secretary, Margaret Bergseng, Department of Botany, University of California, Berkeley. Treasurer: Douglas M. Post, Biology Department, San Francisco State College, San Francisco, California.
1963 | CALDER & TAYLOR: ISOPYRUM 69
A NEW SPECIES OF ISOPYRUM ENDEMIC TO THE QUEEN CHARLOTTE ISLANDS OF BRITISH COLUMBIA AND ITS RELATION TO OTHER SPECIES IN THE GENUS!
J. A. CALDER AND R. L. TAYLOR
In 1957, a party composed of D. B. O. Savile and the authors carried out a botanical survey of the Queen Charlotte Islands from late May until the latter part of August. During the course of the survey, several collec- tions of a new taxon belonging to the genus /sopyrum were made in the Queen Charlotte Ranges. This paper is devoted to the description of this new species and its relation to other members of the genus.
Four species of /sopyrum are presently recognized in North America. Three are found in western United States and their distributions include southern Washington, Oregon, and California. The fourth species, /. biter- natum, is restricted to eastern United States and Canada. The new species, J. savilez, is completely disjunct from these four taxa and is found only in a small area of the Queen Charlotte Islands.
Isopyrum savilet is the fourth endemic to be described from this region as a result of the 1957 survey, the others being: Saxifraga taylori (Calder and Savile, 1959), Saxifraga punctata ssp. carlottae (Calder and Savile, 1960), and Ligusticum calderit (Mathias and Constance, 1959). Another distinct endemic, Senecio newcombet, was previously described by E. L. Greene in 1897 from a collection made by C. F. Newcombe during a survey of the Queen Charlotte Islands in the same year. In addition to these endemic taxa, there still remain a few undescribed entities in our collection from the alpine and subalpine areas in the Queen Charlotte Ranges. Although the endemics are few in number, the degree of endem- ism is high for such a small flora. The phytogeographic significance of this endemism will be fully discussed in a forthcoming treatment of the flora of the Queen Charlotte Islands.
We would like to express our appreciation to the curators of the follow- ing herbaria for the loan of specimens or the opportunity to examine material in their respective institutions: University of California, Berke- ley; University of Oregon; Peck Herbarium, Willamette University; University of Washington; Washington State University; University of Wyoming; New York Botanical Garden; and Gray Herbarium, Harvard University. We would like to express our appreciation to the artist, Miss C. Mentges for the excellent illustrations, to B. Boivin for the Latin diagnosis, and to C. Crompton for technical assistance.
It isa pleasure to name this species after our close friend and colleague, D. B. O. Savile, who has collected widely in the Pacific Northwest and
1 Contribution No. 259 from the Plant Research Institute, Research Branch, Canada Department of Agriculture, Ottawa, Ontario.
Maprono, Vol. 17, No. 3, pp. 69-92. July 15, 1963.
70 MADRONO [Vol. 17
whose suggestions and stimulating discussions have provided a greater insight into the botanical problems of this region.
KEY TO THE NORTH AMERICAN SPECIES OF ISOPYRUM
Tepals 3.5-6.0 mm long, veins few and prominent; filaments flat, narrowly tri- angular; follicles stipitate. Oregon and northern California............ I. stipitatum Tepals 7.0-17.0 mm long, veins many and inconspicuous; filaments filiform, clavel- late to clavate; follicles sessile. Leaflets puberulent beneath; flowers in a cyme. Southern Washington and north- ern: Oresoni 2. eee ee ee I. hallii Leaflets glabrous beneath; flowers solitary. Roots never tuber-like; lobes of leaflets with a shallow glandular notch at apices; tepals usually 12.0-16.0 mm long. Queen Charlotte Islands................ I. savilei Roots often tuber-like; lobes of leaflets glandular apiculate at apices; tepals usually 7.0-10.5 mm long. Tuber-like roots, fasciculate; follicles 10.0-12.0 mm long; styles recurved,
1.0 mm long or less. Central and southern California............ I. occidentale Tuber-like roots never fasciculate; follicles 5.0-6.5 mm long; styles straight, ca. 1.5 mm: long. Hastern North Americas: I. biternatum
Isopyrum savilei Calder and Taylor, sp. nov. Perenne, erectum, gla- brum, valde rhixomatiforme (10.5)—15.0-31.0-(36.0) cm; folia inferne glaucina, bi-ternatisecta; flores solitarii, terminales vel axillare; tepala 5, alba, decidua (9.8)—12.6-15.0-(16.8) mm long; (6.9)—8.2—10.2—(11.2) mm lat.; stamina 40—60, filamentis filiformis, clavatis, 5.0-8.0 mm long; carpellis sessilibus, 2—8; folliculi dense aggregati, arcuati, 11.0-15.0 mm long; semina 2-8, laevigata, ovoidea, apiculata cum raphid, 2.0—-2.3 mm long.
