'«/ PHYTOLOGIA

Bs: An international journal to expedite plant systematic, phytogeographical
oh and ecological publication

Vol. 83 7 July 1997 ; No, 1

CONTENTS

CHEMNICK, J. T.J. GREGORY, & S. SALAS-MORALES, A revision of
Dioon tomasellii (Zamiaceae) from western México, a range extension of e

a merolae, and clarification of D. purpusit.......66...0 ese. eeeee nyse eet eteeen eee: o
- . NESOM, G.L., Review: “A revision of Heterotheca sect. Phyllotheca maine)
Harms (Compositae: Astereaey by d.C. Semple: occa Ce a 7 sleds

-“HALWARD, T., A. HILL; & R. SHAW, RAPD analysis of genetic diversity

es among and within populations of Balduina atropurpurea at Fort Stewart,

Br Geor tia 24 Betis ie ais ches Catenin fect. ye oan) 4 i het
-. TURNER, B.L., Sabazia lapsensis, a new name for S. breedlovei B.L. Turner

_-. (1987) (Asteraceae, Heliantheae), non S. breedlovei B.L. Turner (1976).. 34 ~~
_ SIDDIQUI, M.A., M. ATHAR, & S. AHMED, Effect of farmyard manure and.

mineral fertilizers on coloration, growth, and biomass production of Azolla

as pinnata. Smita srene 6 MA eRe TS Wolcw bk aie siGalb eA iptamla este Calas gc Melee Ww Reds de aw Caw ao Sate estore d hb 36
: ENGEL. Vids Studies « on Geocalycaceae (Hepaticae). X. New taxa and new
combinations in Chiloscyphus Corda for Australasia.,..............00...0.00. 42

~CHEMNICK, J. T.J. GREGORY, & S. SALAS-MORALES, Ceratozamia
mixeorum (Zamiaceae), a new species from Oaxaca, México with comments
» © ondistribution, habitat, and species relationships..................0.000e sees Wat

~MACROBERTS, B.R., M.H.. MACROBERTS, L.M. STACEY, & D.C.
MOORE, The status of Parnassia (Saxifragaceae) in the West Gulf Coastal
Be OR AA sey coho ees chan NeRe tp tare SU Ort tates Fak pdcaai gee uaiat s earuels <sOEEE OG te > bane cote 53
= ~ PALACIOS-CHAVEZ, R., M.de LA LUZ ARREGUIN-SANCHEZ, & D.L.
- QUIROZ-GARCIA, Estudio acaestocagy de las Burseraceae del Estado de
Beets CIPRELANGT ACRICO. each pe ue Wu a ary iratie Fs pan se madtabnuwensiacbics reels 58°"
Se ARREGUIN. SANCHEZ, M. de LA LUZ, R. PALACIOS-CHAVEZ, & D.L.
QUIROZ-GARCIA, Morfologia de las esporas del género Pteris para

Masse he se ae ae gs cepts oe RENE 674.
Pierce s, VOMIMOE ai clases rca hel aon Pet uciuae taseaimtiaw ce Gplun tees 719
Index to reviewers, SHOTAERIES OS feo as Ros ca GOR os Carsales Ba AEE es 80

Published by Michael J. Warnock
185 Westridge Drive Huntsville, Texas 77340 U.S.A.
PHYTOLOGIA is printed on acid free paper.

‘ gtk ¥ ce ¢ oe 0
BHYTOLOGIA. (ssn 00319430) 4 publishes monthly wit
Michael J, Warnock, 185 Westridge Drive,
paid at Ne, 3 TK. © 1996 rae

SRT OFS

“ ‘$44.00. Foreign and/or
issue. as issues. volume |

Phytologia (July 1997) 83(1):1-6.

A REVISION OF DIOON TOMASELLII (ZAMIACEAE) FROM WESTERN
MEXICO, A RANGE EXTENSION OF D. MEROLAE, AND CLARIFICATION OF
D. PURPUSII

Jeffrey Chemnick!, Timothy J. Gregory, and S. Salas-Morales2

1114 Conejo Road, Santa Barbara, California 93103 U.S.A.
2Sociedad para el estudio do lo Recursos Bioticos de Oaxaca, A.C. MEXICO

ABSTRACT

Dioon tomasellii was treated by De Luca et al. 1984 as two varieties; var.
tomasellii and var. sonorense. To quote from the description, “Both the
vegetative and reproductive characters show in general continuous variation
patterns that do not support specific segregation within the range of D.
tomasellii but the variation in the populations of Sonora and northern Sinaloa is
such to warrant segregating them as a distinct variety.” Apparently the
reference to a continuous gradient of patterns was based on conjecture and not
observations in the field, as we have examined both living plants and/or
herbarium vouchers of nearly all known populations and have found no pattern
of continuous variation. Our studies show that the two varieties of D.
tomasellii merit recognition at the species level based on: a host of distinct
morphological characters that are maintained even in cultivation; the lack of
continuous variation between the two varieties; different habitat preferences;
and an RFLP analysis by Moretti et al. 1993. A nomenclatural recombination
(Dioon sonorense) is proposed. Some comments follow on Dioon merolae
and D. purpusii.

KEY WORDS: Dioon, Zamiaceae, México, systematics

Populations of Dioon occurring on the west coast of México inhabit the foothills of
the Sierra Madre Occidental ranging from Sonora in the north to Chiapas in the south.
The ranges of D. sonorense (De Luca et al.) J. Chemnick, T. Gregory, & S. Salas-
Morales and D. tomasellii De Luca, P., S. Sabato, & M. Vazquez Torres are detailed
below. Dioon holmgrenii occurs in southern Oaxaca and D. merolae De Luca, P., S.
Sabato, & M. Vazquez Torres ranges from south-eastern Oaxaca to southwestern
Chiapas. The population of D. tomasellii closest to D. sonorense occurs in Durango
but is decidedly within the described morphological range of other populations of D.
tomasellii further south. Though the Durango plants are much closer geographically to

2 PHY TOLOGIA July 1997 volume 83(1):1-6

the Sonoran populations of D. sonorense than the next closest populations of D.
tomasellii found in Nayarit, no intermediate forms are known to exist. All currently
known populations of D. tomasellii are sufficiently similar to each other to be treated
as a single species and are very different from the known populations of D. sonorense.

A SUMMARY OF CHARACTER DIFFERENCES

Dioon tomasellii can be readily distinguished from D. sonorense by: its fewer but
longer, arching leaves; a thicker rachis and petiole which is densely tomentose when
emerging and occasionally yellow with age; wider, deflexed, falcate, nearly entire,
glabrous dark-green leaflets with conspicuously persistent tomentum on the abaxial
side; and almost no spacing between the margins of the leaflets at the widest point.
Dioon sonorense is distinguished from D. tomasellii by: its crowns of more numerous
leaves which are shorter, upright, sometimes twisted and spirally ascending; a more
slender rachis and petiole which emerge with dense pubescence and generally remain
green with age; considerably narrower, linear-lanceolate leaflets often armed with one
to three small spines on the distal edge of the leaflet; and leaflets that are generally flat,
but occasionally slightly deflexed, or slightly keeled on the petiole and widely spaced
between the margins by almost the width of the leaflets. The newly emerging leaves
of D. tomasellii are covered entirely by a dense golden-brown tomentum and taper
inwardly at the tip. The newly emerging leaves of D. sonorense are light green and
taper outwardly at the tip; the leaflets are only lightly tomentose while the rachis is
covered by a silvery pubescence.

DISTRIBUTION, HABITAT, AND NOTES

Dioon tomasellii is widely but sporadically distributed in oak and oak-pine forest in
the states of Durango, Nayarit, Jalisco, Michoacan, and Guerrero in canyons and
woodlands at elevations ranging from 600-1850 m with an annual rainfall of 1000-
1500 mm. Dioon sonorense is currently distributed entirely within the state of Sonora
(though it has been reported from northern Sinaloa) growing in oak woodland and the
transition zone between high desert (deciduous wet/dry thorn/caudiciform forest) and
oak woodland at an altitude of 615-1200 m. Plants are usually found on steep terrain
growing under extremely dry conditions with an annual rainfall of 250-500 mm.

The taxonomy of Dioon has historically been and still is based almost entirely on
vegetative characters. We are currently examining the megasporophylls of various
species within the genus in order to find other useful characters. We have developed a
process that completely removes the hairs from the megasporophylls and thus reveals
texture, structure, and color beneath. The removal of cone hair is achieved by soaking
the scales in an aqueous solution of 10% w/v sodium hypochloride for 12 hours and
then gently washing them in a steady stream of fresh water. Hopefully other workers
will find this process useful in their search for meaningful cone characters. Because
the systematics of Dioon is not well-understood, we eagerly await the advent of an
accurate and useful DNA fingerprinting process to help determine interspecific

Chemnick et al.: Dioon in Oaxaca 3

relationships and genotypic mutative distance that is invisible to the observer relying
solely upon morphology to determine the disposition of a group of closely allied taxa.
RAPDs hold great promise but are still in the initial stages of application. However,
molecular analysis has already provided some useful insights.

A phylogenetic analysis of all taxa in the genus Dioon was undertaken by Moretti
et al. in 1993 using chloroplast DNA restriction fragment length polymorphism.
Careful examination of their 187 character matrix, drawings, and conclusions support
the separation of D. sonorense and D. tomasellii based on a phenetic approach as
Moretti’s results illustrate (see Moretti et al. 1993, Figure 3). We scored the number
of differences within their character matrix between selected pairings of taxa and found
the following:

tomasellii-sonorense 10differences merolae-califanoi 9 differences
merolae-purpusii 8 differences purpusii-caputot 9 differences
holmgrenii-caputoi 8 differences purpusii-califanoi 6 differences
holmgrenii-purpusii 8 differences edule a.-edule e. 6 differences
holmgrenii-merolae 1 difference spinulosum-rzedowskii 3 differences

There are more differences in the 187 character matrix between Dioon tomasellii
and D. sonorense than between the other pairs above (within the genus Dioon). Thus,
there is molecular evidence to support conferring specific status upon D. tomasellii and
D. sonorense. It is interesting to note that the interspecific comparisons of
spinulosum-rzedowskii and holmgrenii-merolae yielded fewer differences than the
intraspecific comparison of edule angustifolia-edule edule. Perhaps some further
revision of the genus is indicated by these results. We also looked at other pairs of
taxa that we considered more distantly related based on gross morphology to see
whether the RFLP character matrix would support our assessment of those taxa and
scored the differences as follows:

merolae-tomasellii 18 differences purpusii-sonorense 21 differences
caputoi-edule a. 20 differences merolae-sonorense 22 differences
sonorense-edule 29 differences sonorense-spinulosum 82 differences

The results conform with our morphological analyses of the above taxa and yield
nothing that is counter-intuitive to the apparent relationships within the genus Dioon
except the surprising lack of differences between D. holmgrenii and D. merolae; two
taxa that are distinct based on gross morphology. In general, the cpDNA RFLP
analysis seems to be a reliable method for examining interspecific relationships as it
corresponds well with our systematic sense of Dioon based on morphological and
ecological evidence from plants in the field and in cultivation.

In consideration of the ecological, geographical, morphological, and molecular
evidence, this paper therefore confers specific status on:

Dioon sonorense (De Luca et al.) J. Chemnick, T. Gregory, & S. Salas-Morales,
comb. nov. BASIONYM: Dioon tomasellii De Luca, P., S. Sabato, & M.
Vazquez Torres var. sonorense De Luca et al., Brittonia 36:223-227. 1984.

- PHYTOLOGIA July 1997 volume 83(1):1-6

A RANGE EXTENSION FOR DIOON MEROLAE

In their description of Dioon merolae, De Luca et al. (1981) report the distribution
of the species as endemic to the state of Chiapas. They observe that, “It is
noteworthy, furthermore, that D. merolae is well separated orographically from the
other Mexican species by the Isthmus of Tehuantepec.” We wish to report the
existence of two populations of D. merolae in the state of Oaxaca in the Sierra de
Judrez and the Sierra Madre del Sur within the drainage of the Rio Tehuantepec.
These Oaxacan D. merolae are noteworthy because they are quite similar to plants
found in western Chiapas 160 km to the east, yet occur only 30 km east of populations
of Dioon sp. of uncertain affinity, confirming that the Isthmus of Tehuantepec is not
necessarily a geographic barrier to the distribution of the species. We observed
Oaxacan populations of D. merolae during a field trip in May, 1997 and again in
December, 1997. A third population of Oaxacan D. merolae was discovered by S.
Salas-Morales in the eastern region of the Chimalapas growing in oak-pine forest at an
altitude of 810 m. The population in the Sierra Madre del Sur was growing at an
elevation of 1150 m in soil derived from sedimentary rock along a ridge with both SE
and NE exposure in oak/pine forest. The phenology of this population was complex:
mature plants with dried microstrobili and developing megastrobili; active recruitment
of younger but decidedly post-juvenile plants of varying size and several seedlings.
Domestic pigs eat the fruit but, fortunately, pass the seed unharmed. The paucity of
seedlings was probably the result of grazing goats. More than 400 mature plants were
observed. The population in the Sierra de Juarez was growing at an elevation of 1080
m in limestone kaarst with an E/SE exposure. The plants were in association with
Beaucarnea recurvata, Chamaedorea elegans, Agave spp., Plumeria rubra, Bilbergia
sp., Hechtia spp., and Tillandsia brachycaulus. The phenology of the population was
likewise complex; mature plants with dried microstrobili and developing megastrobili;
female plants with recently dehisced cones, and seedlings were observed. The total
number of mature plants observed was in excess of 100 individuals. Herbarium
vouchers from the above localities have been deposited at the Instituto de Ecologia de
Xalapa.

The ethnobotanical use of Oaxacan Dioons is widespread. The leaves are
commonly seen ormamenting windows, doorways, and religious figures during
holidays, especially Easter. Churches often display wreaths of Dioon leaves and
occasionally cultivate plants in the garden to provide a ready source of plant materials
as the closest population of Dioons is often a distant and difficult journey. The
sarcotesta is sometimes a source of food. The sclerotesta is used for games, bracelets,
and necklaces. Dioon leaves are occasionally used in religious and festive ceremonies.
The local common names in the Sierra Madre del Sur are mais viejo (oldtime corn)
and palma espinuda (spiny palm). The sarcotesta has historically been utilized as a
food source in periods of diminished corn supplies. Though none of the authors has
sampled the fleshy yellow sarcotesta, assurance was given that it remains a popular
food item today, prized for its rich flavor. The leaves are used as Christmas party
ornaments and as such, are sold in the markets of Tehuantepec. Occasionally the
mature megastrobili are harvested, apparently by outsiders who sell the cones to a
broker in the port city of Salina Cruz for unspecified uses and destinations but most
likely to meet the foreign demand for propagation. The plant is known locally in the
Sierra de Juarez as palmilla (little palm). It is harvested solely for the leaves which are

Chemnick et al.: Dioon in Oaxaca 5

used in all manner of festivities, Christmas decorations, weddings, and during the
week of Easter. Extreme caution should be exercised whenever traveling or doing
field work in rural mountainous areas of Oaxaca due to the widespread cultivation of
illegal crops. As such, it is imperative to work with a local guide.

CLARIFICATION OF DIOON PURPUSII

Much confusion persists in the proper identification of Dioon purpusii (see Rose
1909) due largely to the brevity of the original description and the remoteness of the
known populations. In spite of the efforts of De Luca et al. (1978) to clarify the
description of D. purpusii, an error persisted in the description of the fronds as “flat in
adult plants, keeled in young plants” and in the comparison between D. purpusii and
D. califanoi which claimed that “It (D. purpusii) differs from the former (D. califanoi)
because its fronds are flat and not keeled (with the exception of the juvenile ones
which are very similar to the fronds of D. califanoi)” (De Luca et al. 1979) .

We visited the type locality in Santa Catarina, Oaxaca as well as three other
populations within the drainages of the Rio La Hondura and the Rio Santo Domingo.
Our awareness of previously unknown populations of Dioon purpusii is the result of
extensive field work by Silvia H. Salas. We examined a number of leaves on various
plants and found that the mature leaves on adult plants are rarely flat; instead they are
moderately to strongly keeled. The leaflets are inserted obliquely on the rachis, angled
forward and usually held at an angle above the rachis of 20- 45 degrees in adult plants
as well as juvenile plants. Unfortunately this misapprehension about flat leaves in D.
purpusii has persisted and undoubtedly contributes to the misidentification and
confusion within this taxon. Consequently, misidentified plants are common in
cultivation. Many Dioons labeled as D. purpusii; with spines on the margin of the
leaflets are likely to be D. merolae, D. holmgrenii, D. caputoi, D. sonorense, or D.
tomasellii, especially if the leaves are flat or deflexed. Dioons with keeled leaves and
spines on the margins of the leaflets are likely to be D. purpusii though D. califanoi
occasions the same habit. De Luca et al. (1980) correctly addressed the leaf aspect of
D. purpusii in the notes of their description of D. caputoi as follows: “Dioon purpusii,
to which were erroneously attributed specimens of D. caputoi, differs in its lightly
keeled fronds. .. .” Sabato & De Luca (1985) amended the matter further in the Key
to Species separating purpusii from califanoi on the basis of “leaf flat or slightly
keeled” rather than “strongly keeled”.

We have recently examined populations of plants in central Oaxaca that apparently
have an affinity with Dioon purpusii, but which produce leaves that are flat to slightly
keeled and moderately to densely tomentose. We are currently cultivating plants of
these populations from seed to compare leaf morphology under uniform conditions.
Further examination of both known and newly discovered populations in what seems
to be emerging as a D. purpusii complex is required to comprehensively determine the
disposition of Oaxacan Dioons.

6 PHYTOLOG?A July 1997 volume 83(1):1-6

ACKNOWLEDGMENTS

We are indebted to Sherwin Carlquist and Dieter Wilken for reviewing the
manuscript. We are grateful to Loran Whitelock for his valuable assistance. We
would also like to thank Peter Fletcher for his help in the field and Leo Schibli of
SERBO.

LITERATURE CITED

De Luca, P. & S. Sabato. 1979 Dioon califanoi (Zamiaceae) a new species from
Mexico. Brittonia. 31:170-173.

De Luca, P., S. Sabato, & M. Vazquez Torres. 1978. Dioon purpusii Rose
(Zamiaceae) a misknown species. Delpinoa 20:31-35.

De Luca, P., S. Sabato, & M. Vazquez Torres. 1980. Dioon caputoi (Zamiaceae) a
new species from Mexico. Brittonia 32:43-46.

De Luca, P., S. Sabato, & M. Vazquez Torres. 1981. Dioon merolae (Zamiaceae) a
new species from Mexico. Brittonia 33:179-185.

De Luca, P., S. Sabato, & M. Vazquez Torres. 1984. Dioon tomasellii (Zamiaceae) a
new species with two varieties from western Mexico. Brittonia 36:223-227.

Moretti, A., P. Caputo, S. Cozzolino, P. De Luca, L. Gaudio, G. Siniscalo Gigliano,
& D.W. Stevenson. 1993. A phylogenetic analysis of Dioon (Zamiaceae).
American Journal of Botany 80:204-214.

Rose, J. N. 1909. Cycadaceae In: Studies of Mexican and Central American Plants.
Contr. U. S. Natl. Herb. 12:260-261.

Sabato, S. & P. De Luca. 1985. Evolutionary trends in Dioon (Zamiaceae).
American Journal of Botany 72:1353-1363.

Phytologia (July 1997) 83(1):7-21.

REVIEW: "A REVISION OF HETEROTHECA SECT. PHYLLOTHECA (NUTT.)
HARMS (COMPOSITAE: ASTEREAE)" BY J.C. SEMPLE

Guy L. Nesom

Texas Research Institute for Environmental Studies, Sam Houston State University,
Huntsville, Texas 77341 U.S.A.

ABSTRACT

A summary and overview are provided for the recent monograph of
Heterotheca sect. Phyllotheca by John Semple (1996), with emphasis on
taxonomic concepts and implications of the formal taxonomic recognition of
sympatric infraspecific taxa. Aspects of the taxonomy of sect. Heterotheca and
sect. Ammodia also are discussed. Three new combinations allow a more
evolutionarily congruent taxonomy for the H. sessiliflora complex: H.
sessiliflora var. thiniicola (Rzed. & Ezc.) Nesom; H. echioides var.
bolanderioides (Semple) Nesom; and H. echioides var. bolanderi (A.
Gray) Nesom. Two combinations necessary in the same group remain to be
formally completed by Semple. Rationale regarding the taxonomic status of
Bradburia (independent genus vs. subgroup within Chrysopsis) is examined.

KEY WORDS: Heterotheca, Bradburia, Chrysopsis, Astereae, Asteraceae,
nomenclature

John Semple (1996) has published “the first comprehensive monograph of the
prairie and montane goldenasters, Heterotheca sect. Phyllotheca (Nutt.) Harms.”
“This study was based on more than 10,300 herbarium specimens (6,844 separate
collection numbers)” and includes specimen citations, typification, and a detailed
illustration and distribution map for every taxonomic entity recognized in the treatment.
An intuitive phylogenetic diagram (p. 6), drawn from molecular and morphological
information, shows Semple’s view of relationships among the goldenaster genera
(subtribe Chrysopsidinae, sensu Nesom 1994) and provides a summary of the
taxonomy and species relationships within sect. Phyllotheca. There are taxonomic
rearrangements, and two new species are described; one species is newly raised from
varietal rank. The treatment also provides background for understanding
nomenclatural combinations in sect. Phyllotheca that were published earlier (Semple
1987, 1992, 1994). Most immediately, the value of the treatment is evident to anyone
needing to identify plants of Heterotheca, but details of the nomenclature,

8 PHYTOLOGIA July 1997 volume 83(1):7-21

morphological accounts, and maps make it much simpler to comprehend the genus at
all levels.

A synopsis of the history, morphology, and distribution of the goldenaster genera
is presented at the beginning of the Phyllotheca monograph. The segregation of
Heterotheca, Chrysopsis, and Pityopsis makes sense morphologically, cytologically,
and phyletically, and those generic delimitations have gained increasing acceptance
over the 20 years since publication of evidence for this system (Semple 1977; Semple
et al. 1980). Among the goldenaster genera, Heterotheca is the largest and most
taxonomically difficult. The complexity of the variation patterns apparently has long
postponed a treatment of the largest part of the genus (sect. Phyllotheca), and in the
wake of Semple’s comprehensive study, it seems unlikely that anyone will be eager to
begin any detailed process of reevaluation of the whole group.

An earlier review of Semple’s treatise (Burk 1996, p. 219) speculated, however,
that “because of its inherent variability, sect. Phyllotheca will continue to present
difficulties for field biologists.” The treatment will be subject to “the inevitable
revisions of the 21st Century,” and “if [Semple’s taxonomic] structure is in time
dismantled, he has nonetheless brought together here the building blocks to shape
another.” As with any study that pulls together such a large amount of information,
unresolved problems also are brought to clearer focus, and the treatment provides an
invaluable basis for further studies of the biology and evolution of these species. The
present review provides an overview and perspective for some of the more interesting
conclusions and questions that arise from the Phyllotheca monograph.

The 20 species (as recognized by Semple) of sect. Phyllotheca are a mixture of
narrow endemics (e.g., Heterotheca rutteri, H. marginata, H. jonesii, H. brandegeei,
H. pumila, H. barbata, H. shevockii, H. monarchensis, H. mexicana) and entities
more widespread to varying degrees (e.g., H. villosa, H. canescens, H. stenophylla,
H. camporum, H. mucronata, H. zionensis, H. fulcrata, H. viscida, H. echioides, H.
sessiliflora). The most complex taxa are H. villosa (nine varieties, no subspecies) and
H. sessiliflora (four subspecies, seven basic entities). Heterotheca mucronata, H.
camporum, and H. stenophylla have two varieties each and H. fulcrata has four
varieties.

Semple has dealt with the complex variation and difficulties in identification in a
forthright way by separating specimen citations for collections that deviate from the
typical form of the taxon. These are given in paragraphs (often several) after citations
of “typical” collections with the heading of “aff. [the taxon under consideration]”
followed by a parenthetical explanatory expression (e.g., “approaching var. minor” or
“possible hybrids with H. zionensis”). A commentary on unusual variation for each
taxon also is provided, and the indications of “aff.” status are shown by distinct
symbols on the distribution maps.

Taxonomic delimitations in sect. Phyllotheca are based in part on multivariate
morphometric analyses “on more than 600 specimens including 76 type specimens,” to
be published separately (p. 2). Their publication will correspondingly contribute to an
understanding of variation in sect. Phyllotheca and its taxonomic treatment. And “a
cytogeographic study of the whole genus with a review of all previously published
counts and new reports for several hundred individuals is in preparation” (p. 23).

Nesom: Review of Heterotheca revision 9

Discussion of evolutionary processes underlying the variation patterns are found in the
commentaries by Semple on individual species.

Among the most interesting features of the variation patterns described by Semple
are the strongly overlapping geographic ranges in infraspecific taxa of most of the
widespread species (especially see H. sessiliflora, H. villosa, H. fulcrata, and H.
mucronata). Are these now sympatric entities recently spread from originally
allopatric, more restricted ranges, with extensive hybridization resulting in blurred
morphological boundaries in regions of overlap? Or, do these sympatric entities
maintain their evolutionary independence to a significant degree? Evidence suggests
that both situations may exist in sect. Phyllotheca.

“Within species, intervarietal hybrids are common in areas of sympatry” (p. 24),
but these are usually between plants at the same ploidy level. Interspecific
hybridization, however, is generally uncommon between diploids of sect. Phyllotheca
but more common among tetraploids, suggesting that the difference in ploidy (between
diploids and tetraploids) provides an effective isolating mechanism (see various
comments below). Triploids are rarely encountered.

Taxonomic concepts

Indication of Semple’s general approach toward fitting a nomenclatural system to
the variation patterns is provided in commentary regarding Heterotheca villosa. “The
races [= varieties, of Heterotheca villosa] fit well with the concept of variety in that
each occurs in pure form in some populations, and the overall ranges are sympatric to
a considerable degree with at least one other variety. Some taxa have sufficiently non-
overlapping ranges that subspecies status might be considered. ... Each variety most
likely evolved in isolation and adapted to a different set of habitat parameters, but by
and large no variety now occurs in isolation” (1996, p. 108). The biology and
taxonomy of H. villosa and others (where the only infraspecific category is “var.”’)
contrast in the Phyllotheca monograph with that of H. sessiliflora (where both
“subsp.” and “var.” are used).

For a more detailed explanation of his concepts of subspecific and varietal
categories, Semple refers to an earlier study of the genus Xanthisma: “A subspecies is
characterized by all members exhibiting a particular morphology distinct from other
individuals in the species and by the allopatric distribution of these members from the
rest of the species” [citing various references] (Semple 1974, p. 4). “The variation
between subspecies can be described as discontinuous, except for the few hybrids”
(1974, p. 8). “A variety is characterized by all members of a population exhibiting a
particular morphology distinct from other individuals in the species. The distribution
of these populations is sympatric with populations whose members are not within the
same variety, and also many populations of morphological intermediates exist [citing
various references]. ... Wan Steenis described varietal level variation as being
continuous with other varieties, although the continuum would have pronounced
modes” (Semple 1974, p. 8-9).

Another perspective on Semple’s varietal concept is found in his comments on
Heterotheca brandegeei, which is markedly variable in glandularity and density of
indument. The species is narrowly endemic to Sierra San Pedro Martir in Baja

10 PHYTOLOGIA July 1997 volume 83(1):7-21

California and is not suspected of intergrading with any other. “Even in a species with
a limited distribution and a relatively few number of populations the full range in
indument variation can be encountered. In other species with a greater range in
[geographic] distribution than H. brandegeei (e.g., H. sessiliflora, H. fulcrata, H.
mucronata, and H. villosa) past periods of isolation apparently have allowed fixation
of different alleles controlling indument features in different-portions of the range
resulting in morphologically more well-defined races (generally labeled in this
treatment as varieties)” (p. 66).

The taxonomic approach taken by Semple (recognition of numerous sympatric
varieties) is perhaps by necessity a first step simply in providing a documented
phenetic framework for the variation in this biologically complex group. Semple notes
that this approach serves a related practical purpose. In discussing the strong
similarity between Heterotheca villosa vars. ballardii and foliosa (both mostly
tetraploid), he observes that if the diagnostic distinction of the former provides
insufficient grounds for its formal recognition, “it then would be logical to merge all
other varieties together with no infraspecific taxa being recognized in H. villosa. This
would result in the loss from the formal nomenclature of a great deal of information on
variation and distribution in what is admittedly a difficult species complex. Splitting
seems justifiable in this case, and it maintains a nomenclature that parallels what has
been adopted with less hesitation for other species in the section” (p. 114).

Still, if entities can be identified with some degree of consistency (as implied by
the maps and specimen citations), and if they are sympatric and similar in habitat and
phenology, some degree of internal reproductive isolation might be inferred to exist.
Alternatively, segregation of linked genes controlling the character suites by which
these taxa are identified may have a large effect on the variation patterns. Needed for
interpretation, but missing in most cases, are observations on variation within
populations of the taxa concerned. For those species where isolation does exist among
the infraspecific taxa, the taxonomic approach could be shifted more toward an
evolutionary perspective. Alternative taxonomic interpretations are possible, based on
the same evidence and information.

Semple’s approach to variation patterns and taxonomic applications in various
species of Heterotheca is discussed below.

Heterotheca villosa/stenophylla var. angustifolia

Semple has transferred var. angustifolia of Heterotheca villosa to H. stenophylla.
The latter species then becomes “divided into two seemingly quite distinct [and
strongly sympatric] varieties that differ in gland and hair density” (p. 88). The transfer
of var. angustifolia was made on the basis of “field experience and the results of
multivariate analyses” showing that “the type of var. angustifolia is morphologically
closer to many individuals of var. stenophylla than it is to either H. canescens or
typical H. villosa” (p. 94).

Semple speculates that “tetraploid var. angustifolia originated from diploid var.
stenophylla and subsequently converged toward tetraploid H. canescens due to
putative occasional hybridization with the latter. . . . Alternatively, var. angustifolia
might have originated via allopolyploidy from more hairy and less glandular diploid H.

Nesom: Review of Heterotheca revision 11

stenophylla var. stenophylla and H. canescens’ (p. 94). Heterotheca stenophylla and
H. canescens are shown as sister species in Semple’s phylogram.

Based on Semple’s estimate of its evolutionary origin, var. angustifolia could
justifiably be treated within or close to either of the two contributors to its genome: (a)
Heterotheca stenophylla and (b) H. canescens.

(a) “The range of var. angustifolia is generally the same as that of var. stenophylla
from Oklahoma northward, except that var. angustifolia occurs over a slightly greater
area and in the gaps between the disjunct populations of var. stenophylla” (p. 94).
“The two varieties occur in pure and mixed populations throughout the range of the
species” (p. 53). Cytological evidence seems unequivocal in suggesting that var.
angustifolia is genetically isolated from var. stenophylla. Most chromosome number
reports for var. stenophylla have been of diploids, while all of many reports for var.
angustifolia have been of tetraploids. “Several triploid counts [have been reported]
from putative intervarietal hybrids” (p. 92).

(b) Heterotheca canescens also is mostly diploid over its range and also is broadly
sympatric with var. angustifolia. The only intermediate collections cited by Semple for
H. canescens are those “aff. H. canescens (close to H. stenophylla var. angustifolia)”
(p. 100). Presumably, var. angustifolia - H. canescens hybrids are triploid.

Because var. angustifolia is broadly sympatric with both of its putative parents and
apparently genetically isolated from them to a significant degree, its treatment at
specific rank also is a possibility. It presumably is an evolutionarily distinct entity and
its morphogeographic circumscription is the same regardless of its taxonomic
placement.

If Semple’s hypothesis of origin for var. angustifolia is correct, placement of it
within Heterotheca stenophylla is better than within H. villosa. Inclusion of var.
angustifolia, however, only slightly increases the morphological complexity of H.
villosa, as defined by Semple, and occupies a part of the overall geographic range
where its sympatry with conspecific varieties is relatively less (Figs. 39 and 40).

Heterotheca villosa

Heterotheca villosa is “highly variable in diagnostic features” and is “difficult to
define as a species, although each infraspecific taxon has a diagnostic suite of traits.”
The species is “very variable in stem height, leaf base shape, stem and leaf indument
traits, numbers of heads per capitulescence and florets per head” (p. 105). It is
“defined by what it lacks rather than what it possesses” (p. 108).

Nine varieties are recognized within Heterotheca villosa in 1996, but Semple’s
concepts of these taxa have fluctuated. In 1990, he placed a number of names as
synonyms of H. villosa var. hispida (= H. villosa var. minor of 1996) with the
following comment: “Included are morphotypes that I have previously accepted as
species or subspecies (Semple 1987), but have come to view as sometimes semi-
distinct regional ‘races’ that grade into each other to such an extent that continued
recognition cannot be justified with the data available to me at present.” Later (1994,
1996), apparently based on multivariate studies, he returned to his earlier position of

12 PHYTOLOGIA July 1997 volume 83(1):7-21

formally recognizing these races, including four varieties within H. villosa from the
same 1990 list of synonyms. “A number of morphotypes [of H. villosa] appear
sufficiently distinct to warrant recognition. .. . All races have well defined
geographic distributions which overlap to a considerable degree in some cases (Figs.
39-40). The highly plastic nature of the species and undoubted hybridization make
identification to variety difficult in numerous cases” (1996, p.- 108).

“The diploid races [of Heterotheca villosa] are usually distinct from each other, but
each has given rise to one (or more) tetraploid lines [exception noted below].
Tetraploids [‘more common than diploids in H. villosa’] tend to look more alike
because the diploid traits are less pronounced and because the tetraploids are more
likely to have hybridized, thus further blurring the distinctions between the races.
Possible occasional hybridization with tetraploids in other species may also have
further buffered the distinctive morphology of the tetraploid level of the pillar
complex” (p. 108).

Some infraspecific taxa of Heterotheca villosa are more distinct than others. Two
have been regarded as species in recent floristic treatments: (a) var. nana (as H.
horrida, e.g., Correll & Johnston 1970; Dorn 1988) and (b) var. depressa (as H.
depressa, Dorn 1988). The distinctiveness of these entities is further emphasized by
the relatively few collections cited for them as “aff.” Treatment of var. nana and var.
depressa at species rank appears to be a reasonable alternative potentially providing a
closer match between taxonomy and the evolutionary pattern.

(a) Var. nana (diploid, many counts, without tetraploid populations) is almost
completely overlaid in its geographic range by var. foliosa (diploid and tetraploid but
tetraploid in its area of overlap with var. nana, many reports) and by var. minor
(diploid and tetraploid, numerous reports). Intermediates between var. nana and var.
scabra occur in the Four Corners area; the closest relative of var. nana is the narrow
endemic var. sierrablancensis (diploid), which occurs at the southeastern corner of the
range of var. nana.

(b) Var. depressa (tetraploid, several counts, with only a speculative evolutionary
connection to diploids) is endemic to habitats connected with hot springs and geyser
basins mostly in the area of Yellowstone National Park. Putative hybrids have been
observed between var. depressa and var. minor, which is sympatric but different in
habitat.

Heterotheca villosa var. pedunculata also has distinctions that set it apart from
other taxa within H. villosa. “Semple (1990) included it among tentative synonyms of
var. villosa, but the results of multivariate analyses conducted since then indicate that it
is sufficiently distinct from var. villosa to warrant recognition even when only non-
diagnostic traits are used in the discriminant analysis. It is the only usually tetraploid
taxon in sect. Phyllotheca that has very densely pubescent leaves” (1996, p. 124).
Intergrades occur between var. pedunculata and var. minor (tetraploid) and var. scabra
(tetraploid).

Var. pedunculata “is similar to the recently described Heterotheca mexicana, which
has achenes with a weakly developed short outer pappus whorl. If the more pubescent
forms of the Mexicana complex [H. mucronata, H. gypsophila, H. mexicana] are
primitive in the section, then var. pedunculata is likely to be similar to the ancestral

Nesom: Review of Heterotheca revision 13

form of H. villosa from which other taxa evolved . . .” (1996, p. 124). These
comments seem to imply that var. pedunculata is closely related to the Mexicana
species, but there apparently is no further development of the hint that the Mexicana
complex may be primitive within sect. Phyllotheca. Nor does the 1996 phylogram
support this point of view. The phylogram also places H. villosa in a position widely
separated from the Mexicana complex. .

Heterotheca mucronata

Semple has described Heterotheca mucronata var. harmsiana (var. nov.) from the
northeast Mexican states of Tamaulipas, Nuevo Leén, San Luis Potosi, and Coahuila.
Var. harmsiana differs from the typical variety in its leaves with “fewer hairs and more
glands,” illustrating Semple’s observation (p. 94) that “most other species [of sect.
Phyllotheca] include both more glandular and more hairy races.” War. mucronata and
var. harmsiana have essentially congruent geographic distributions and both have been
collected from at least six of the same localities or localized areas from a relatively
small region within Nuevo Leén and Coahuila (see specimen citations for the two
taxa): the Pefia Nevada area; east of Iturbide; Chipinque; Sierra de la Viga; Sierra de
Arteaga; and Cafion de San Lorenzo.

My own field and herbarium experience have indicated that only a single
evolutionary entity exists among plants identified as Heterotheca mucronata. Plants
from Tamaulipas and near Linares in southeastern Nuevo Le6én have eglandular leaves
and a more densely sericeous vestiture of thinner-based trichomes than those in the
remainder of the Mexican range of the species (pers. observ.), but the distribution of
these variants does not match the distribution of var. mucronata described by Semple.
Putative intergrades with H. fulcrata (see below) have been collected around Saltillo,
Coahuila, and slightly to the south in northern Zacatecas.

The recognition within Heterotheca mucronata of closely sympatric varieties with
no apparent difference in habitat or phenology suggests that the taxa recognized are
inter- or infra-populational variants differing in the expression of two types of
trichomes. Local adaptation and genetic segregation could account for differentiation
among and within populations. An independent evaluation would be useful to resolve
the differences in perception of these variation patterns, but differences in our concepts
of the varietal category apparently preclude any chance of taxonomic agreement.