A delicate upright perennial, glabrous throughout, strongly rhizoma- ous (10.5)—15.0-31.0-(36.0) cm high; leaves glaucous beneath, twice ternately compound, leaflets strongly 2—3-lobed, lobules entire to 3-lobed, with shallow glandular notches at apices, basal leaves usually one, cauline 1—-several; flowers solitary, terminal or axillary; tepals 5, white, occa- sionally tinged pink at apex, readily deciduous, (9.8)—12.6—15.0—-(16.8) mm long, (6.9)—8.2-10.2-(11.2) mm wide; stamens usually 40-60, filaments filiform, clavate, 5.0-8.0 mm long; carpels sessile, 2—8; fruit a head of upright to strongly arcuate follicles, follicles 11.0-15.0 mm. long with recurved beaks; seeds 2-8, essentially smooth, ovoid, promi- nently apiculate with distinct raphe, 2.0—2.3 mm long.
Type: 20 miles south of Moresby Logging Camp near an alpine lake, Moresby Island, Queen Charlotte Islands, British Columbia, Calder et al. 23055 (DAO).
GRAHAM IsLAND: Empire Anchorage, Athlow Bay, Calder & Savile 21464; head of McClinton Bay, Masset Inlet, Calder et al. 21578; east side of Shields Bay, Rennell Sound, Calder & Taylor 23294; mountain north of Mt. Stapleton, Shields Bay, Rennell Sound. Calder & Taylor 23375. MorEesBy IsLtanp: Mt. de la Touche, Fairfax Inlet, Tasu Sound, Calder & Taylor 23566; mountain at west end of Mosquito Lake, Caider & Taylor 23721; 20 miles south of Moresby Logging Camp near an alpine lake, Foster & Joslin 56 (UBC); Tasu Inlet, June 26, 1961, Foster & Bigg (UBC).
Isopyrum savilei is restricted to the Queen Charlotte Ranges at high elevations except on the west coast where subalpine conditions extend
1963 | CALDER & TAYLOR: ISOPYRUM (p
Fic. 1. Isopyrum savilei: A, habit, « ™%; B, stamen, x 5; C, fruit, x 2; D, seed, se 10:
down to sea level. It is a species usually found in moist, shady, rock run- nels or in cliff crevices, but it occasionally extends onto talus slopes where suitable habitats exist. It is associated with many species, but is fre- quently found with Lloydia serotina, Saxifraga mertensiana, Anemone narcissifiora (s.l.), Romanzo fia sitchensis, Heuchera glabra, and Pingui-
WD MADRONO [Vol. 17
Fic. 2. Leaflets of North American species of Jsopyrum: A, I. savilei; B, I. biter- natum; C, I. occidentale; D, I. hallii; E, I. stipitatum. (All ca. X 2.)
cula vulgaris. Although noted in all alpine areas surveyed, it was never a conspicuous element of the vegetation.
A detailed comparison has been made between the five North American species, emphasizing those morphological characters which we feel are most diagnostic (table 1). We fully realize that additional characters such as the shape of the leaflets, stamen number, and follicle shape, which
1963 | CALDER & TAYLOR: ISOPYRUM fle
have been used by other authors, could have been included. However, the number of stamens and shape of the follicle is variable and cannot be used readily to separate the species. On the other hand, leaflet characters are distinct, but difficult to describe adequately. For this reason we have included detailed morphological comparisons of actual leaves utilizing a chloral hydrate/sodium hydroxide clearing technique (fig. 2). This method of comparison of leaf types clearly shows the venation patterns, lobule apiculation (or lack of same), and the degree and types of lobing. The distribution of the four western species is shown in Figure 3.
Isopyrum savilei is strikingly distinct from the other specie of /so- pyrum that occur in North America in several morphological characters, e.g., the strongly rhizomatous nature of the root system, the shallowly notched tips of the ultimate leaf segments, the large showy flowers, the arcuate follicles, and the essentially smooth apiculate seeds with promi- nent raphes (fig. 1).