Heterotheca fulcrata

The distinctive species Heterotheca fulcrata comprises four varieties in Semple’s
concept. Numerous reports of diploid chromosome numbers have been reported for
all of them (plus one “unconfirmed” tetraploid count for var. fulcrata). Vars. fulcrata,
arizonica, and senilis are sympatric with nearly congruent ranges in the montane
habitats of the Chihuahuan Desert region in northeastern México and trans-Pecos
Texas and from there into southern New Mexico and Arizona. I have identified these
plants in México as a single evolutionary entity (= H. fulcrata). The overall
geographic distributions of var. fulcrata and var. amplifolia (sensu Semple) also are
remarkably similar, as are those of var. arizonica and var. senilis. In fact, given the
apparent cohesiveness of the species, evidence suggests that the varieties (sensu

14 PHYTOLOGIA July 1997 volume 83(1):7-21

Semple) of H. fulcrata are better regarded as local variants in the sense of most current
botanists, perhaps treated with taxonomic status as “forma,” which would retain the
formal nomenclature desired for these entities by Semple (see “Taxonomic Concepts,”
above).

Heterotheca sessiliflora complex

Within the primarily Californian Heterotheca sessiliflora complex, Semple has
fashioned an amalgum of greatly increased complexity by combining H. echioides, H.
camphorata, H. bolanderi, H. fastigiata, and H. sessiliflora into a single species (H.
sessiliflora). Four of these are treated at subspecific rank (subspp. echioides,
bolanderi, fastigiata, and sessiliflora), H. camphorata is treated as a variety and placed
within subsp. echioides. Varieties are recognized within subsp. fastigiata (2 vars.)
and subsp. echioides (3 vars.); subspp. sessiliflora and bolanderi are monotypic.
Heterotheca monarchensis is a narrow endemic from the Kings River canyon in
Fresno County. It is similar to H. echioides but is morphologically distinct and
geographically separated from other members of the H. sessiliflora complex.

Semple’s Figure 14 maps the geographic distribution of the basic taxa of
Heterotheca sessiliflora as he has defined that species. Four varieties of H. sessiliflora
are sympatrically overlaid in Los Angeles County, three each in San Bernardino and
Ventura counties. Var. camphorata is closely sympatric with var. echioides in
Monterey, Santa Clara, and Santa Cruz counties.

Without disagreement regarding delimitation of the basic evolutionary units of the
Heterotheca sessiliflora complex, they can be positioned in a way that more closely
matches the evolutionary situation by essentially eliminating sympatric entities within a
single species. A taxonomic arrangement to accomplish this is suggested below (Fig.
1), contrasted with Semple’s arrangement of the same basic entities (Fig. 2).
Continuing elimination of natural habitats and creation of hybrid habitats by human
activities might drive this whole complex toward a genetic swarm, but the suggested
alternate arrangement preserves the morphological coherence of the taxa involved and
provides a more comprehensible tool for dealing with the current morpho-geographic
pattern of variation. Based on the information presented by Semple, and in my
experience, the H. sessiliflora complex (sensu Semple) is significantly different from
most other Heterotheca species of this treatment that are divided into sympatric
varieties.

Semple’s basic units in the Heterotheca sessiliflora complex are a mix of
subspecies and varieties. He did not treat the entities subsp. sessiliflora and subsp.
bolanderi at varietal rank, apparently because both are restricted to coastal strand
habitats and neither is geographically overlapping with any other taxa (see definitions
above of variety and subspecies). Formal varietal combinations were not provided for
“var.” fastigiata and “var.” echioides, although it appears that this was intended, as
they are repeatedly referred to as “var. fastigiata” and “var. echioides” and shown on
the phylogram as entities coordinate with other varieties. The count of “24 varieties”
in the Abstract also must include “var. fastigiata” and “var. echioides.”

Nesom: Review of Heterotheca revision 15

Heterotheca sessiliflora (Nutt.) Shinners
var. sessiliflora
var. fastigiata (Greene) Semple, ined. [nom. nud. in Semple 1996]
var. sanjacintensis Semple
var. thiniicola (Rzed. & Ezc.) Nesom

Heterotheca echioides (Benth.) Shinners
var. echioides
var. bolanderioides (Semple) Nesom
var. bolanderi (A. Gray) Nesom

Heterotheca camphorata (Eastw.) Semple

Heterotheca monarchensis York, Shevock, & Semple

Figure 1. Alternate taxonomy for the Heterotheca sessiliflora complex. Except for
var. fastigiata, nomenclatural combinations to formally complete this are provided
below.

Heterotheca sessiliflora (Nutt.) Shinners

subsp. sessiliflora

subsp. fastigiata (Greene) Semple
var. fastigiata (Greene) Semple, ined. [nom. nud. in Semple 1996]
var. sanjacintensis Semple

subsp. echioides (Benth.) Semple
var. echioides (Benth.) Semple, ined. [nom. nud. in Semple 1996]
var. bolanderioides Semple
var. camphorata (Eastw.) Semple

subsp. bolanderi (A. Gray) Semple

Heterotheca thiniicola (Rzed. & Ezc.) B.L. Turner

Heterotheca monarchensis York, Shevock, & Semple

Figure 2. Semple’s taxonomy for the Heterotheca sessiliflora complex. See
comments in text regarding “ined.” nomenclature.

16 PHY TOLOGIA July 1997 volume 83(1):7-21

The alternate arrangement adopts Semple’s suggestions in combining Heterotheca
fastigiata with H. sessiliflora and H. bolanderi with H. echioides, adding a newly
described variety to each species. Heterotheca bolanderi (diploid) is discrete in
geography and habitat and might be kept as a distinct species, but it is closely similar
to var. echioides and var. bolanderioides and may have been directly involved in the
parentage of the latter, which is primarily tetraploid (fide Semple, p. 49). Heterotheca
camphorata (mostly diploid) is kept as a separate species (with a combination made
earlier by Semple) and H. thiniicola is brought within H. sessiliflora (comments
below). This arrangement does not eliminate difficulties in identifying hybrids,
introgressants, and other intermediates for whatever reason, but such problems exist
no matter what taxonomic superstructure is laid over the basic evolutionary units. The
most common interspecific hybrids in the suggested alternate arrangement appear to be
between H. echioides (var. echioides) and H. sessiliflora (var. fastigiata) where they
are sympatric in San Bernardino, Los Angeles, and Ventura counties.

Heterotheca sessiliflora (s. str.), like H. bolanderi, is a coastal strand entity
discrete in geography and habitat, but Semple’s proposal to unite it with H. fastigiata
is a good one. The four varieties of H. sessiliflora (as suggested here) are exclusively
diploid and distributed allopatrically in southwestern California and northwestern
México (Baja California and Sonora). A sericeous indument of short hairs and leaves
with distinctly wavy margins unite this group of plants and give it an immediately
recognizable appearance.

In a treatment of Mexican Heterotheca (Nesom unpublished), H. thiniicola (a
desert habitat population from northwestern Sonora) has been included in the same
circumscription as the type of H. fastigiata. Semple, in contrast, has maintained H.
thiniicola at specific rank, noting (p. 54) that “While similar to var. fastigiata, H.
thiniicola is sufficiently different to warrant recognition as a separate taxon. Its unique
habitat indicates that it is more than just a disjunct population of the montane var.
fastigiata.”

The only differences I can confirm to separate Heterotheca thiniicola from H.
sessiliflora var. fastigiata are those noted by Semple: the absence of osteiform (Type
A) trichomes on the disc corollas of the former, its distinctly desertic habitat at 110
meters elevation, and a geographic disjunction of about 200 miles from other H.
sessiliflora. Var. fastigiata, however, occurs in habitats at “(150) -300-1800- (2200)”
meters elevation, low enough to include “desert washes,” although its primary habitat
is higher in “pine forests and transition chaparral” (p. 40). Semple has made the
useful observation that the cuisistent occurrence of osteiform trichomes on the disc
corollas is evidence for monophyly of the H. sessiliflora complex (sensu Semple,
including H. monarchensis) --- the absence of these trichomes on H. thiniicola corollas
almost certainly has resulted from a recent evolutionary loss (vs. primitive absence)
and does not suggest the species should be considered apart from var. fastigiata, to
which it is otherwise nearly identical. To formally recognize the evolutionary
independence (via geographic isolation) of the Sonoran population and its small degree
of morphological divergence, it is treated here at varietal rank within H. sessiliflora,
coordinate with the other three varieties.

Nesom: Review of Heterotheca revision 17

Heterotheca sessiliflora (Nutt.) Shinners var. thiniicola (Rzed. & Ezc.) Nesom,
comb. nov. BASIONYM: Haplopappus thiniicola Rzed. & Ezc., Cienc.
Interamer. 26:16. 1986. Heterotheca thiniicola (Rzed. & Ezc.) B.L. Turner,
Phytologia 63:128. 1987.

Heterotheca echioides (Benth.) Shinners var. bolanderioides (Semple) Nesom,
comb. nov. BASIONYM: Heterotheca sessiliflora (Nutt.) Shinners var.
bolanderioides Semple, Phytologia 73:450. 1992.

Heterotheca echioides (Benth.) Shinners var. bolanderi (A. Gray) Nesom,
comb. nov. BASIONYM: Chrysopsis bolanderi A. Gray, Proc. Amer. Acad.
Arts 6:543. 1866. Heterotheca bolanderi (A. Gray) Harms, Brittonia 26:61.
1974.

Species concepts in sect. Heterotheca

Semple recognizes seven species of sect. Heterotheca (see p. 25: “Key to
Heterotheca sect. Heterotheca [after Wagenknecht, 1960, with modifications]’),
noting that my approach (Nesom 1990) contrasted with that of Wagenknecht. He
adopted Wagenknecht’s definitions of taxa without commenting on the suggestion that
H. subaxillaris be broadened to include H. latifolia (including varieties), H.
psammophila, and H. chrysopsidis. Regional morphological tendencies in H.
subaxillaris can be recognized, but my brief study was unsuccessful in sorting out
morpho-geographic “nodes” in this phenotypically malleable complex that could be
unarbitrarily recognized. Nor has anyone provided a documented (specimen-based)
map showing the distribution of these taxa. Commenting on previous studies of sect.
Heterotheca, including Wagenknecht’s, Harms (1968, p. 9) observed that “Perhaps
this entire [H. subaxillaris] complex should still be accepted as a single, polymorphic,
polytypic species.” Lammers (1997), in contrast, apparently has identified H. latifolia
(as distinct from H. subaxillaris) with confidence and is able to distinguish all three
varieties of H. latifolia.

Semple’s key to sect. Heterotheca gives an overview of the typological concepts in
the H. subaxillaris complex that may be applied to indicate that one or another plant
approaches the typical morphology of a named taxon, but either extensive interregional
gene flow or weak primary differentiation, or both (see comments by Burk 1961,
1966), have not made it simple to find geographic patterns to which a meaningful
(predictive) taxonomy can be applied. Field and lab study may yet show that such
patterns and evolutionary entities exist, but as indicated earlier (Nesom 1990), it will
be a considerable challenge to provide this evidence.

Taxonomy of Heterotheca sect. Ammodia
A taxonomic study of the single species of sect. Ammodia (Heterotheca oregona)

was published earlier (Semple et al. 1988). The treatment of infraspecific variation
there is similar to that in sect. Phyllotheca. Four partially sympatric entities with “no

18 PHYTOLOGIA July 1997 volume 83(1):7-21

indication of any pronounced differences in habitat preference” (p. 554) were found to
separate with little or no overlap in a multivariate analysis. Following earlier criteria
(e.g., Semple 1974), “varietal rank was determined to be most appropriate. The
ranges of the four races overlap to a great extent in California, which precludes
subspecies status, although each has a unique range” (1988, pp. 549-550).
Chromosome numbers have been reported for three of the varieties: all are diploid.

“The varietal differences [within Heterotheca oregona] are thought to have evolved
as a consequence of geographic isolation. During the Holocene, migration and range
expansions have eliminated the spatial isolation and the sympatric races now
hybridize” (1988, p. 553). Non-overlap in multivariate analyses, however, and a low
frequency of intermediacy (“about 10% of all herbarium specimens . . . studied”) seem
to indicate that the infraspecific taxa may be separated by substantial internal
reproductive isolation. As presented by Semple et al. (1988), information suggests
that these closely sympatric but little intergrading entities with small morphological
differences may be biclogical microspecies.

Status of Bradburia

Semple notes that “circumscription of all the generic limits of the goldenasters
remains in turmoil” (p. 7). His only example, however, of problematic generic limits
is the question of taxonomic rank for Bradburia (as a separate genus vs. a subgroup
within Chrysopsis). Turmoil is not evident, and given increasing agreement with
Semple’s arrangement of Heterotheca, Chrysopsis, and Pityopsis, the only
controversy appears to involve the Bradburia question and what it may imply (for
consistency) about the relationship of sect. Heterotheca to the rest of the genus (sensu
Semple).

In contrast to my decision to merge the genus Bradburia with Chrysopsis as sect.
Bradburia (enlarged to two species with the addition of Chrysopsis pilosa, Nesom
1991a), Semple has decided to retain Bradburia as a separate genus including the same
two species. He has observed the close similarity and relationship between B. hirtella
and C. pilosa (Semple & Chinnappa 1984) and accepts the results of recent
morphological analyses (Nesom 1991a) and molecular analyses (Lane et al. 1996) that
place them as sister species. These two, in turn, are the sister group to the rest of
Chrysopsis in phyletic analyses including other taxa of goldenasters (Nesom 1991b;
Lane et al. 1996) as well as in Semple’s own diagram of goldenaster relationships
(1996, p. 6).

Semple’s published justification for maintaining Bradburia at generic rank is solely
his view that a ditypic Bradburia could serve as an “alternative solution to the generic
limits problem surrounding the goldenasters” (p. 7). He has neither indicated on what
grounds he prefers one alternative rather than the other nor provided any discussion of
the relative merits or problems regarding the choice of options. Based on his
comments and distribution maps, the distinctions between the two genera are
summarized as follows.

1. Perennial; leaves and stems with “distinctly flagelliform hairs”; cells of disc corolla
throat with elongate crystals; Florida to Mississippi, Louisiana, and Texas, but
mostly east of the Mississippi RIVET. ..5.”.... ...0.<<--276--0- 20452 see Chrysopsis

Nesom: Review of Heterotheca revision 19

1. Annual or perennial; leaves and stems with “less to non-flagelliform hairs”; cells of
disc corolla throat without crystals or crystals reduced in size; Texas and Louisiana
to Missouri and Kansas, Tennessee, Mississippi, and Alabama, but mostly west of
the MMississippi River: oc. sis. ceseseeess-oosee-sermmecane. osaeteasrines sree. ae Bradburia

Additionally (Nesom 1991a), these two species differ as a pair from other Chrysopsis
in longer flowering branches, scarious-margined phyllaries, sharp-pointed sweeping
hairs on the style branches, and karyotype.

If the characterization of Bradburia hirtella and Chrysopsis pilosa as sister species
is correct, and if these two are phylogenetically coordinate with the rest of Chrysopsis,
taxonomic treatment of a ditypic Bradburia at either rank (within or distinct from
Chrysopsis) is consistent with the phylogeny. My study also noted that the
enlargement of an independent Bradburia was an alternative solution (Nesom 1991a,
p. 111): “Chrysopsis pilosa and Bradburia are so distinct as a pair that C. pilosa might
justifiably be transferred to Bradburia.” Does available evidence support a decision
regarding the taxonomic placement of ditypic Bradburia? And which treatment is more
consistent with existing taxonomic arrangements within the Chrysopsidinae?

Within the goldenaster group (subtribe Chrysopsidinae), ditypic Bradburia is
united with Chrysopsis (sensu Semple) by a set of cytological and morphological
features: reduced base chromosome number (x=5 or 4; shared with the genus
Osbertia); long, smooth-walled osteiform trichomes often conspicuously drawn out
into flexuous, filamentous extensions; achene shape obovate and asymmetric (shared
with sect. Heterotheca); achene surfaces with thick, rounded ridges, the nerves
completely below the epidermal surface; and pappus insertion inset from the shoulder
rim of the achene apex. A significant degree of genetic similarity between the two
segments of Chrysopsis was demonstrated by hybrids betweeri C. pilosa (sect.
Bradburia) and C. gossypina (sect. Chrysopsis) synthesized by Semple (1981), who
then viewed C. pilosa as the sister species to C. gossypina and justifiably treated
within Chrysopsis. As noted above, molecular data also indicates that
Bradburia/Chrysopsis is monophyletic. ;

The relationship of ditypic sect. Bradburia to the rest of Chrysopsis appears to be
analogous to the relationship of sect. Heterotheca, and perhaps of sect. Ammodia, to
the rest of the genus Heterotheca (sect. Phyllotheca). Semple (1996) considers sect.
Heterotheca to be the sister group to rest of the genus, sect. Ammodia phyletically
coordinate with sect. Heterotheca. My cladistic analysis (Nesom 1991b) placed sect.
Ammodia basal within the genus and sect. Heterotheca among other clades, but only
weak characters supported this. Sect. Heterotheca is a distinct and clearly
monophyletic group, but Harms (1965) synthesized viable hybrids between H.
subaxillaris (sect. Heterotheca) and H. canescens (sect. Phyllotheca). The option
(“alternative solution”) of segregating sect. Heterotheca as a small genus within the
goldenasters has often been followed, with the remainder of the Heterotheca species
placed into an expanded Chrysopsis. If sect. Heterotheca were segregated today,
however, the generally accepted redefinition of Chrysopsis would necessitate
recognition of a new genus to accommodate the species of sects. Phyllotheca and
Ammodia. Nevertheless, Semple’s rationale for segregrating Bradburia as a genus

20 PHY TLOLOGIA July 1997 volume 83(1):7-21

provides a similar one for the treatment of Heterotheca s. str. The monotypic sect.
Ammodia also has been treated as a separate genus (Nuttall 1841) and could be again.

Available evidence and the current taxonomy of the Chrysopsidinae indicate to me
that ditypic Bradburia (in the current view of its phylogeny) is better viewed as a well-
defined subgroup of Chrysopsis rather than a weakly separated genus.

ACKNOWLEDGMENTS

I am grateful to John Strother, John Kartesz, and Billie Turner for their comments
on the manuscript.

LITERATURE CITED

Burk, C.J. 1961. Environmental variation in Heterotheca subaxillaris [from Texas].
Rhodora 63:243-246.

. 1966. Rainfall periodicity as a major factor in the formation of flowering
races of camphorweed (Heterotheca subaxillaris). Amer. J. Bot. 53:933-936.

. 1996. Book Review [Semple 1996, Revision of Heterotheca sect.
Phyllotheca]. Rhodora 98:218-219.

Correll, D.S. & M.C. Johnston. 1970. Manual of the Vascular Plants of Texas.
Texas Research Foundation, Renner, Texas.

Dorn, R.D. 1988. Vascular Plants of Wyoming. Mountain West Publishing,
Cheyenne, Wyoming.

Harms, V.L. 1965. Cytogenetic evidence supporting the merger of Heterotheca and
Chrysopsis (Compositae). Brittonia 17:11-16.

. 1968. Nomenclatural changes and taxonomic notes on Heterotheca,
including Chrysopsis, in Texas and adjacent states. Wrightia 4:8-20.

Lammers, T.G. 1997. Heterotheca latifolia (Asteraceae), new to the flora of Iowa.
Sida 17:841-842.

Nesom, G.L. 1990. Taxonomy of Heterotheca sect. Heterotheca (Asteraceae:
Astereae) in México, with comments on the taxa of the United States. Phytologia
69:282-294.

___. 1991a. Union of Bradburia with Chrysopsis (Asteraceae: Astereae),
with a phylogenetic hypothesis for Chrysopsis. Phytologia 71:109-121.

. 1991b. A phylogenetic hypothesis for the goldenasters (Asteraceae:
Astereae). Phytologia 71:136-151.

. 1994. Subtribal classification of the Astereae (Asteraceae). Phytologia
76:193-274.

Nuttall, T. 1841. Descriptions of new species and genera of plants in the order
Compositae. Trans. Amer. Phil. Soc. ser. 2, 7:283-454.

Semple, J.C. 1974. The phytogeography and systematics of Xanthisma texanum
DC.: proper usage of infraspecific categories. Rhodora 76:1-19.

. 1977. Cytotaxonomy of Chrysopsis and Heterotheca (Compositae -
Astereae): A new interpretation of phylogeny. Canad. J. Bot. 55:2503-2513.

Nesom: Review of Heterotheca revision 2

1981. A revision of the goldenaster genus Chrysopsis (Nutt.) Ell.

nom. cons. (Compositae: Astereae). Rhodora 83:323-384.

. 1987. New names, combinations and lectotypifications in Heterotheca
(Compositae: Astereae). Brittonia 39:379-386.

. 1992. The goldenasters of California, Heterotheca (Compositae:
Astereae): New names and combinations. Phytologia 73:449-455.

. 1994. New combinations in the Heterotheca villosa (Pursh) Shinners
complex (Compositae: Astereae). Novon 4:53-54.

. 1996. A revision of Heterotheca sect. Phyllotheca (Nutt.) Harms
(Compositae: Astereae). Univ. Waterloo Biol. Ser. 37:i-iv, 1-164.

Semple, J.C., V.C. Blok, & P. Heiman. 1980. Morphological, anatomical, habit,
and habitat differences among the goldenaster genera Chrysopsis, Heterotheca,
and Pityopsis (Compositae: Astereae). Canad. J. Bot. 58:147-163.

Semple, J.C. & C.C. Chinnappa. 1984. Observations on the cytology, morphology,
and ecology of Bradburia hirtella (Compositae: Astereae). Syst. Bot. 9:95-101.

Semple, J.C., C. Leeder, C. Leuty, & L. Gray. 1988. Heterotheca sect. Ammodia
(Compositae: Astereae): a multivariate study of H. oregona and specimens of
Brewer’s (golden)aster. Syst. Bot. 13:547-558.

Wagenknecht, B.L. 1960. Revision of Heterotheca sect. Heterotheca (Compositae).
Rhodora 62:61-76, 97-107.

Phytologia (July 1997) 83(1):22-33.

RAPD ANALYSIS OF GENETIC DIVERSITY AMONG AND WITHIN
POPULATIONS OF BALDUINA ATROPURPUREA AT FORT STEWART,
GEORGIA

Tracy Halward!, Alison Hill2, and Robert Shaw!

1Center for Ecological Management of Military Lands (CEMML), Colorado State
University, Fort Collins, Colorado 80523 U.S.A.

and

2U.S. Army Construction Engineering Research Laboratories (USACERL),
Champaign, Illinois 61826 U.S.A.

ABSTRACT

Compared with other federal land management agencies, the Department of
Defense (DoD) has a disproportionately large number of threatened,
endangered, and sensitive (TES) plant species known to occur on its lands
(Flather et al. 1994). In some instances, this has resulted in a conflict between
measures necessary to meet conservation requirements for TES species and the
ability of the installation to train troops and test weapons and equipment to
assure military readiness. To support the mission of the U.S. Army and Fort
Stewart, researchers at the U.S. Army Construction Engineering Research
Laboratories (USACERL) undertook a multi-scoped project to investigate
various aspects of Balduina atropurpurea, a federal ‘species of concern’
(formerly, category 3C under the Endangered Species Act) that is state listed as
‘rare’ in Georgia (Smith 1994). This particular portion of the project was
undertaken to determine relative levels of genetic diversity among and within
on-post populations of B. atropurpurea. Seedlings from five representative
on-post populations were evaluated using Random Amplified Polymorphic
DNA (RAPD) marker analysis. Very little genetic variation was detected
among or within the on-post populations evaluated. The variation observed
was randomly and approximately equally distributed among populations and
among individuals within populations.

KEY WORDS: Balduina atropurpurea, RAPD analysis, genetic diversity

22

Halward et al.: Genetic diversity in Balduina 23

INTRODUCTION

Compared with other federal land management agencies, the Department of
Defense (DoD) has a disproportionately large number of threatened, endangered, and
sensitive (TES) plant species known to occur on its lands (Flather et al. 1994). In
some instances, this has resulted in conflict between measures necessary to meet
conservation requirements for TES species and the ability of the installation to train
troops and test weapons and equipment to assure military readiness. There is,
therefore, great interest in pursuing innovative ways to manage and monitor TES
species on DoD lands. To support the mission of the U.S. Army and Fort Stewart,
researchers at the U.S. Army Construction Engineering Research Laboratories
(USACERL) undertook a multi-scoped project to investigate various aspects of
Balduina atropurpurea Harper, a federal ‘species of concern’ (formerly, category 3C
under the Endangered Species Act) that is state listed as ‘rare’ in Georgia (Smith 1994)
and known to occur on Fort Stewart.

Balduina atropurpurea (Asteraceae) is a perennial herb that occurs in wet areas of
peaty pitcher plant bogs, pine flatwoods, and pine savannas with seasonal standing
water (Patrick 1994). The species is endemic to the southeastern Coastal Plain area of
the United States (Lutz 1995). Extant populations are known to occur in scattered
locations in south to south-central Georgia, northeastern Florida, southeastern
Alabama, and southern Mississippi (Helton 1995; Mississippi Natural Heritage
Program 1991).

The largest, healthiest known populations of Balduina atropurpurea are thought to
occur on the U.S. Army’s Fort Stewart, Georgia, where 21 populations, distributed
across five training areas, have been identified. These populations range in size from
< 10 to > 2,000 individuals and cover areas of approximately 1 m2 to 19,500 m2
(Helton 1995; Fort Stewart Natural Resources Office, FSNRO 1996, pers. comm).
Prior to 1995, only six populations of B. atropurpurea were known to occur at Fort
Stewart. These populations were identified by The Nature Conservancy (TNC) during
a survey of the installation conducted between March 1992 and October 1994 (U.S.
Department of Defense 1994). An additional fourteen populations were discovered in
1995 during a survey of the installation by C. Helton (1995) and one population was
identified in 1996 by Fort Stewart Personnel (FSNRO 1996, pers. comm.). These
populations potentially double the number of known individuals of B. atropurpurea in
the state of Georgia. The Fort Stewart populations are, therefore, of particular
significance to the recovery and future listing status of the species.

Throughout its range, potential threats to the survival of Balduina atropurpurea
include: alterations to the hydrological regime; loss of habitat to agricultural,
commercial, and residential development; and inappropriate site management,
particularly fire suppression, resulting in increased shading by shrubs and trees (Smith
1994). Military training exercises that alter the hydrological regime, cause excessive
soil disturbance, or suppress the occurrence of fire could negatively impact the Fort
Stewart populations. At least 43% of the Fort Stewart populations show significant
impacts from tank maneuvers and/or off-road vehicle traffic. Most (>71%) of the
populations are in need of prescribed burning to reduce the encroachment of shrubs
and woody vegetation and to encourage the establishment of a healthy herbaceous
layer. Currently, no U.S. Fish and Wildlife Service recovery plan has been prepared

24 PHYTOLOGIA July 1997 volume 83(1):22-33

for B. atropurpurea, and there are no existing management plans in place at Fort
Stewart specifically designed for this species.

Development of a recovery plan for Balduina atropurpurea will be especially
challenging as very little is known about the reproductive biology of this species. In
the field, individuals typically produce a rosette the first year, with inflorescences
produced in the second and subsequent years. Under greenhouse conditions, we
observed individuals flowering during their first year of growth. Parker & Jones
(1975) reported that B. atropurpurea is self-incompatible, and that interspecific
hybridization does not occur among species of Balduina. They also reported the
occurrence of vegetative reproduction from root stocks. R. Determann, Atlanta
Botanical Garden, successfully propagated seeds of B. atropurpurea following four
weeks of cold stratification; the majority of the seeds germinated and produced robust
rosettes (R. Determann 1996, pers. comm.). Investigations into the phenology,
reproduction, seed dispersal, and seedling establishment of B. atropurpurea are
needed. Studies evaluating the effects of disturbance, as well as fire frequency and
intensity, on the reproduction and health of this species are also necessary.

Knowledge regarding relative levels of genetic diversity among and within
populations of Balduina atropurpurea at Fort Stewart would aid in determining
whether any on-post populations contain unique genetic characteristics. Such
populations should be given priority for conservation as their destruction would lead to
the potential loss of genetic diversity necessary for adaptation to environmental
changes or habitat disturbances. In addition to genetic diversity, other factors such as
population health and community structure should be considered when determining the
overall biological value of each on-post population.

The objectives of this study were (1) to evaluate the relative levels of genetic
diversity among and within a representative sample of on-post populations of Balduina
atropurpurea; and (2) to examine the relationship among genetic diversity,
morphological diversity, and habitat diversity for this species. The information
obtained from this study will aid in the development of a management plan for B.
atropurpurea at Fort Stewart.

METHODOLOGY

Seedlings from five Fort Stewart populations of Balduina atropurpurea were
obtained from R. Determann, Atlanta Botanical Garden. Seeds collected from the
remaining populations were either immature or non-viable and failed to germinate.
The five populations evaluated represent a diversity of habitats among the Fort Stewart
sites in which the species is found. The seedlings were transported to Colorado State
University, transplanted into pots containing a commercial, soilless potting medium
(Metro Mix) and placed in a greenhouse. Six individuals each from populations ‘1’,
Sars and ‘4’, and four individuals from population ‘5’ were included in the genetic
analysis.

DNA was extracted from fresh leaf tissues according to procedures adapted from

Halward et al.: Genetic diversity in Balduina 25

Stewart & Via (1993) (Appendix 1). Random Amplified Polymorphic DNA (RAPD)
analysis was conducted on the DNA extracts according to procedures adapted from T.

Lowrey (unpubl.) (Appendix 2). When the amplification process was complete, 10 jl
of electrophoresis tracking dye was added to each reaction tube and the reactions
loaded into individual wells on 2.0% agarose gels. The first well on each half of the
gel (upper and lower) contained a molecular weight marker of known band sizes. A
negative control, without DNA, was included with each set of reactions. Gels were
electrophoresed at 80 - 120 mA for approximately 16 - 18 hrs, stained with ethidium
bromide for 1 - 2 hr, destained with ddH2O for 2 - 3 hr, and photographed over UV
light using Polaroid Type 665 positive/negative film. Variations in banding pattern
among amplification products were analyzed from the resulting photographs.

DNA samples from four randomly chosen individuals were used for initial primer
screening. Among 100 primers screened, 59 successfully amplified the DNA samples
and were subsequently used in RAPD analysis on the entire set of individuals
described above. Individual bands produced with each primer were numbered and
scored as present or absent in each amplification product. Comparisons among and
within populations were based on the presence or absence, in a given individual, of
specific bands produced during amplification. To minimize potential scoring errors
resulting from any uncontrollable variation in the amplification environment, only
bands of moderate to high intensity (major bands) were scored for evaluation. Faint
(minor) bands were excluded from the analysis. Due to the low level of variation in
RAPD banding patterns detected among amplification products, quantitative analysis
was not deemed appropriate, and the results were evaluated qualitatively.

RESULTS

Fifty-nine of the 100 primers initially screened produced strong amplification
produgts and were evaluated across all samples. Of these, only seven (11.9%)
revealed genetic differences among the individuals evaluated, producing a total of 25
scorable bands. Only ten of the 25 bands showed variability among samples (Figure
1). The slight differences in banding patterns produced by these primers were
randomly, and approximately equally, distributed among populations and among
individuals within a population. No particular individual(s) showed an especially high
level of variation for RAPD banding patterns and no particular population(s) contained
an especially high proportion of variable individuals (Table 1).

Thirty-one (52.5%) of the primers evaluated revealed no variation for RAPD
banding patterns among samples, producing a total of 78 scorable bands (Figure 2).

Despite producing strong amplification products with the four samples evaluated
during primer screening, 21 (35.6%) of the primers tested repeatedly produced a large
number of non-scorable amplification products when evaluated over all samples.
Fifteen (71.4%) of the non-scorable primers failed to amplify a large number of the
samples and produced very faint bands among many of the samples that did amplify.
High levels of non-specific amplification occurred among a majority of the samples

26 PHYTOLOGIA July 1997 volume 83(1):22-33

with two (9.5%) of the non-scorable primers. This resulted in a considerable amount
of background smearing, thus preventing definitive scoring of bands. It could not be
determined whether either of these two classes of non-scorable primers would have
revealed variability among individuals if band scoring had been possible across all
samples. No variation was observed, however, among those samples that were
successfully amplified with any of these primers.

Four primers (19.0%) produced unique, repeatable banding patterns for each of
the samples evaluated. These were also considered non-scorable, as they could not be
used to identify relative levels of genetic variation among individuals or populations.

DISCUSSION

Despite showing strong amplification during primer screening, several primers
produced non-scorable banding patterns when evaluated across populations. This was
most likely due to high levels of contaminants, particularly polysaccharides,
complexed with many of the extracted DNA samples (Demeke & Adams 1992; Fang et
al. 1992). Several methods for removing the contaminants were attempted (Demeke &
Adams 1992; Fang et al. 1992; Maniatis et al. 1982; Murray & Thompson 1980; Ranu
1996, pers. comm.), with varying degrees of success. To compensate for the high
levels of complexed contaminants, the concentration of template DNA in each reaction
mixture was reduced to 0.5 - 1.0 ng. This allowed for adequate amplification of
template DNA with most of the primers evaluated, while reducing interference from
the complexed contaminants. Several primers still failed to yield consistently clean
amplification products across samples, resulting in either a large number of non-
amplified extracts or a high degree of background smearing.

The results of the RAPD analysis indicate the Fort Stewart Balduina atropurpurea
populations evaluated are quite similar in genetic composition. While only 25% of the
Fort Stewart populations were evaluated in this study, these populations represent 2
diversity of the habitats found among on-post populations. The five populations
chosen for genetic diversity analysis also had previously been sampled for
morphological variation (D. Lincicome, unpubl. data). Once morphological analysis is
complete, these two measures of diversity will be compared. An in-depth study of
habitat characteristics also should be conducted for these populations and the results
evaluated against morphological and genetic diversity. Site and population data
obtained thus far include population size; approximate number of individuals per
population; associated herbaceous, shrub, and tree species; soil type and nutrient
content; evidence of fire or disturbance; and general site quality (Helton 1995; unpubl.
data). The majority of the site and population data have not yet been analyzed. Soil
sample analyses revealed similarities among the on-post sites. The Fort Stewart
populations occur on slightly acidic (pH 3.9 - 5.0) soils with a sandy loam texture and
a relatively low organic matter content (1.0 - 6.5%). Soil nutrient content varied
considerably from site to site, particularly levels of phosphorus, potassium, and iron
(unpubl. data).

Halward et al.: Genetic diversity in Balduina 27

T2234 5 CEN 7 28 9510 HAZ B46. 1748 19 20-21

22 23 24 25 26 27 28 29 30 31

Figure 1. Banding patterns from amplification of Ba/duina
atropurpurea DNA using RAPD analysis.
Amplification products show variation among
DNA extracts. Both inter- and intra-population
variation is evident. Lanes 1 and 22: molecular
weight marker; Lanes 2-7: DNA extracts from
population 1; Lanes 8-13: DNA extracts from
population 2; Lanes 14-19: DNA extracts from
population 3; Lanes 20-21 and 23-26: DNA
extracts from population 4; Lanes 27-30: DNA
extracts from population 5; Lane 31: negative
control (no DNA).

28 PHYTOLOGIA July 1997 volume 83(1):22-33

1234567 8 9 1011 1213 14 15 16 17 18 192021 22

23 24 25 26 27 28 29 K 31

Figure 2. Banding patterns from amplification of Ba/duina
atropurpurea DNA using RAPD analysis.
Amplification products show no variation among
DNA extracts. Lanes 1 and 23: molecular
weight marker; Lanes 2-7: DNA extracts from
population 1; Lanes 8-13: DNA extracts from
population 2; Lanes 14-19: DNA extracts from
population 3; Lanes 20-22 and 24-26: DNA
extracts from population 4; Lanes 27-30: DNA
extracts from population 5; Lane 31: negative
control (no DNA).

Halward et al.: Genetic diversity in Balduina 29

Although the morphological analysis is not yet complete, initial evaluations
revealed differences among the on-post populations for several vegetative and seed
characters (D. Lincicome, unpubl. data). The variance observed for those characters,
however, is high, thereby reducing differences among character means for the
populations (D. Lincicome 1996, pers. comm.). Thus, levels of morphological
variation within and among populations may be similar. This is consistent with the
results from the RAPD analysis. While little genetic variation was detected overall
with RAPD analysis, the variation that was observed occurred with similar frequency
both among and within populations.

If considerable inter- or intra-population variability for morphological characters
should ultimately be found, this would not necessarily reflect the presence of high
levels of genetic variation among or within populations. Morphological variation is a
product of differential gene expression and may not be correlated with underlying
levels of genetic variation. Populations may be very similar in genetic composition,
yet show considerable morphological variation. This can result from differences in
phenotypic expression in response to environmental differences (Williams et al. 1995)
or from interactions among a small number of genes controlling morphological
characters (Kochert et al. 1991).

Given the out-crossing nature of Balduina atropurpurea, it was not unexpected to
observe genetic variation within populations, as well as among populations. Many
questions remain unanswered regarding the reproductive biology of this species
including identification of the primary pollinator(s), the distance viable pollen can
“travel” between populations, and the origin(s) of the on-post populations. Thus, the
degree to which populations might be genetically differentiated from one another
cannot be predicted. Among the populations evaluated, population ‘S’ exists in a
distinct location relative to the others and would have the lowest probability for gene
exchange with surrounding populations. Population ‘5S’, however, did not show a
high proportion of genetic variation relative to the other populations evaluated nor was
it genetically distinct from the remaining populations based on RAPD banding
patterns.

It would be beneficial to evaluate the remaining sixteen on-post populations using
RAPD analysis to determine whether any of these populations contain unique genetic
characteristics. Such analyses would be time-consuming and expensive given the low
levels of genetic variation detected among the populations evaluated thus far. In the
short term, conducting in-depth habitat characterizations for each population, in
combination with evaluations of reproductive success and population sustainability,
would aid management decisions. Ideally, genetic evaluations should occur
simultaneously with habitat characterization studies since long-term survival of the
species is ultimately dependent upon maintaining adequate genetic diversity among and
within populations. Genetic diversity provides a species with a means for better
adaptability to environmental changes and/or habitat disturbance.

To obtain a better understanding of the relative value of the Fort Stewart
populations compared to surrounding populations, we recommend that habitat
characteristics, population parameters, morphological variation, and genetic diversity
also be evaluated for several off-post populations and compared to results obtained
from the on-post populations. A diverse range of populations should be sampled,
including nearby off-post populations in Tattnall and Bulloch counties, Georgia, as

30 PHYTOLOGIA July 1997 volume 83(1):22-33

well as more distant populations known to occur in Georgia, Florida, Alabama, and
Mississippi. This would aid in the development of a recovery plan for Balduina
atropurpurea within the region as a whole, in addition to a management plan for the
species at Fort Stewart.