The discovery of this endemic is significant as it provides further evi- dence of the close relationships between the North American and Japa- nese species. Close scrutiny of the distinctive western North American species, Jsopyrum hall, reveals that it possesses many similar charac- ters to the Japanese species, /. raddeanum, such as: pubescence of leaves, apiculate tipped leaflets, and seed coat characters. These observations confirm those of Drummond and Hutchinson (1920. p. 154) who stated, “The remarkably close affinity of two species of this genus, /. [nemuion | Raddeanum from Manchuria, and FE. Hallw from Oregon, is worthy of note.” Another group of Japanese species have glandular, notched tips on the ultimate leaf segments, smooth seeds, rhizomatous root systems, and usually two carpels; characters which are also found in Jsopyrum savilet. On the other hand, /. savilei also shows close relationships with the American taxa with respect to lack of staminodia, clavate filaments, follicle size, and deeper lobation of leaf segments.
The North American species have been segregtated under Enemion (Drummond and Hutchinson, 1920) and this segregation was based on the tenuous character of carpel number and the presence or absence of petals; two morphological units which are extremely difficult to eval- uate in the family Ranunculaceae. They recognized seven genera; how- ever, only /sopyrum and Enemion are pertinent to the present discus- sion. Hnemion was proposed by Rafinesque (1821) to include a group of species differing from /sopyrum by the absence of petals and this sepa- ration was supported by Drummond and Hutchinson. It should be em- phasized that the use of the terms petals and sepals with respect to the genera in question was not supported by anatomical studies by either Rafinesque, or Drummond and Hutchinson. Indeed, the latter authors have based their segregation primarily on phyletic grounds rather than on critical evaluation of morphological characters. We believe the outer showy organs arranged in a spiral fashion are not sepals, but are best classified as tepals in accordance with modern terminology. In addition
74
MADRONO
[Vol. 17
e. bY Ks) WZ
‘>
Lsovilee
L halts
L.stip/tatum I. occidentale FZ
Fic. 3. The geographical distributions of western North American Jsopyrum.
WwW
CALDER & TAYLOR: ISOPYRUM
1963 |
9}e[noIde
ATyvomM pue ‘punol ‘gsosni A][3UO0I}s ‘SUO] WW /*[-p'T a}ye}yIds ! passoid -de pue yystidn
: SUO[ UW Q'8-0'S Jepnsuel} A[MOIeU -Snouriquisw ‘yey SuO|T WU 8°7-6'T [el1a}eW polip
UI JUIUTWIOId SUTOA :OpIM WU 7° 7-S'T ‘SUO] WW (Q'9)- s's-2'p—-(S'e) ‘ory IIM0IqO ‘paley A[suo01}s xode
AIBPI[OS SIIMOLY
snoiqeys !ayepnoide Ie[npurls saqo] $JOOI (UWIIOJISN}] ) a¥T-19qn} ‘pauaysry} ‘JAOYS JO VpOIOse] :SNo}eWLOZIYI OU
Ws (S°2T)-O'IT-S's WN4D41G 14S * J
dye[noide ATyeom pue ‘proao ‘aso[n3sni :SU0[ WU 7°7-0'Z a[ISSas !yUISIBAIP AJPPIM 0} yYUsIIdN :SUO] WW O'/-O'p I} RAPID
APYPIM ‘UAOF YY
SUO] WW 7Z°8-]'S
[el1a}e8Ul potip
UI snondIdsuOdUT SUIOA ‘ OPIM WUT Co—E"¢
: SUO] WW ¢°6—-S*'Y ‘OTM
poiey A[BUOI}S JaAoU xode
ISOWAD SIIMOT]
y}yeaueq JUs[N -daqnd ‘oazernoide Ie[Npur[s soqo]
4904S}001 APOOM ‘ynoys ‘y10Ys -SnOJeUIOZIYI JOU
wd (0°L9)-O'S9-0'S¢S myDY *]
yuadsaqnd
AJaynutu ‘}UsuTUTOId oydei ‘proao ‘asopn3ni APysys 10 y}oouS ‘SUO, WW 9Q°7-1'Z J[ISses !JUIBIIAIP ATapIM 0} YYysIUIdN
- sUO,T WU S°9-0'S
IJET[PAP]O ‘WO FT,
suo, Wu 7°¢~-1°7
[PI19}eW Pelip UI sno -noidsuosutl suraA Sapir Wu 9'9-7'p ‘suo, WU (ZIT)-T6-Z'2 “oy poiey
A[SuO1}s JaAoU xode
AIeP]OS SIOMOY
SnNOIqe[S 5 9}e|
-noide re[npurys saqo] SIITIISEJ
UI JaAdU Jnq ‘sno
-19qN} SaWIT}IWIOS s}OOI1 :Sno}ewoziyt ApyVoM
Ws O'Te-O'TT WUNIDULIILG "J
aye[noido AyT}UoU -Tm01d pue ‘ploAo ‘yyoows A][eUassa ‘BuO, WU ¢'7-0'Z
a[ISsos a} eNII1e A[Su01}s 0} JYysIAdN ‘ 3UO] WU O'S T—-O'TT
9} PAPO APYeaM ‘ULIO FY
BUOT WW Q'8-0'S
[el1oyeW palip ul Snonotdsuoour suloA
-OpIm WW (7° TT)— 201-78
—(6°9) ‘Buoy WU (g°9T)— O'ST-9°ZI-(8'6) ‘9}TYM
poley A[BUO01}S JaAouU xode
AIE}OS SIIMOY
snoiqeys - Yoyou Ae[NyF ueps MOT[VYS B YIM soqo]
WIOflUdl ‘asopnsni
: SUO[ WU Q'7—-/'T J[ISSIS | }JUIBIIAIP AJapIM 0} 7YBIUdN :SUO] WU Q'ZI-O'0I I}BALTO
AP{RIM ‘UWIOFTTY
suo] WU ¢'S—-7'7
[el10}eW ~pelip
UI SNONSTdSUOSUT SUTIOA -Opim wut /°o-8'¢-(Z'¢) “SUOT UWL (E"1T)—£°6-0'L yuid Ay[euotsess0 ‘aqyTyM
poley A[BUOI}S IaAoU xode
AIVPIOS SIOMOLT
SNOIqR]S $9}°] -noide aepnpurys saqo]
SJOOI (UIIOJISN]) dy]
pousayoty} JOU s}OOI -Jaqny ‘pauayory} ‘y10Ys Jo -SnO}eULOZIYA A[SUOI}S a[IIOSeJ {Ssno}eWOZIYyI jou
Wd (O'9¢)— (OP ES Ok E(USu oy) 1a]LAUDS *]
wd (0'Or)- 0'0£-0'ST-(0'L) 91D]UIP1IIO “J
WOUAdOST JO SAIOUdS NVOIWAWY HLYON AHL JO NOSIYVdWOD YW ‘1 ATav
Spaas
SofH[O} dIN}eIA]
S}UIUILTI
SUITES
sjeday,
apunpedg
JUDISIIOYUT
Sjoyea’]
UI9}SAS JOO
1Y4SIOH
76 MADRONO [Vol. 17
to the tepals, Drummond and Hutchinson have noted small petaloid structures in the European /. thalictroides, and stalked and bilobate structures in the Japanese species, /. stolonifera and I. trachyspermum, respectively. We feel they have misinterpreted these petaloid structures as petals and have subsequently placed too much stress on their value as characters for segregation of the genera. After careful examination of these structures, we have concluded they are staminodia and represent a transitional series from a sessile petaloid organ to one which is stalked bilobate. These structures are readily identifiable as staminodia and close- ly approximate the stamens with clavellate filaments found in several Isopyrum species. It should be pointed out that we have not completed a detailed ontogenetic study of these structures.
Drummond and Hutchinson considered that both /sopyrum and Ene- mion were derived from Paraquilegia, a “primitive ancestoral”’ genus comprised of four species from the mountainous regions of central and southern Asia, and that they represent separate and distinct lines of divergence. We have no evidence that Paraquilegia does not represent the ancestoral progenitor of this complex and we do not disagree with their concept that a ‘“‘natural” group exists in Japan and North America. How- ever, we do think that the two groups, i.e., the Japanese and the North American, do not represent divergent lines of evolution, but rather that they represent overlapping stages within a single line of development. These two groups are closely related and there is no significant morpho- logical evidence for separating them into two separate and distinct genera, hence we consider that the North American taxa belong in the genus Isopyrum.
Plant Research Institute
Canada Department of Agriculture Ottawa, Ontario
LITERATURE CITED
Caper, J. A. and D.B.O. Savite. 1959. Studies in Saxifragaceae—II. Saxifraga sect.
Trachyphyllum in North America. Brittonia 11:228-249. . 1960. Studies in Saxifragaceae. III. Saxifraga odontoloma and lyallii, and
North American subspecies of S. punctata. Canad. Jour. Bot. 38:409-435.
DrumMmMonpD, J. R. and J. Hutcurnson. 1920. XXIII. A revision of Isopyrum (Ra- nunculaceae) and its nearer allies. Kew Bull. No. 5:145-169.
GREENE, E. L. 1897. New or noteworthy species XX. Pittonia 3:249.
Maruzas, M. and L. Constance. 1959. New North American Umbelliferae—III. Bull. Torrey Club 86:374-382.