ACKNOWLEDGMENTS

The authors express sincere thanks to the following individuals, without whom
this project would not have been possible: Dr. Bert Bivings, U.S. Army FORSCOM,
for supporting research efforts on Balduina atropurpurea at Fort Stewart; Christopher
Nelson and Kristina Vinsonhaler, Research Assistants, Colorado State University, for
providing technical assistance in RAPD analysis; Ron Determann, Atlanta Botanical
Garden, for growing the seedlings used in genetic analysis; David Lincicome,
University of Illinois, for collecting the seeds used for growing the seedlings, for
providing valuable information regarding the populations evaluated, and for his helpful
editorial comments; Carol Helton, for her botanical expertise and significant
contribution in the field at Fort Stewart; and The National Seed Storage Laboratory,
for providing us with laboratory space to carry out the study. The content and quality
of this manuscript were greatly enhanced by the critical reviews of Julie Laufmann,
National Seed Storage Laboratory and Christine Bern, Department of Range Science,
Colorado State University. This work was supported by the Strategic Environmental
Research and Development Program (SERDP), and by the U.S. Forest Service,
Rocky Mountain Range and Experiment Station, Fort Collins, Colorado.

LITERATURE CITED

Demeke, T. & R.P. Adams. 1992. Effects of plant polysaccharides and buffer
additives on PCR. BioTechniques 12(3):332-334.
Determann, R. 1996. Botanist, Atlanta Botanical Garden. Personal communication.
Fang, G., S. Hammar, & R. Grumet. 1992. A quick and inexpensive method for
removing polysaccharides from plant genomic DNA. BioTechniques 13:52-57.
Flather, C.H., L.H. Joyce, & C.A. Bloomgarden. 1994. Species endangerment
patterns in the United States. Gen. Tech. Rep. RM-241. U.S. Department of
Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment
Station, Fort Collins, Colorado. 42 pp.

Fort Stewart Natural Resources Office (FSNRO). 1996. Personal communication.

Helton, C. 1995. Final Report: Balduina atropurpurea Survey, Fort Stewart,
Georgia, December 1995.

Kochert, G., T. Halward, W.D. Branch, & C.E. Simpson. 1991. RFLP variability
ere (Arachis hypogaea L.) cultivars and wild species. Theor. Appl. Genet.

1:565-570.

Lincicome, D. 1996. Research assistant, University of Illinois. Personal
communication.

Lutz, K. 1995. Plant characterization abstract: Balduina atropurpurea. Draft Final

Halward et al.: Genetic diversity in Balduina 31

Report of Fort Stewart Inventory. The Nature Conservancy.

Maniatis, T., E.F. Fritch, & J. Sambrook. 1982. Molecular Cloning: A Laboratory
Manual. Cold Springs Harbor Laboratory. Cold Springs Harbor, New York.
Mississippi Natural Heritage Program. 1991. Unpublished list of rare vascular plant
taxa of Mississippi, from: Plants National Database, U.S. Department of

Agriculture, Natural Resources Conservation Service, Washington, DC.

Murray, M.G. & W.F. Thompson. 1980. Rapid isolation of high molecular weight
DNA. Nucleic Acid Research 8:4321-4325.

Parker, E.S. & S.B. Jones. 1975. A systematic study of the genus Balduina
(Compositae, Heliantheae). Brittonia 27(4):355-361.

Patrick, T. 1994. Personal communication with Inge Smith, North Carolina Natural
Heritage Program, Raleigh, North Carolina, October, 1994. (see Smith, I.K.,
1994, below)

Ranu, R.S. 1996. Professor of molecular biology, Colorado State University.
Personal communication.

Smith, I.K. 1994. Species stewardship summary: Balduina atropurpurea. Draft.
Element Stewardship Abstract for Balduina atropurpurea.- North Carolina Natural
Heritage Program, Raleigh, North Carolina.

Stewart, C.N. & L.E. Via. 1993. A rapid CTAB DNA isolation technique useful for
RAPD fingerprinting and other PCR applications. BioTechniques 14(5):748-750.

U.S. Department of Defense. 1994. Fort Stewart Inventory. Conducted by The
Nature Conservancy of Georgia. Prepared for the Department of the Army,
Headquarters 24th Mechanized Division, Fort Stewart, Georgia.

Williams, D.G., R.N. Mack, & R.A. Black. 1995. Ecophysiology of introduced
Pennisetum setaceum on Hawaii: the role of phenotypic plasticity. Ecology
76(5): 1569-1580.

APPENDIX 1
Methodology used for DNA extractions.
Adapted from Stewart & Via (1993).

For each sample, approximately 0.1 g of fresh leaf tissue was harvested and
mechanically ground in an individual tissue grinder. Five 1] B-mercapto-ethanol and 1
ml warm (600°C) CTAB extraction buffer (2% w/v CTAB, 1.42 M NaCl, 20 mM
TRIS-HCI pH 8.0, 2 % w/v PVP, and 5 mM ascorbic acid) were added, and the
mixture incubated in a 600C water bath for 30 min. Each sample was transferred to a
clean eppendorf tube and 500 yl chloroform:isoamy] alcohol (24:1) was added. The
samples were placed on a mechanical shaker at 100 rpm for 15 min., followed by
centrifugation at 10,000 rpm for 5 min. The upper phase of each sample was

transferred to a fresh eppendorf tube using a pasteur pipette and re-extracted with
chloroform:isoamyl alcohol. The DNA was precipitated out of each sample by adding
an equal volume of ice-cold (approximately 0°C) isopropanol to the tube and gently
inverting the mixture. Samples were placed in a -200C freezer overnight to further
precipitate the DNA. This was followed by centrifugation at 5,000 rpm for 5 min.

The supernatant was discarded and the pellets washed with 500 ul of a 0.2 M sodium

ae PHYTOLOGIA July 1997 volume 83(1):22-33

acetate/70% ethanol solution. The sodium acetate/ethanol mixture was added to the
tube, and the pellet was dislodged and allowed to soak for 10 min. The samples were
briefly centrifuged and the supernatant discarded. The pellets were air-dried and

resuspended in 200 pl Tris-EDTA. Extracted DNA samples were stored at -200C.

The quantity of DNA obtained per sample was measured using a
spectrophotometer, based on relative absorption of 260 and 280 nm wavelengths UV
radiation passing through the sample. The quality of the DNA extracted from each
sample was determined by electrophoresing a subsample through a 1.0% agarose gel,
staining the gel with ethidium bromide, and exposing the gel to UV light.

APPENDIX 2
Methodology used for RAPD analysis.
Adapted from T. Lowrey, University of New Mexico (unpubl.).

Each RAPD reaction mixture was prepared by adding the following reagents to a
sterile microcentrifuge tube: 17 wl sterile ddH2O, 5 pl Master Mix [10x
Electrophoresis Reaction Buffer (Boehringer Mannheim); 10 mM each dATP, dCTP,
dGTP, and dTTP; ddH2O; and 1 M MgCl? (magnesium chloride), (bringing the total
MgCl? concentration to 2 mM)], | pl (5 picamoles) primer (Operon Technologies,
Inc., Alameda, California), and 1 11 diluted DNA sample (0.5 - 1.0 ng). The reaction
mixture was gently vortexed, then briefly centrifuged to collect the mixture at the
bottom of the tube. Each reaction mixture was overlain with approximately 50 jl
electrophoresis grade mineral oil to prevent evaporation during amplification. The
reaction tubes were placed into individual wells, to which one drop of mineral oil had

been added, in a DNA thermal cycler (MJ Research Inc. PTC 100 Programmable
Thermal Controller). The amplification program used was as follows: (Step 1) “Hot

Start’ of 2 min. @ 940C; (Step 2) addition of 0.5 unit Taq DNA polymerase
(Boehringer Mannheim) to each reaction tube, @ 800C (held for 20 min.); (Step 3)
time delay of 3 min. @ 940C; (Step 4) 35 cycles, each consisting of 1 min. @ 940C
(denaturing), 1 min. @ 380°C (first annealing), 30 sec. @ 540°C (second annealing), 2
min. @ 72°C (elongation); (Siep 5) 15 min. @ 720°C (final elongation); (Step 6)
indefinite soak @ 49C.

Table 1. Variation in RAPD Banding Patterns Among and Within Populations of
Balduina atropurpurea.\,2

Primer # OPZ-7

Pattern (+++) I (B1, B7, B11, B13); II (B22, B35, B38); II (B41, B58, B67);
IV (B70, B75, B79); V (B85, B88)

Pattern (+-+) I (B3, B14); II (B26, B34, B39); III (B46, B49, B64); IV (B71,

Halward et al.: Genetic diversity in Balduina 33

B74, B76); V (B87)

Primer # OPZ-9*
Pattern (++) 1(B1, B14); I (B34); I (B58, B64); IV (B76, B79); V (B82)
Pattern (+-) I (B3, B7, B11, B13); Il (B22, B26, B35, B38, B39); I (B41,

B46, B49, B67); IV (B70, B71, B74); V (B85, B87, B88)
* B75 missing (did not amplify with this primer)

Primer OPT-20*
Pattern (++++++) I(B1, B3, B7, B14); Il (B26, B35, B38, B39); Ill (B58, B67);
IV (B71, B75, B79); V (B85, B87, B88)
Pattern (++++-+) I (B11, B13); I (B34); I (B41, B46, B49, B64); IV (B70,
B74, B76); V (B82)
* B22 missing (did not amplify with this primer)

Primer OPAL-18*

Pattern (+++) I (B7); 0 (B26, B34); Il (B46, B64, B67); IV (B71); V (B85,
B88)
Pattern (+-+) I (B3, B11, B13, B14); I (B35, B38, B39); Il (B41, B49,

B58); IV (B70, B74, B75, B76, B79); V (B87)
* B1, B22, and B82 missing (did not amplify with this primer)

Primer OPA-7*
Pattern (+++) I (B13); I (B46, B58); IV (B76)
Pattern (++-) I (B1, B3, B7, B14); 0 (B22, B26, B34, B35, B38, B39); II

(B41, B49, B64, B67); IV (B70, B71, B74, B75, B79); V (B82,
B85, B87, B88)
* B11 missing (did not amplify with this primer)

Primer OPJ-13

Pattern (+++) I (1, B3, B7, B11, B13, B14); I (B22, B26, B34, B35, B38,
B39); Ill (B41, B46, B49, B58, B64, B67); IV (B70, B71, B74,
B79); V (B82, B85)

Pattern (+-+) IV (B76); V (B87)

Pattern (-++) IV (B75); V (B88)

Primer OPJ-10*
Pattern (+4++++) _ III (B46, B58); IV (B70, B74, B76)
Pattern (++++-) I (B3, B7, B11, B13); Il (B22, B26, B35, B39); Ill (B41, B49,
B64); IV (B71, B75, B79); V (B82, B85, B87)
Pattern (+++-+) I(B1)
Pattern (++-+-) II (B38)
* B14, B34, B67, and B88 missing (did not amplify with this primer)

1] - V refer to population numbers.

2 Bl, B3, etc. refer to DNA extracts from individual plants evaluated in RAPD
analysis.

Phytologia (July 1997) 83(1):34-35.

SABAZIA LAPSENSIS, A NEW NAME FOR S. BREEDLOVE! B.L. TURNER
(1997) (ASTERACEAE, HELIANTHEAE), NON S. BREEDLOVE! B.L.
TURNER (1976)

B.L. Turner
Department of Botany, University of Texas, Austin, Texas 78713 U.S.A.

ABSTRACT

A new name Sabazia lapsensis, is provided for a later homonym of
Sabazia breedlovei.

KEY WORDS: Sabazia, Heliantheae, Asteraceae, nomenclature, later
homonym

K. Gandhi, ‘keeper’ of the Gray Card Index, Harvard University, has called to
my attention a grievous lapse in my recent description of Sabazia breedlovei, a
proposed new species from Guerrero, México, noting that I had already used the latter
name (based upon a different type, Turner 1976) for a Chiapan species, although the
latter was subsequently transferred to the genus Alepidocline (Turner 1990). In short,
Sabazia breedlovei B.L. Turner (1997) is a later homonym of S. breedlovei (1976)
and is therefore illegitimate.

To correct this embarrassing error I propose the following:

SABAZIA LAPSENSIS B.L. Turner, nom. nov. Based upon Sabazia breedlovei
B.L. Turner, Phytologia 82:278. 1997. (TYPE: Breedlove & Almeda 65204);
non S. breedlovei B.L. Turner, Wrightia 5:303. 1976. (TYPE: Breedlove &
Smith 77632).

The specific name is obviously derived from Latin Japsis, meaning “slip” or
“sleep”, a name which I have opted to pin on the taxon for the lapsitic error concerned.

34

Turner: New name in Sabazia 35

LITERATURE CITED

Turner, B.L. 1976. New species and combinations in Sabazia (Heliantheae,
Galinsoginae). Wrightia 5:302-305.

---------- . 1990. A reevaluation of the genus Alepidocline (Asteraceae, Heliantheae,
Galinsoginae) and description of a new species from Oaxaca, México. Phytologia
69:387-392.

a--------- . 1997 [1998]. Sabazia breedlovei (Asteraceae, Heliantheae), a new species
from Guerrero, México. Phytologia 82:278-279.

Phytologia (July 1997) 83(1):36-41.

EFFECT OF FARMYARD MANURE AND MINERAL FERTILIZERS ON
COLORATION, GROWTH, AND BIOMASS PRODUCTION OF AZOLLA
PINNATA

M. Akmal Siddiqi, Mohammad Athar*, & Shahbaz Ahmed

National Agricultural Research Center, Pakistan Agricultural Research Council, Park
Road, Islamabad 45500, PAKISTAN

*Present address: Department of Environmental Horticulture, University of
California-Davis, Davis, California 95616 U.S.A.; e-mail: matariq@ucdavis.edu.

ABSTRACT

An experiment was conducted in ponds to determine the effect of farmyard
manure and mineral fertilizers on the coloration, growth, and biomass
production of Azolla pinnata. Dark green color was observed in the treatment
supplied with farmyard manure and brilliant green color in the treatments
supplied with phosphorus, and phosphorus plus nitrogen. Plants without any
fertilizer were light green in color. The number of plants, and fresh and dry
weight of A. pinnata were significantly higher (P < 0.05) in farmyard manure
than other treatments. Addition of phosphorus, and phosphorus plus nitrogen
also significantly (< 0.05) increased the number of plants, and fresh and dry
weights of A. pinnata as compared to the control treatment. Results suggest
that farmyard manure is more effective and cheaper than phosphorus, or
phosphorus plus nitrogen for propagation and multiplication of A. pinnata.

KEY WORDS: Azolla, Azollaceae, fertilizer, ecology

INTRODUCTION

Azolla is a free floating aquatic fern belonging to the cryptogamic family
Azollaceae. It forms symbiosis with the nitrogen-fixing cyanobacterium Anabaena
azollae (Watanabe 1982). The importance of Azolla as an organic input in rice
cultivation is extensively reported (Yanni 1991; Kumarasinghe & Eskew 1993; Yanni
et al. 1994; Kundu & Ladha 1995). The genus is widely distributed throughout
temperate and tropical regions with A. pinnata as the most prevalent species in Asia
(Lumpkin 1987; Liu & Zheng 1992; Kundu & Ladha 1995).

36

Siddiqui et al.: Effects of fertilizer on Azolla 37

Azolla varies in color under different environmental conditions. Growth of Azolla
is also greatly affected by physico-chemical factors (Lumpkin 1987; Siddiqi et al.
1987; Bonetto & Carcano 1995). The Azolla-Anabaena symbiosis has been observed
both in the presence of combined nitrogen and in a nitrogen free medium (Lumpkin
1987; Watanabe 1982). But the optimum growth of Azolla requires fertilization with
phosphorus and sometimes with nitrogen (Watanabe et al. 1980; Aziz & Watanabe
1983; Watanabe & Ramirez 1990; Yanni et al. 1994). However, use of chemical
fertilizers for Azolla culture is limited due to economic reasons and environmental
restrictions (Kundu & Ladha 1995).

Azolla is found naturally in ditches, ponds and roadside streams in central Punjab,
Pakistan. It is gaining importance in rice growing areas due to its nitrogen-fixing
ability and green manuring property (Kumarasinghe & Eskew 1993; Kundu & Ladha
1995). The Azolla-Anabaena symbiosis can produce 1 ton of green manure per
hectare per day containing 3 kg of fixed nitrogen which is equivalent to 15 kg of
ammonium sulphate of 7 kg of urea (El-Bassel et al. 1994). The benefit of Azolla
incorporation on the yield of rice has been well demonstrated. Azolla has been used as
green manure for rice in China, Philippines, and northern Vietnam (Watanabe et al.
1989). The use of Azolla as green manure in wetland rice was summarized by
Watanabe (1987) from internationally coordinated work by the International Network
on Soil Fertility and Fertilizer Evaluation for Rice (INSFFER). Watanabe (1987)
mentioned that incorporation of one crop of Azolla gave an increased yield equivalent
to that given by 30 kg urea N per hectare. Siddiqi et al. (1985) and Bonetto &
Carcano (1995) reported 16-19% grain yield increase over control when Azolla was
incorporated into rice fields. The use of Azolla by farmers is limited by a number of
constraints. The most important constraint faced in the use of Azolla is the large size
of inoculum needed for a limited field area, especially with the difficulties of
preservation and transportation (Kulasooriya 1991). Besides, Azolla must be
produced and distributed fresh among rice farmers just before use. Attempts have
been made to identify factors affecting growth and nitrogen-fixing ability of Azolla-
Anabaena and to develop conditions suitable for its propagation, transportation and
utilization in rice (Siddiqi et al. 1987; Kumarasinghe & Eskew 1993; Kulasooriya et
al. 1994; Yanni et al. 1994). The current studies were designed to determine the effect
of farmyard manure and mineral fertilizers on the coloration, growth, and biomass
production of Azolla.

MATERIALS AND METHODS

The experiment was conducted in ponds constructed at the National Agricultural

Research Center, Islamabad (33° 42’ N, 73° 7’ E, elevation approximately 518 m).
Azolla pinnata was obtained from the Rice Research Institute, Kala Shah Kaku,
Punjab, Pakistan. Azolla plants were cultured in ponds measuring 3 x 3 x 1 m filled
with irrigation water from a nearby well. The bottom and side walls of all ponds were
lined with polyethylene sheets. There were four treatments replicated four times in a
completely randomized design. T-1 was control without addition of any fertilizer. T-2
received farmyard manure (FYM) at the rate of 10 ton/ha fresh weight. Farmyard

38 Ps PHYTOLOGIA July 1997 volume 83(1):36-41

manure used in this experiment contained buffalo and cow dung, liquid animal excreta,
and organic matter includmg rice and wheat straw. Farmyard manure contained
1.24% nitrogen, 0.75% phosphorus, and 1.07% potassium on a dry weight basis. T-
3 was supplied with phosphorus as single super phosphate (SSP) at the rate of 5
kg/ha. T-4 was ameliorated with 5 kg/ha of phosphorus from single super phosphate
plus one kg/ha nitrogen from urea. One kg (fresh weight) of healthy Azolla plants was
added to each pond during late March. Population was recorded after two and five
months in each pond. Sampling was done with a wooden quadrat measuring 50 x 50
cm taking five samples from each pond. Change in the color of Azolla was observed
visually and population was recorded by counting the total number of plants per
quadrat at each sampling. Biomass production per quadrat was expressed both as
fresh and dry weight. Dry weight was determined by drying the plants in an oven at

60° C for 48 hours. Results were analyzed statistically by the method of Steele &

Torrie (1980). Means were compared with the LSD multiple comparison test at P <
0.05.

RESULTS AND DISCUSSION

Azolla pinnata changed color from green to red within two weeks after application
of treatments and remained red until the first week of April when the plants were four

weeks old. With the increase in average day and night temperature from 20°/12° C

during March to 26°/15° C during April, the color began to change to green. The
average humidity was 85% during March and 61% during April. Azolla in all the four

treatments was green in May, when average day/night temperature was 29°/20° C and
average humidity was 71%. The color was again red in June/July when average

day/night temperature was 37°/22° C and 36°/25° C, and average humidity was 54%
and 70%, respectively.

The color of Azolla pinnata was light green in the control treatment but it was dark
green when treated with farmyard manure. Azolla plants in the phosphorus treatment,
and in phosphorus plus nitrogen treatment were brilliant green. Similar observations
were made by Watanabe et al. (1980, 1989), Siddiqi et al. (1985) and Watanabe &
Ramirez (1990) who reported that deficiency of phosphorus activates the production of
anthocyanin which causes the appearance of red color in Azolla.

The research on Azolla culture points out several constraints to its universal use
(Kulasooriya 1991; Kulasooriya et al. 1994; Bonetto & Carcano 1995). Among them,
excessive light and temperature have been mentioned (Chung 1987; Fiore & Gutbrod
1987; Watanabe et al. 1989). It has been reported that red coloration in Azolla due to
anthocyanin formation does not affect its growth and nitrogen-fixing ability (Chung
1987; Lumpkin 1987). The change in the color of Azolla at the beginning of the
experiment may be due to sudden change in environment including low temperature
which produced physiological stress in the plants. The average temperature increased

from 26°/15° C during April to 29°/20° C during May, and humidity also changed
from 61% during April to 71% during May providing optimum conditions for the

Siddiqui et al.: Effects of fertilizer on Azolla 39

growth of Azolla. Azolla showed green coloration and grew vigorously during this
period. The humidity for the normal growth of Azolla is above 60% but optimal
values range from 85% to 90% (Watanabe 1982). Average temperature increased to

37° C in June reducing the humidity by as much as 54%. Plants developed red color
indicating stress conditions. The next change in color was observed during
September/October when average humidity was 50% and average day/night

temperature was 33°/21° C and 25°/15° C, respectively. The color again changed to
red with a drop in temperature below 22°/10° C during the later months.

The number of plants and biomass production of Azolla pinnata in various
treatments is presented in Table 1. The number, fresh and dry weight of Azolla plants

two months after treatment were significantly higher (P < 0.05) with farmyard manure
than with other treatments. Addition of phosphorus, and phosphorus plus nitrogen

also significantly (P < 0.05) increased the growth and biomass of A. pinnata as
compared to the control treatment. Similar trends of growth and biomass production
were obtained in all treatments after five months. Fertilizing Azolla with compost can
be a good practice for the mass cultivation of Azolla but conclusive results are not
available (Liu & Zheng 1992; Bonneto & Carcano 1995). Phosphorus is also essential
for the optimal growth of Azolla (Watanabe et al. 1980; Aziz & Watanabe 1983;
Watanabe & Ramirez 1990; Yanni 1991; Yanni et al. 1994). Best results can be
obtained if phosphorus is supplemented with some nitrogenous fertilizer because
synergistic effects of amelioration promote multiplication and enhance growth of
Azolla (Yanni 1991; Yanni et al. 1994; Bonneto & Carcano 1995).

Table 1. Effect of various fertilizers on the population and biomass production in
Azolla pinnata.

Values in each column followed by the same letters are not significantly different at P

< 0.05.
FYM stands for farmyard manure and SSP for single super phosphate.

Results of this experiment suggest that farmyard manure is more effective and
cheaper than phosphorus, and phosphorus plus nitrogen for multiplication and

40 PHYTOLOGIA July 1997 volume 83(1):36-41

biomass production of Azolla pinnata. Azolla may be a promising crop for vast areas
of rice growing regions and it can also be a good substitute for substantial amounts of
expensive nitrogen without environmental damage. Azolla can either be repeatedly
harvested and incorporated as green manure or intercropped with rice to meet the
fertilizer requirements of rice. However, further experiments are required to identify
the stress factors affecting growth and multiplication of Azolla. Appropriate
techniques which are agronomically feasible and socially acceptable also need to be
developed for its economical propagation and use in rice. Future studies will be
directed along these lines to work out solutions to these problems for sustainable
farming systems.

ACKNOWLEDGMENTS

The authors are indebted to the Director, the Rice Research Station, Kala Shah
Kaku for supplying Azolla pinnata, and the Watershed and Agro-meteorology
Program of the Council for providing meteorological data. Thanks are also due to the
late Dr. G.R. Sandhu and Dr. Murray Dawson for many helpful suggestions and
criticisms during this experiment. Appreciations are also extended to anonymous in-
house reviewer(s) of the manuscript for the comments which undoubtedly improved
the quality of the paper. Technical assistance of Mr. Rashid Akhtar Chaudhary in
constructing, and Mr. Bashir Ahmad in maintaining the Azolla ponds is highly
appreciated.

REFERENCES

Aziz, T. & I. Watanabe. 1983. Influence of nutrients on the growth and mineral
composition of Azolla pinnata. Bangladesh J. Bot. 12:166-170.

Bonetto, C.A. & L.R. Carcano. 1995. Azolla utilization, rice yield and nitrogen
balance in Argentina. Intern. J. Ecol. Environ. Sci. 21:37-44.

Chung, C.-L. 1987. Re-evaluation of Azolla utilization. Bulletin International Rice
Research Institute, Manila, Philippines, pp. 67-75.

El-Bassel, A., F.A. Ghaz2!, & O. El -Farouk. 1994. Azolla cultivation in Egypt. In:
A.N. Hegazi, M. Fayez, & M. Monib (eds), Nitrogen Fixation with Non-
Legumes, pp. 155-156. The American University in Cairo Press, Cairo, Egypt.

Fiore, M. & K. Gutbrod. 1987. Use of Azolla in Brazil. In: Azolla Utilization,
International Rice Research Institute, Manila, Philippines. pp. 45-53.

Kulasooriya, S.A. 1991. Constraints for the widespread use of Azolla in rice
production. In: M. Polsinelli, R. Materassi, & M. Vincenzi (eds.), Nitrogen
Fixation, pp. 473-479. Academic Publishers, New York, New York.

Kulasooriya, S.A., G. Seneviratne, & C. Wijesundera. 1994. An evaluation of some
factors that limit the widespread use of Azolla in rice cultivation. In: A.N.
Hegazi, M. Fayez, & M. Monib (eds.), Nitrogen Fixation with Non-Legumes, pp.
469-474. The American University in Cairo Press, Cairo, Egypt.

Kumarasinghe, K.S. & D.L. Eskew. 1993. Isotopic Studies of Azolla and Nitrogen
Fertilizer on Rice. Kluwer Academic Publishers, Dordrecht, Germany.

Siddiqui et al.: Effects of fertilizer on Azolla 41

Kundu, D.K. & J.K. Ladha. 1995. Enhancing soil nitrogen use and biological
nitrogen fixation in wetland rice. Expl. Agric. 31:261-277.

Liu, C.-C. & W.-W. Zheng. 1992. Nitrogen fixation in Azolla and its utilization in
agriculture in China. In: G.-H. Hong (ed.), The Nitrogen Fixation and its
Research in China, pp. 525-537. Springer-Verlag, New York, New York.

Lumpkin, T.A. 1987. Environmental requirements for successful Azolla utilization in
agricultural production. In: Azolla Utilization, International Rice Research
Institute, Manila, Philippines. pp. 67-75.

Siddiqi, M.A., M. Athar, S. Ahmed, & G.R. Sandhu. 1987. Effect of pH on growth
and biomass of Azolla pinnata under pot conditions. Pak. J. Sci Ind. Res. 30:301-
302.

Siddiqi, M.A., M. Athar, N.A. Mirza, & G.R. Sandhu. 1985. Effect of Azolla
manuring on the yield of IR-6 variety of rice. In: K.A. Malik, S.H.M. Naqvi, &
M.LH. Aleem (eds.), Nitrogen and Environment, pp. 269-274. Nuclear Institute
of Agriculture and Biology, Faisalabad, Pakistan.

Steele, R.G.D. & J.H. Torrie. 1980. Principles and Procedures of Statistics.
McGraw Hill, New York, New York. ;

Watanabe, I. 1982. Azolla-Anabaena symbiosis: its physiology and use in tropical
agriculture. In: Y.R. Dommergues & H.G. Diem (eds.), Microbiology of
Tropical Soils and Plant Productivity, pp. 170-185. Martinus-Nijhoff Publisher,
New York, New York.

Watanabe, I. 1987. Summary report of the Azolla Program of the International
Network on Soil Fertility and Fertilizer Evaluation for Rice (INSFER). In: Azolla
Utilization, International Rice Research Institute, Manila, Philippines. pp. 197-
205.

Watanabe, I., N.S. Berja, & D.C. Del Rosario. 1980. Growth of Azolla in paddy
fields as affected by phosphorus fertilizer. Soil Sci. Plant Nutr. 26:301-397.

Watanabe, I., C. Lin, C. Ramirez, M.T. Lapis, T. Santiago-Ventura, & C.C. Liu.
1987. Physiology and agronomy of Azolla-Anabaena symbiosis. Plant and Soil
110:57-62.

Watanabe, I. & C. Ramirez. 1990. Phosphorus and nitrogen contents of Azolla
grown in the Philippines. Soil Sci. Plant Nutr. 36:319-331.

Yanni, Y.G. 1991. Efficiency of rice fertilization schedules including cyanobacteria
under soil application of phosphate and molybdate. World J. Microbiol.
Biotechnol. 7:415-418.

Yanni, Y.G., S.N. Shalaan, & M. El-Haddad. 1994. Potential role of Azolla as
green manure for rice in Nile delta under different levels of inorganic fertilization.
In: A.N. Hegazi, M. Fayez, & M. Monib (eds.), Nitrogen Fixation with Non-
Legumes, pp. 127-132. The American University in Cairo Press, Cairo, Egypt.

Phytologia (July 1997) 83(1):42-46.

STUDIES ON GEOCALYCACEAE (HEPATICAE). X. NEW TAXA AND NEW
COMBINATIONS IN CHILOSCYPHUS CORDA FOR AUSTRALASIA

John J. Engel
Department of Botany, the Field Museum, Chicago, Illinois 60605-2496 U.S.A.

ABSTRACT

Chiloscyphus subg. Lophocolea is a new _ combination.
Chiloscyphus erosus, C. fertilis, C. suboppositus, C. edentatus,
C. tuberculatus, C. connatifolius, C. parvispinus, C. semiteres
var. retusus, C. mittenianus var. obtusus, and C. mittenianus var.
symmetricus are described as new species and varieties from Australasia.
Chiloscyphus subporosus var. inflexifolius is a new combination.

KEY WORDS: Chiloscyphus, Geocalycaceae, Hepaticae, Australasia

The following new taxa and new combinations are the result of a systematic study
of the Australasian species of the genus Chiloscyphus, and a treatment of the genus for
a second volume of a Manual of New Zealand, Hepaticae. The names are here
published separately to make them immediately available for use.

1. Chiloscyphus subg. Lophocolea (Dum.) Engel & Schust., comb. nov.
BASIONYM: Jungermannia sect. Lophocolea Dum., Syll. Jungerm. Eur. 59.
1831. [TYPE: Jungermannia bidentata L.]._ Lophocolea (Dum.) Dum., Recueil
d’ Observ. Jungerm. 17. 1835.

2. Chiloscyphus connatifolius Engel, spec. nov. HOLOTYPE: AUSTRALIA.
Tasmania: Gordon River, Gorge Creek, near Pine Landing, sea level, Engel
14648 (F); Isotype: HO.

Folia dorsaliter connata. Amphigastrium bidentatum vel usque 0.30-0.45
mm longitudinis bilobum, segmentis semper magnioribus quam armaturis
foliorum ceteris praeditum, apicem cetera armatura adjecta carentem evolutum,
margines laminarum utrinsecus dente vel cilio armati. Tuberculae cellulorum
foliorum conspicuae perevolutaeque. Cellulae amphigastriorum grandes,
laeves, etuberculatae.

42

Engel: New Geocalycaceae 43

A combination of features will separate this species from all other species in sect.
Leucophylli. The leaves are dorsally connate and conspicuously tuberculate.
Underleaves are bidentate to bilobed to at most 0.45 mm, and the segments are
uniformly larger than any other underleaf armature. Underleaf cells, however, are
devoid of tuberculae.

3. Chiloscyphus edentatus Engel, spec. nov. HOLOTYPE: AUSTRALIA.
Tasmania: Cradle Mtn.-Lake St. Clair Natl. Park, Ballroom Forest, SW side of
Lake Dove, 950-1050 m, Engel 13993 (F); Isotype: HO.

Plantae ad 2.2 mm latae. Ramificatio precipue vel omnino terminalibus.
Folia dorsaliter discreta, verticale vel subverticale, valde dorsaliter assurgentia
et precipue erecta vel suberecta, transversa vel subsuccuba. Apex foliorum
interdum anguste rotundatus, retuse-bidentatus, symmetrice vel asymmetrice
breviter bifidus vel 1-dentatus. Corpora oleosa 2(3), elliptica ad longe linearia.

Chiloscyphus edentatus is allied to C. suboppositus, but differs from that species
by the predominantly to exclusively Frullania type branching; the variable leaf apices,
which are narrowly rounded to retuse-bidentate to symetrically or asymmetrically short
bifid; the vertical to subvertical, strongly dorsally assurgent and mostly erect to
suberect leaves which are consistently free dorsally; and the fewer number of oil-
bodies per cell, being 2(3) vs. 3-5 per cell in C. suboppositus.

4. Chiloscyphus erosus Engel, spec. nov. HOLOTYPE: NEW ZEALAND.
North Is., South Auckland Prov., Plateau E of Waiotapu Valley, ca. 1800 ft.,
Allison 3569 (CHR!).

Caules demum plerumque flagelliformes; cellulae caulis parietibus distincte
tenuibus praeditae. Amphigastria aspectu ventrali convexa, bifida usque 0.9
longitudinis, marginibus laminae utrinque omnino vel maximam parte
appendiculo dentiformi vel laciniiformi armatis. Gemmae abundantes.

This species is related to Chiloscyphus perpusillus (Hook. f. & Tayl.) Engel, but
differs in several respects. The leaf apex (gemmiparous leaves) becomes
progressively more erose, and with continued gemmae formation the lobes disappear

altogether (the leaf apex then + broadly rounded). The leaves often with 1-2 accessory
lobes at apex lending a ragged appearance. Vegetative branches are all or mostly
intercalary (both lateral and ventral).

5. Chiloscyphus fertilis Engel, spec. nov. HOLOTYPE: AUSTRALIA. New
South Wales: Lane Cove, Forsyth 60 as L. bridellii--c. sporo.+ (male) (NSW!).

Ramificatio tantum lateri-intercalaris, ramis terminalibus carens. Caules
cellulis corticalibus medullaribusque parietibus percrassis praeditis vestiti.
Apex marginesque foliorum integri; trigonae gangliiformes. Bracteola
foeminea 0.2-0.4 areae bracteae occupans. Lobi perianthii non divisi et integri

a PHYTOLOGIA July 1997 volume 83(1):42-46

vel repandi vel subdenticulati vel usque 1/2 numerii toti loborum parvi-bifidi.
Plantae fructibus persaepe adsunt.

The species is related to Chiloscyphus semiteres, but differs in 1) branching
strictly lateral-intercalary; 2) stems with cortical and medullary cell walls very thick; 3)
perianth lobes entire or repand-sparsely denticulate, or with 1-2 of lobes short-bifid,
but never with all 3 lobes bifid; and 4) female bracteole 0.2-0.4 bract area.

6. Chiloscyphus mittenianus (Col.) Engel

Chiloscyphus_ mittenianus (Col.) Engel var. obtusus Engel, var. nov.
HOLOTYPE: NEW ZEALAND. South Is., Otago Prov.: Mt. Maungatua, W of
Mosgiel, 760 m, Engel 17768 (F); Isotype: CHR.

A varietate typica foliis plerumque dorsaliter connatis, apice non diviso,
integro, angustate vel interdum late rotundato differt.

Chiloscyphus mittenianus (Col.) Engel var. symmetricus Engel, var. nov.
HOLOTYPE: NEW ZEALAND. South Is., Westland Prov.: Westland Natl.
Park, track to Alex Knob, off track to Louisa Peak, 1170 m, Engel 18973 (F);
Isotype: CHR.

A varietate typica foliis uniformiter dorsaliter liberis, apice subaeque vel
aeque bifido segmentis piliferis praedito, segmento ventrali in stratum
uniseriatum e 5-7(-8) cellulis compositum terminanti differt.

7. Chiloscyphus parvispinus Engel, spec. nov. HOLOTYPE: NEW
ZEALAND. South Is., Otago Prov.: S side of Mt. Cargill, just below summit, N
of Dunedin, ca. 2200 m, Engel 17563 (F); Isotype: CHR.

Plantae dioecae. Folia tota bifida, pagina dorsali hispida, supra luminem
cuiusque cellulae prominenta brevi-conica e 1(2) cellulis composita obsita,
pagina ventrali perlaevi. Segmenta amphigastriorum integra vel 1-2 dentibus
armata, armatura nullo modo regulariter opposita. Perianthium et paginae
utrinque bractearum gynoicialium hispidae.

Chiloscyphus parvispinus differs from the related C. gippslandicus Engel &
Schust. of Tasmania and Australia by the 1(2) celled surface teeth, which are
juxtaposed over the lumen of most lamina cells; the uniformly hispid, consistently
bifid leaves; and the hispid perianth and gynoecial bract surfaces.

8. Chiloscyphus semiteres (Lehm.) Lehm. & Lindenb.

Chiloscyphus semiteres (Lehm.) Lehm. & Lindenb. var. retusus Engel, var.
nov. HOLOTYPE: AUSTRALIA. New South Wales: Murrumbidgee River,
Rules Point, 37 km NW of Adaminaby, Streimann 7482 (CBG!).

Engel: New Geocalycaceae 45

A varietate typica apicibus saepe retusis vel curto-bifidis, lobis plerumque
rotundatis differt.

The Chiloscyphus semiteres complex also includes the following:

Chiloscyphus platensis (Mass.) Engel, comb. nov. BASIONYM: Lophocolea
platensis Mass., Atti Accad. Sci. Mediche Natur. Ferrara 80(3/4):12. 1906 of NE
Argentina and SE Brazil.

9. Chiloscyphus suboppositus Engel, spec. nov. HOLOTYPE: AUSTRALIA.
Tasmania: Cradle Mt.-Lake St. Clair Natl. Park, Pine Valley, Cephissus Falls,
NNW of L. St. Clair, 850 m, Engel 14247 - c. sporo. (F); Isotype: HO.