RAFINESQUE, C.S. 1820. Enemion biternatum. Jour. Phys. Chim. Hist. Nat. 91:70.
1963 | TURNER: HYMENOXYS del
CYTOPHYLETIC ANALYSIS OF HYMENOXYS ODORATA: A RECAPITULATION!
B. L. TURNER
In a recent article in Madrono, Speese and Baldwin (1963)? stated, “The basic number for H. odorata is 11. (We assume the report [sic] of n = 15 for this species to be incorrect.)’’ In spite of this statement, the authors succinctly summarized the reported chromosome counts for H. odorata by referring to 3 populational counts from the United States (reported independently by 2 different groups of workers) as being di- ploid with n = 11; they also referred to chromosome counts by myself from 2 Mexican populations which had n = 15. To convince the reader that the published count of n = 15 for H. odorata might be in error, they pointed out that this count (Turner, Beaman & Rock, 1961) was made from “‘pollen-mother-cell smears of buds fixed in the field during the sum- mer of 1959. Our experience has been that preparations from material so fixed are often difficult to interpret, and especially so if the weather were hot at the time of fixation.” This, in spite of the fact that a camera lucida drawing included in the published account showed a meiotic figure with n = 15.
Upon reading Speese and Baldwin’s comments, I felt compelled to take a second look. The following account, though phrased in an admittedly personal way, is my version of the story:
When I first received the pickled buds of H. odorata from Mexico in 1960, I was reluctant to examine these since I recognized the species as
TABLE 1. CHROMOSOME COUNTS OF HYMENOXYS ODORATA FROM MEXICO
COLLECTION CHROMOSOME COUNT SOURCE
Coahuila: 12 mi S of Saltillo al (buds)
Powell & Edmondson 528. TEX 211 = 30 (root tips) Nuevo Leon: 39 mi S of Saltillo
Powell & Edmondson 543. TEX one = 30 (root tips) Nuevo Leon: 24 mi S of Galeana
Crutchfield & Johnston 5860. TEX if ema Et (buds) Nuevo Leon: 41.2 mi S of Saltillo
Rock M264. TEX n= 15 (buds) Nuevo Leon: 1 mi S of San Roberto.
Thompson & Doolin 2163. TEX N15 (buds)
1 The rhetorical definition of recapitulation is preferred here being “A form of peroration in which the respective processes, as of explanation, conviction, excita- tion, and persuasion, pursued in a discourse, are concisely repeated for the purpose of more complete effect.” (Funk and Wagnalls, New Standard Dictionary, 1945 edition.)
2 The authors were apparently unaware of an earlier report of n= 15 for H. anthemoides from Argentina (Solbrig, 1962).
78 MADRONO [Vol. 17
being one already counted by Speese and Baldwin (1952). However, I thought the single previous count needed checking, so I examined the material and to my surprise it showed n — 15. I double checked this by counting the meiotic material from several different florets and heads; all counts were n — 15. As indicated by Speese and Baldwin in their ref- erence to a Johnston collection (5860), also from Mexico, I again counted n = 15 for H. odorata; this time I was not surprised, but I noted in my lab book and indicated on the collector’s label, ““‘n = 15, clearly, det. B. L. Turner from PMC’s.”’
My next encounter with the chromosomes of H. odorata came in the fall of 1961 when a graduate research assistant, A. M. Powell, counted n = 15 for a collection of his from Nuevo Leon, Mexico, from the same general area of the previous Mexican collections with n = 15. By this time, / was beginning to question the counts of n = 11 reported for H. odorata by Speese and Baldwin (1952) as well as those of Raven and Kyhos (1961). My curiosity now being whetted, I collected in the spring of 1962, buds from 3 populations of H. odorata (2 in Texas and 1 in California, the latter from the same area from which Raven and Kyhos’ counts were obtained). I was delighted to find n = 11 in all these collec- tions (Powell and Turner, 1963); my confidence in my scientific col- leagues now re-established I forgot the issue until the recent publication of Speese and Baldwin (1962), statements from which are quoted above.
My first reaction to Baldwin and Speese’s comments was mild irrita- tion; this soon gave way to the haunting fear that my observational senses were being affected by the too frequent exposure to acetocarmine fumes, to say nothing of the PDB to which we are all accustomed; finally I couldn’t bear the onus of my conscience and decided to germi- nate seeds from the original Mexican collection or collections to deter- mine if indeed the mitotic counts might not tell a different story, one told from the cool confines of a petri dish instead of the hot atmospheric con- finement accorded the original material. In spite of the trivial