Plantae dioecae, magnae, usque 5 mm latae. Ramificatio maximam partem
intercalaris. Folia subopposita horizontalia, late patentia, vulgo dorsaliter connata;
apices non congruente breviter bifidi segmento dorsali perparviori; segmentum
ventralis folii varium, acutum vel acuminatum interdum apiculatum; margines folii
dorsales ventralesque integri. Amphigastria usque 0.5 longitudinis divisa.

Chiloscyphus suboppositus is a close relative of C. trialatus (Gott.) Engel &
Schust., but may be distinguished from that species by 1) the shallowly divided
underleaves, divided to at most 0.35 mm; 2) the dioecious condition; and 3) the
opaque and rigid texture.

10. Chiloscyphus subporosus var. inflexifolius (Steph.) Engel, comb. & stat
nov. BASIONYM: Lophocolea inflexifolia Steph., Spec. Hep. 6:278. 1922.

11. Chiloscyphus tuberculatus Engel, spec. nov. HOLOTYPE: NEW
ZEALAND. South Is., Southland Prov.: Fiordland Natl. Park, Tutoko River, W.
of Milford Sound, 50 m, Engel 18844 (F); Isotype: CHR.

C. aculeato Mitt. similis, sed ramificatione frullanioidea plerumque
terminali, tuberculis folii bene evolutis in paene toto celularum conspicuis,
foliis dorsaliter liberis usque 0.30-0.35 longitudinis bifidis, cellulis folii

medianis 17-23 ym latis x 20-25 um longis, lobulis bractearum mascularum
saccati, a latere viso sacculi verrucoso-mamillato differt.

The species is closely allied to Chiloscyphus aculeatus Mitt., but may be
distinguished from it by predominately terminal, Frullania-type branching; the well-
developed tuberculae, which are conspicuous on nearly all leaf cells; the more deeply
bifid leaves (divided to 0.30-0.35 mm), which are free dorsally; and the smaller leaf
cell size.

The following combination also is required:

46 PHYTOLOGIA July 1997 volume 83(1):42-46

Chiloscyphus profundus subsp. cladogynus (Schust.) Engel, comb. nov.
BASIONYM: Lophocolea heterophylla subsp. cladogyna Schust., Hep. Anthoc.
N. Amer. 4:223. 1980.

Phytologia (July 1997) 83(1):47-52.

CERATOZAMIA MIXEORUM (ZAMIACEAE), A NEW SPECIES FROM
OAXACA, MEXICO WITH COMMENTS ON DISTRIBUTION, HABITAT, AND
SPECIES RELATIONSHIPS

Jeffrey Chemnick!, Timothy J. Gregory, & S. Salas-Morales?

1114 Conejo Road, Santa Barbara, California 93103 U.S.A.
2Sociedad para el Estudio de los Recursos Bioticos de Oaxaca, A.C. MEXICO

ABSTRACT

Ceratozamia mixeorum spec. nov., from Oaxaca, México is described
and illustrated. The species differs from others in the genus by the presence of
both remarkably long peduncles bearing the megastrobili and microstrobili,
and large, arching leaves with numerous, wide leaflets. Its affinity is
unresolved at present, but it is likely to be close to C. matudae. Ceratozamia
mixeorum is known only from cloud forest on montane peaks of the Sierra
Mixes in central Oaxaca, ranging in elevation from 1440 m to 1895 m.

KEY WORDS: Ceratozamia, Zamiaceae, Mexico, Oaxaca, systematics

Ceratozamia mixeorum Chemnick, Gregory, & S. Salas-Morales, spec. nov.
TYPE: MEXICO. Oaxaca: Vicinity of Juquila Mixes, May 1997, Chemnick,
Gregory, & S. Salas-Morales. HOLOTYPE: HNT; Isotypes: to be distributed to
FTG and XAL.

Truncus semihypogaeus, ad 48 cm altus; folia pauca, usque 8, glabra;
petiolus teresve, 61-81 cm longus, parte infima dilatatus, pauca spinis armatus;
rachis subteres, supra bisulcata, in dimido inferiore, paucis spinis armata,
supra fere inermis vel inermis, in cuspidem 10-25 mm longam excurrens;
foliala opposita vel subopposita, 30-40 juga, lanceolata vel falcata, 27-37 cm
longa, 23-27 mm lata, coriacea, falcata, basi attenuata, apicam lanceolata
acuminata, margine integerrima, revoluta; strobilus microsporangiatis lineari-
cylindricus, 22-24 cm longus, 70-75 mm latus; pedunculus tomentosus, 13.5-
15.0 cm longus, 18-20 mm latus; strobilus megasporangiatis cylindricus
pendulus, apice mucronatus, 24.0-30.6 cm longus, 12.2-15.2 cm _ latus;
pedunculus tomentosus, 12.5-23.0 cm longus, 1.5-1.9 latus.

47

48 PHY TOLOGIA July 1997 volume 83(1):47-52

Stems mostly solitary, semihypogeous, cylindric 34-125 cm long, 14-18 cm in
diameter, smooth, medium brown, with no protruding leaf bases, approximately 20-
25% of the mature plants bifurcate, some individuals with up to 4 branches of nearly
equal length originating below grade, branches originate from subterranean
procumbent stems that are often gnarled and in varying degrees of decomposition;
leaves 1.46-1.98 m long, usually in whorls of 5-11, ascending pendulous, recently-
emerged and juvenile leaves bright pea-green, turning dark green with age, glabrous,
slightly lighter in color on abaxial surface, adult plants with up to 3 crowns of leaves;
petiole 45-85 cm long, green, round with an expanded base that is dark reddish brown
and forms a distinct ridge at junction with the petiole, 25 mm in diameter at petiole
base tapering to 10 mm in diameter at the mid-way point, moderately armed with
simple spines 3-5 mm long gradually decreasing in frequency distally, adaxial surface
shallowly bisulcate with grooves arising just above the petiole base and extending
distally to the first pair of leaflets; rachis round, arching, 50-85 cm, sparsely armed
with spines 3-5 mm long gradually decreasing in frequency distally, nearly unarmed
on the distal 25%; leaflets linear-lanceolate, acuminate, often falcate, moderately
coriaceous with margins slightly revolute and turned upward, with veins neither
conspicuously raised nor visible, flat to deflexed on rachis except basal 3-5 “pairs” that
are keeled, the median leaflets 24-39 cm long and 21-29 mm wide decreasing slightly
in length towards apex, total number of leaflets 49-77 arising opposite to subopposite
along rachis inserted 3-4 cm apart; microsporangiate strobilus elongate-conical,
solitary, 22-24 cm in length, 7.0-7.5 cm in diameter, tapering gradually towards apex,
microsporophylls 14-15 mm wide and 7-8 mm long, yellow green, peduncle 13.5-
15.0 cm in length and 18-20 mm in diameter, green but covered with reddish-brown
tomentum; megasporangiate strobilus cylindrical, 23.5-30.5 cm in length and 12-15
cm in diameter with mature megasporophylls arranged in 12 vertical “columns” and 8
horizontal “rows”, solitary, apiculum truncate, megasporophylls 24-28 mm long and
42-50 mm wide, green suffused with yellow, horns 5 mm long and inserted 10 mm
apart, decumbent at tip, with a 3 mm long by 3 mm wide triangulate process evident
on the upper facet of the megasporophyll between the horns; megastrobilus horizontal
at receptivity and pendant at maturity; peduncle 12.5-23.0 cm in length and 15-20 mm
in diameter, green with reddish-brown tomentum; sclerotesta ovoid, smooth, tan, 25-
32 mm in length and 18-20 mm in diameter; initial seedling leaf usually with 4 leaflets.
Developing strobili of both sexes appear orange in color from a distance due to the
concentration of dark red tomentum on light yellow green scales and are borne on
peduncles of nearly mature length even when the immature strobili are only 6-7 cm
long.

Etymology: The species is named for the people inhabiting the region of
distribution.

DISTRIBUTION AND HABITAT

Ceratozamia mixeorum is known only from cloud forest covering the peaks of two
adjacent mountains in the extreme eastern Sierra Norte de Oaxaca (Sierra Mixes),
ranging in elevation from 1440 m to 1895 m. Precipitation occurs throughout the
year.

Chemnick et al.: New Ceratozamia from Oaxaca 49

Figure 1. Ceratozamia mixeorum. A, Habit of plant with megasporangiate strobilus.
B, Megasporangiate strobilus. C, Microsporangiate strobilus. Died WE}
Megasporophyll with attached seed at maturity in two aspects showing details of
sarcotesta. F, Seed showing details of sclerotesta in two aspects.

47-52

volume 83(1)

PHYTOLOGIA July 1997

50

H, Leaf detail; section from mid-rachis.

Figure 2. Ceratozamia mixeorum.

Chemnick et al.: New Ceratozamia from Oaxaca 51

Habitat consists of very steep slopes with small pockets of remnant primary forest
now interdigitated with coffee and secondary growth. The cloud forest consists of
two arboreal strata. The upper canopy is comprised chiefly of Weinmannia pinnata,
Liquidambar sstyraciflua, Cyathea mexicana, Alchornea latifolia, Ticodendron
incognitum, Clethra mexicana, Oreopanax xalapensis, Quercus excelsa, and
Dendropanax. The understory consists mainly of Hedyosmum mexicanum,
Phyllonoma laticuspis, Rondeletia, and Ternstroemia oocarpa among others.
Ceratozamia mixeorum is the dominant bushy plant accompanied in the higher
elevations of its range by Eugenia. Though there is a paucity of herbaceous ground
cover, the overstory is laden with an abundance of epiphytes, predominantly orchids
and bromeliads. Ceratozamia mixeorum occurs on heavily shaded east- and west-
facing slopes in primary forest with Chamaedorea, Geonoma, Melastoma, Acanthus,
Ficus, Begonia, and Selaginella. The substrate consists of a light-colored crumbly,
rocky clay soil with a pH of 5 and outcroppings of sedimentary rock.

The entire locality is rapidly being cleared and planted almost to the tops of the
peaks and thus this cycad must be considered threatened. Local prohibition of further
deforestation to protect the watershed is a likely benefactor for this species as well.
Since ex situ specimens of Ceratozamia mixeorum are unknown, it appears that habitat
destruction is currently the sole threat to its existence. We have withheld the exact
locality to protect it from the depredations of collectors. In our most recent survey of
the locality in May, 1997, we observed approximately 500-1000 plants during one day
of field work. Seedlings were abundant. Continuous recruitment into the population
was evidenced by the occurrence of many juvenile and older plants in a gradation of
size up to coning plants. Nearby peaks of the surrounding mountains are likely to
contain additional populations of C. mixeorum but their existence is yet to be
determined because accessibility is difficult. It is noteworthy that numerous
individuals persist in the dense scrublike secondary growth just below the primary
forest of the mountain tops. The local name of C. mixeorum is “carrete” (ox cart)
because the children play with the microstrobilus in a related manner.

RELATIONSHIP TO OTHER SPECIES OF CERATOZAMIA

Ceratozamia mixeorum is most likely allied to C. matudae Lundell (1939) because
both taxa possess a long peduncle that is atypical for the genus. Apparently there are
several populations of C. matudae-like plants with long peduncles that occur in
Chiapas currently under investigation (Miguel A. Perez Farrera, pers. comm.). The
occurrence of various populations of Ceratozamia with elongated peduncles suggests a
complex that ranges throughout Chiapas and into central Oaxaca. The Sierra Madre
Sur contains several peaks of a similar elevation between the known localities of C.
mixeorum and C. matudae and further field studies will undoubtedly uncover new
populations of plants in what is emerging as the “C. matudae complex”.

Ceratozamia matudae occurs in cloud forest and is characterized by pendant cones
borne on elongated peduncles as in C. mixeorum. However, the leaves are much
shorter and the leaflets narrower in C. matudae and the cones are much smaller.
Ceratozamia zaragozae Medellin-Leal (1963) is characterized by a small, pendant

52 PHYTOLOGITA July 1997 volume 83(1):47-52

megastrobilus borne on an elongated peduncle and its leaves are small, spirally
ascending, and unarmed. Additionally, C. zaragozae occurs in a much drier habitat far
to the north. The other known Ceratozamia with elongated peduncles have larger
leaves than does C. matudae but are still much smaller than the leaves of C. mixeorum.
Vegetatively, C. mixeorum is similar to the various C. mexicana Brongniart (Vovides
et al. 1983; Stevenson et al. 1986) ecotypes by the presence of large, arching leaves.
However, the cylindrical-long shape, smooth texture, and branching habit of the stems
are distinct. The peduncle length, combined with the size of the leaf and character of
the stem, are diagnostic for C. mixeorum and thus it is easily separated from the other
similar species of Ceratozamia that occur in Oaxaca, C. robusta Miquel (Vovides et al.
1983; Stevenson et al. 1986), and C. whitelockiana Chemnick-Gregory (1995).

ACKNOWLEDGMENTS

We are grateful to Sherwin Carlquist and Dieter Wilken for reviewing the
manuscript and providing valuable assistance. We would like to thank Peter Fletcher,
Leo Schibli and the office and staff of SERBO (Sociedad Para El Estudio De Los
Recursos Bioticos de Oaxaca) for assistance and contributions in the field. SERBO
has also provided logistical and botanical assistance in the city of Oaxaca for which we
are grateful. We would like to thank Loran Whitelock for his aid in both the field
work and preparation of this manuscript. We appreciate the help of Alan Smith and
John Strother in constructing the species epithet. Finally we are indebted to Sergio
Castro for the illustrations.

LITERATURE CITED

Chemnick, J., & T.J. Gregory. 1995. A new species of Ceratozamia (Zamiaceae)
from Oaxaca, México with comments on distribution, habitat and relationships.
Phytologia 79:51-57.

Lundell, C.L. 1939. Studies of Mexican and Central American plants. VII. Lloydia
2:75-76.

Medellin-Leal, F. 1963. A new species of Ceratozamia from San Luis Potosi.
Brittonia 15:175-176.

Stevenson, D.W., S. Sabato, & M. Vazquez Torres. 1986. A new species of
Ceratozamia (Zamiaceae) from Chiapas, Mexico with comments on species
relationships, habitats and vegetative morphology in Ceratozamia. Brittonia 38:17-
26

Vovides, A.P., J.D. Rees, & M. Vazques Torres. 1983. Zamiaceae in Flora de
Veracruz, Fasiculo 26, INIREB, Xalapa, Veracruz, México.

Phytologia (July 1997) 83(1):53-57.

THE STATUS OF PARNASSIA (SAXIFRAGACEAE) IN THE WEST GULF
COASTAL PLAIN

B.R. MacRoberts', M.H. MacRoberts', Lynn M. Stacey”, & David C. Moore?’

"Bog Research, 740 Columbia, Shreveport, Louisiana 71104 U.S.A. & Herbarium,
Museum of Life Sciences, Louisiana State University, Shreveport, Louisiana 71115
U.S.A.

2USDA Forest Service, Kisatchie National Forest, Winn Ranger District, Winnfield,
Louisiana 71483 U.S.A.

Louisiana Natural Heritage Program, Dept. of Wildlife and Fisheries, Baton Rouge,
Louisiana 70898 U.S.A.

ABSTRACT

A recent find of Parnassia grandifolia DC. from central Louisiana led to a
review of the status of Parnassia in the West Gulf Coastal Plain.

KEY WORDS: Parnassia grandifolia DC., Parnassia asarifolia Vent.,
Saxifragaceae, Louisiana, Texas

Parnassia asarifolia Vent., a southeastern species, is found in only one area of east
Texas: northern Nacogdoches and southern Rusk counties (Correll & Johnston 1970;
Correll & Correll 1972; Godfrey & Wooten 1981; Johnston 1990; Jones et al. 1997).
It is not known from Louisiana, Oklahoma, Arkansas, or Mississippi. The closest
known occurrence to the Texas site is in Coosa Co., Alabama, 775 km east (Alfred
Schotz, pers. comm.).

Parnassia grandifolia DC., a southeastern species, is considered to be rare,
threatened, or of special concern throughout most of its range - Virginia south to
Florida and westward to Texas and Oklahoma (Correll & Johnston 1970; Correll &
Correll 1972; Eakes 1989; Godfrey & Wooten 1981; Johnston 1990; Jones et al.
1997). It is found in northern Arkansas (Smith 1988). There are several P.
grandifolia sites in southern Mississippi (Stone, Perry, Forrest, Newton, and Pearl
River counties) (Lowe 1921; Eakes 1989; Steve Leonard, pers. comm.; Ken Gordon,
pers. comm.) and in Alabama (Freeman 1978). Although there are numerous seeps in
the coastal plain area of Oklahoma, Taylor & Taylor (1978) found P. grandifolia in
only one, where it was common (Connie Taylor, pers. comm.). It has been reported

33

54 PHYTOLOGIA July 1997 volume 83(1):53-57

from two sites in Texas. Reverchon found P. grandifolia in 1902 near Swan, Texas,
in Smith County; Bob O’ Kennon saw it in northern Newton County in 1985 (Lindsay
Woodruff, pers. comm.; Geraldine Watson, pers. comm.; Bob O’Kennon, pers.
comm.), but there is no voucher for the Newton Co. locality. It was reported from
Louisiana by Riddell in 1852, a report that has been overlooked by subsequent
collators of Louisiana plant taxa (Thomas & Allen 1982; MacRoberts 1984, 1989;
Louisiana Natural Heritage Program 1995). We searched for Riddell’s collection of
Parnassia from Louisiana but were unable to locate it at US or NO, two repositories
where his specimens and those of other early Louisiana botanists may have been
housed (MacRoberts 1984). Riddell referred to the specimen as “Parnassia
Caroliniana, Michx. var. grandifolia.” Where in Louisiana it was found remains
unknown.

On March 21, 1997, while surveying for rare plants on the Winn Ranger District
of the Kisatchie National Forest, Natchitoches Parish, Louisiana, we found dozens of
Parnassia grandifolia growing in a small (5 m x 3 m) seepage area in a pine-
hardwood forest ravine at the base of a 20° west-facing slope, approximately 200 feet
from the top of the adjacent ridge and approximately 20 feet from a 2nd order
perennial stream. The plants were growing in sandy, mucky soil at the edge of slow
moving water that was surfacing at the seep. The plants were not in flower, but from
leaf form and dried scapes of the previous year they were unmistakably Parnassia. To
confirm our initial identification, we monitored them until they flowered in mid-
October. On October 29, 1997, the population had 32 flowering stems: 2 in flower
and 30 buds.

Soil information is given in Table 1. Soil at two places in the seep was taken from
the upper 10 cm next to the Parnassia and analyzed by A & L Laboratories, Memphis,
Tennessee.

Table 1. Soil characteristics of Parnassia grandifolia site.

yea, ae Exchangeable Ions (ppm) ods A

ee eee ee

The Louisiana Parnassia soils are essentially the same as those analyzed by Eakes
(1989) for southern Mississippi Parnassia sites except that there may be slightly more
calcium and magnesium in the Louisiana samples.

Associated species were Aster sp., Athyrium filix-femina (L.) Roth, Callicarpa
americana L., Carex atlantica Bailey, Carex crinita Lam., Carex debilis Michx.,
Chasmanthium laxum (L.) Yates, Dichanthelium sphaerocarpon (Ell.) Gould,

MacRoberts et al.: Parnassia in West Gulf Coastal Plain 55

Eupatorium fistulosum Barratt, Gelsemium sempervirens (L.) St. Hil., Hypericum
hypericoides (L.) Crantz, Lyonia ligustrina (L.) DC., Melanthium virginicum L.,
Mitchella repens L., Myrica cerifera L., Myrica heterophylla Raf., Osmunda
cinnamomea L., Osmunda regalis L., Oxypolis rigidior (L.) Raf., Platanthera
clavellata (Michx.) Luer, Rhododendron canescens (Michx.) Sw., Rhododendron
oblongifolium (Small) Millais, Rhynchospora glomerata (L.) Vahl., Rhynchospora
gracilenta A. Gray, Rubus spp., Scleria triglomerata Michx., Smilax glauca Walt.,
Smilax laurifolia L., Solidago patula Muhl. ex Willd. var. strictula Torrey & A. Gray,
Solidago rugosa P. Mill., Toxicodendron vernix (L.) Kuntze, Vaccinium spp., and
Viola primulifolia L. Canopy species included Acer rubrum L., Magnolia virginiana
L., Nyssa sylvatica Marsh., and Pinus palustris P. Mill., with a midstory of Nyssa
sylvatica and Persea borbonia (L.) Spreng. The seep was well shaded with total
canopy cover.

The ravine in which the seep occurred and surrounding ravines had numerous
seeps and three hillside pitcher plant bogs, but none had Pamassia grandifolia.
However, very few sites appeared to be permanently flowing like the P. grandifolia
Site.

The distribution of Parnassia asarifolia and P. grandifolia in Arkansas, Texas,
Oklahoma, Mississippi, and Louisiana is shown in Figure 1.

DOCUMENTATION

Parnassia grandifolia: LA: Natchitoches Parish, MacRoberts, MacRoberts, Stacey,
Moore s.n. [NLU]; MacRoberts & MacRoberts 3341, 3736 [| LSU]. OK: Choctaw
Co., Taylor & Taylor 23223, 23596, 23797, 27365 [DUR-BRIT]. TX: Smith Co.,
Reverchon s.n. [BRIT].

Parnasgsia asarifolia: TX: Rusk Co., Banks 3957 [ASTC]; Nacogdoches Co., Nixon
15086 [ASTC], Lacey 789 [BRIT], MacRoberts & MacRoberts 3344 [TEX],
MacRoberts & MacRoberts 3343 [LSU].

ACKNOWLEDGMENTS

The plants were found while the senior authors were engaged in a Challenge Cost-
Share Agreement with the Kisatchie National Forest. R. Dale Thomas aided with
distribution information. Steve Leonard, Clifton Eakes, and Ken Gordon provided
information on Mississippi populations. Bob O’Kennon, Lindsay Woodruff, Dorinda
Scott, Geraldine Watson, and Tom Wendt provided information on Texas Parnassia.
Robert E. Evans and Michael Dehnisch showed us Pamassia asarifolia populations in
Nacogdoches County, Texas. Connie Taylor provided information on the Oklahoma
gents. Alfred Schotz provided information on Pamassia in Alabama. Anne

56 PHYTOLOGIA July 1997 volume 83(1):53-57

e x ion of Parnassia grandifolia
(circles and R) and P. asarifolia (triangles) in
rkansas, Louisiana, Mississippi, Oklahoma, and

MacRoberts et al.: Parnassia in West Gulf Coastal Plain 57

Bradburn and Susan L. Richardson provided information about J.L. Riddell and
Parnassia. Connie Taylor and R. Dale Thomas reviewed an earlier version of the

paper.

LITERATURE CITED

Correll, D.S. & M.C. Johnston. 1970. Manual of the Vascular Plants of Texas.
Texas Research Foundation, Renner, Texas.

Correll, D.S. & H.B. Correll. 1972. Aquatic and Wetland Plants of the
Southwestern United States. Vol Il. Stanford University Press, Stanford,
California.

Eakes, C.B. 1989. Studies on the life history and habitat of the Bogstar (Parnassia
grandifolia DC.) in south Mississippi. MS Thesis, University of Southern
Mississippi, Hattiesburg, Mississippi.

Freeman, J.D. 1978. State Record for Parnassia (Saxifragaceae) in Alabama.
Castanea 43:191-192.

Godfrey, R.K. & J.W. Wooten. 1981. Aquatic and Wetland Plants of the
Southeastern United States: Dicotyledons. Univ. of Georgia Press, Athens,
Georgia.

Johnston, M.C. 1990. The Vascular Plants of Texas. Published by the author.
Austin, Texas.

Jones, S.D., J.K. Wipff, & P.M. Montgomery. 1997. Vascular Plants of Texas: A
Comprehensive Checklist including Synonymy, Bibliography, and Index.
University of Texas Press, Austin, Texas.

Louisiana Natural Heritage Program. 1995. Rare Plant list. Unpublished report,
Louisiana Department of Wildlife and Fisheries, Baton Rouge, Louisiana.

Lowe, E.N. 1921. Plants of Mississippi: A List of Flowering Plants and Ferns.
Bull. No. 7, Mississippi State Geological Survey, Jackson, Mississippi.

MacRoberts, D.T. 1984. The Vascular Plants of Louisiana: An Annotated Checklist
and Bibliography of the Vascular Plants Reported to Grow without Cultivation in
Louisiana. Bulletin of the Museum of Life Sciences, No. 6. Louisiana State
University, Shreveport, Louisiana.

MacRoberts, D.T. 1989. A Documented Checklist and Atlas of the Vascular Flora of
Louisiana. Bulletin of the Museum of Life Sciences, No. 9. Louisiana State
University, Shreveport, Louisiana.

Riddell, J.L. 1852. Catalogus Florae Ludovicianae. New Orleans Med. Surg.
Journal 8:734-754.

Smith, E.B. 1988. An Atlas and Annotated List of the Vascular Plants of Arkansas.
Fayetteville, Arkansas.

Taylor, R.J. & C.E. Taylor. 1978. Additions to the vascular flora of Oklahoma.
Sida 7:36 1-368.

Thomas, R.D. & C.M. Allen. 1982. A Preliminary Checklist of Dicotyledons of
Louisiana. Contributions of the Herbarium of Northeast Louisiana University,
No. 3. Monroe, Louisiana.

Phytologia (July 1997) 83(1):58-66.

ESTUDIO PALINOLOGICO DE LAS BURSERACEAE DEL ESTADO DE
QUERETARO, MEXICO*

Rodolfo Palacios-Chavez**
Ma. de La Luz Arreguin-Sanchez**
D. Leonor Quiroz-Garcia**

Escuela Nacional de Ciencias, Biolégicas del IPN, Prol. Carpio y Plan de Ayala, Col.
Sto.Tomés 11340 México, D.F

RESUMEN

Se estudia e ilustra la morfologia de los granos de polen de la familia
Burseraceae del Estado de Querétaro. Los géneros y las especies que
comprende son: Bursera fagaroides (HBK.) Engl. var. fagaroides, B.
galeottiana Engl., B. lancifolia (Schlecht.) Engl., B. morelensis Ramirez, B.
palmeri S. Wats., B. schlechtendalii Engl., B. simaruba (L.) Sarg., y Protium
copal (Schlecht. & Cham.) Engl. var. copal.

El polen de las especies de Bursera resulto ser aspidado, tricolporado o
triporado, esferoidal y con la ornamentacién estriada reticulada. En Protium
copal var. copal existen diferencias notables con el del género anterior no es
aspidado, es tricolporado con colpos bien definidos y la ornamentacién es lisa
o psilada. Ambos géneros pudieron ser separados por medios palinoldégicos.

PALABRAS CLAVE: Burseraceae, morfologia polen, sistematica

ABSTRACT

Pollen grains of genera and species belonging to Burseraceae from
Querétaro state, México, are studied and illustrated. The taxa are: Bursera
fagaroides (HBK.) Engl. var. fagaroides, B. galeottiana Engl., B. lancifolia
(Schlecht.) Engl., B. morelensis Ramirez, B. palmeri S. Wats., B.
schlechtendalii Engl., B. simaruba (L.) Sarg., and Protium copal (Schlecht. &
Cham.) var. copal.

*Trabajo subsidiado por la Direccién de Posgrado e Investigacion del IPN.
**Becarios de COFAA.

58

Palacios-Chavez et al.: Burseraceae pollen 59

Pollen of Bursera species usually is tricolporate or triporate, spheroidal and
with ornamentation striate reticulate and pollen of Protiwm lacks aspides, is
tricolporate and prolate, and usually with ornamentation smooth or psilate.
These differences facilitate the separation of genera by palynology, according
to the key included.

KEY WORDS: Burseraceae, pollen morphology, systematics

INTRODUCCION

La familia Burseraceae la constituyen cerca de 20 géneros y 680 especies que se
encuentran distribuidas en las partes tropicales de ambos hemisferios, (Engler 1913).
En México segtin Rzedowski y Guevara (1992) existen dos taxa, Bursera y Protium.

El primero agrupa mas de 100 especies que progresan en los bosques tropicales
caducifolios y matorrales xeréfilos, se encuentra distribuido desde el sur de Estados
Unidos hasta sudamérica, principalmente en la vertiente del Pacifico en México.

A Protium \o constituyen cerca de 75 especies que progresan en las zonas
tropicales de Asia, Madagascar, Mascarefias, y América (Rzedowski y Guevara 1992).

En el estado de Querétaro se encuentran las siguientes especies: Bursera
fagaroides (HBK.) Engl. var fagaroides (HBK.) Engl., B. galeottiana Engl., B.
lancifolia (Schlecht.) Engl., B. morelensis Ramirez, B. palmeri S. Wats., B.
schlechtendalii Eng|., B. simaruba (L.) Sarg., y Protium copal (Schlecht. & Cham.)
Engl. var. copal.

Los estudios palinolégicos, realizados en México de la familia Burseraceae
generalmente versan sobre el género Bursera y hasta ahora no ha sido descrito, el
polen de Protium tal vez por ser menos frecuente y contener un numero menor de
especies en el pais.

El polen de Bursera es de gran importancia en otros estudios como en las floras
fdésiles, donde es considerado como un indicador de clima caliente. Se encuentra en
forma abundante en las lluvias de polen actual y como fésil ha sido encontrado por
Graham (1976) en los sedimentos del Mioceno Superior en la formacion de Paraje
Solo de Coatzacoalcos, Veracruz.

Se ha encontrado también en los sedimentos del Mioceno Inferior del norte de
Chiapas (Palacios y Rzedowski 1993), cabe agregar que en dichos depdsitos también
fue encontrado polen fésil de Protium.

Sobre la morfologia del polen actual de Bursera existen varios trabajos regionales
de México. Palacios Chavez (1966) describe el polen de siete especies del estado de
Morelos; Gonzalez (1969) estudia el de dos taxa perteneciente a la flora del Valle del
Mezquital; Palacios Chavez (1984) en un trabajo mds amplio describe el de 49
especies. Rzedowski y Palacios (1985) mediante el andlisis de los granos de polen de

60 PHYTOLOGIA July 1997 volume 83(1):58-66

Bursera encuentran que dos de las especies pertenecen a Commiphora. Palacios
Chavez et al. (1986) describen el polen de dos especies de la flora del Valle de México,
y Palacios Chavez et al. (1991) incluyen la morfologia de una de las especies que se
encuentran en la flora de Sian Ka’an, Quintana Roo. Algunos otros investigadores .
también han estudiado la palinologia de la familia Burseraceae; entre ellos Erdtman
(1966), Huang (1972), La Asociacién de los palindélogos de la lengua francesa (1974),
Lobreau et al. (1975), Mitra et al. (1977), Roubik & Moreno (1991), y Harley & Daily
(1995) quienes describen las Burseraceae de la tribu Protieae.

MATERIALES Y METODOS

Todas jas muestras florales donde se obtuvieron los granos de polen fueron
tomadas del Herbario (ENCB) del Departamento de Botanica de la Escuela Nacional de
Ciencias Bioldégicas del Instituto Politécnico Nacional, se procuro que todas las
colectas fueran de diferentes localidades del estado de Querétaro o de otros lugares
cercanos; principalmente cuando no se encontraron ejemplares pertencecientes al
estado o las muestras carecian de polen, como es el caso de Bursera galeottiana. Los
granos de polen fueron procesados con la técnica de la acetolisis de Erdtman levemente
modificada y las observaciones se hicieron con el microscopio de luz. En las
descripciones se utiliza la terminologia de Ertdman (1966), la de Faegri & Iversen
(1975), y la de Punt et al. (1994). Las medidas de los ejes y diametros se expresan en
micras.

DESCRIPCION DE LOS GRANOS DE POLEN

Bursera fagaroides (HBK.) Engl. var. fagaroides 6 km. al NNW de Querétaro,
Municipio de Querétaro, J. Rzedowski 16240. Lam. I, Figuras 1 a 4.

Polen triporado, aspidado, semitectado, esferoidal, de 30(34)38 x 28(32)35 um.

P/E=1.06. Vista polar circular de 30(32)35 jum de didmetro. Exina de 2.4 um de
grosor, con la sexina de mayor espesor que la nexina, engrosada a la altura de los
poros hasta 4.5 «1m, superficialmente estriada reticulada. Poros mds 0 menos
circulares de 5(6)7 [zm de didmetro.

Bursera galeottiana Engl. Ixtla, Municipio el Grande, Estado de Guanajuato, J.
Rzedowski 37545. Lam. I, Figuras 5 a 7.

Polen tricolporado, aspidado, semitectado, esferoidal, de 30(31)33 x 28(30)34

uum. P/E= 1.03. Vista polar de 30(32)34 rm de didmetro. Exina de 3 jum de grosor,
con la sexina y la sexina de igual espesor, engrosandose a la altura de los poros hasta 5

Palacios-Chavez et al.: Burseraceae pollen 61

im, superficialmente estriada reticulada, colpos cubiertos con membranas lisas. Poros
lalongados de 5(6)8 [um de largo por 4(5)6 tm de ancho.

Bursera lancifolia (Schlecht.) Engl. 2 km al S de Ayutla, Municipio de Arroyo Seco,
Querétaro, R. Ferndndez 2461. Lam. I, Figuras 8 a 11.

Polen triporado, aspidado, semitectado, esférico de 30(32)34 um de didmetro.
Vista polar circular de 30(32)36 tm de didmetro. Exina de 3 tim de grosor, con la
sexina y la nexina de igual espesor, engrosdndose hasta 6 tum a la altura de los poros,
superficialmente estriada reticulada. Poros lalongados de 5(6)7 jm de largo por 2(3)4
im de ancho.

Bursera morelensis Ramirez Vista hermosa, Municipio de Cadereyta, Querétaro, J.
Rzedowski 43102. Lam. I, Figuras 12 a 15.

Polen tricolporado, aspidado, semitectado, esferoidal de 30(33)36 m x 28(31)34

pm. P/E=1.01. Vista polar circular de 30(31)34 j1m de didmetro. Exina de 2.4 um
de grosor, con la sexina y la nexina de igual espesor, engrosdndose a la altura de los

poros hasta 4 jm, superficialmente estriada reticulada. Colpos cortos, difusos y mal
definidos. Poros lalongados de 6(7)8 jm de largo por 4(5)6 tm de ancho.

Bursera palmeri S. Wats. 5 km al NNW de Querétaro, sobre la carretera a San Luis
Potosi, Querétaro, J. Rzedowski 16242. Lam. II, Figuras 17 a 20.

Polen tricolporado, aspidado, semitectado, esferoidal de 26(28)32 x 22(26)30

uum. P/E=1.1. Vista polar circular de 25(28)31 {1m de didmetro. Exina de 2.4 tum de
grosor, con la sexina y la nexina de igual espesor, engrosdndose a la altura de los

poros hasta 4 1m, superficialmente estriada reticulada. Colpos cortos y someros.
Poros lalongados de 5(6)8 jim de largo por 3(4)5 pm de ancho.

Bursera schlechtendalii Engl. 11 km. al S de Camargo, Municipio de Pefiamiller,
Querétaro, R. Fernandez 2876. Lam. Il, Figuras 21 a 24.

Polen triporado, aspidado, semitectado, esferoidal de 25(27)29 x 24(26)29 um.

P/E=1.1. Vista polar circular de 25(27)29 um de didmetro. Exina de 3 pm de grosor,
con la sexina y la nexina de igual espesor, engrosdndose a la altura de los poros hasta
4 um, superficialmente estriada reticulada. Poros lalongados de 3(4)5 tm de largo por
2(3)4 tum de ancho.

62 PHYTOLOGIA July 1997 volume 83(1):58-66

Lam. I. Bursera fagaroides var. fagaroides. 1. Vista ecuatorial superficial; 2. Seccién
transversal; 3. Vista polar superficial; 4. Corte 6ptico de la vista polar. B. galeottiana.
5. Vista ecuatorial superficial, mostrando ornamentacién y aberturas; 6. Corte dptico;
7. Corte 6ptico de la vista polar. B. lancifolia. 8. Vista ecuatorial superficial; 9. Corte
Optico de la vista ecuatorial, 10. Vista polar superficial; 11. Seccién dptica. B.
morelensis. 12. Vista ecuatorial superficial en 6ptica; 13. Vista ecuatorial seccién
Optica; 14. Vista polar superficial; 15. Vista polar en seccién 6ptica.

Palacios-Chavez et al.: Burseraceae pollen 63

32

Lam. Il. Bursera palmeri. 17. Vista ecuatorial superficial; 18. Vista ecuatorial, corte
Optico; 19. Vista polar superficial; 20. Corte dptico de la vista polar. B.
schlechtendalii. 21. Vista ecuatorial superficial; 22. Corte dptico; 23. Vista polar
superficial; 24. Corte dptico de la vista superficial. B. simaruba. 25. Vista ecuatorial;
26. Corte dptico de la vista ecuatorial; 27. Vista polar superficial; 28. Vista polar en
corte 6ptico. Protium copal var. copal. 29. Vista ecuatorial superficial; 30. Vista
ecuatorial en corte 6ptico; 31. Vista polar superficial; 32. Corte 6ptico de la vista polar.

64 PHYTOLOGIA July 1997 volume 83(1):58-66

Bursera simaruba (L.) Sarg. 2 km al NE de Ayutla, sobre el Cafién del rio Sta. Maria,
Municipio de Arroyo Seco, Querétaro, J. Rzedowski 43192. Lam. Il, Figuras 25
a 28.

Polen tricolporado, aspidado, semitectado, esferoidal de 28(33)38 x 27(32)36

yum. P/E=1.03. Vista polar circular de 25(27)29 jum de didmetro. Exina de 2.5 um
de grosor, con la sexina de mayor espesor que la nexina. Colpos cortos y cubiertos

con membranas lisas. Poros lalongados de 8(9)11 jm de largo por 3(5)6 ym de
ancho.

Protium copal (Schlecht. & Cham.) Engl. var. copal La Isla, 6 km al N de Carrizal,
Municipio de Jalpan, Querétaro, R. Fernandez 4474. Lam. Il, Figuras 29 a 32.

Polen tricolporado, tectado, prolato de 28(32)37 x 20(23)27 um. P/E=1.39.

Vista polar angular de 20(23)27 um de didmetro. Exina de 2.4 um de grosor, con la
sexina ligeramente de mayor espesor que la nexina, superficialmente psilada,
ligeramente reticulada en los polos. Colpos delgados y bien definidos, cubiertos con
membranas lisas. Poros lalongados de 8(9)12 jm de largo por 2.0(3.5)4.0 jum de
ancho, cubiertos con membranas granulosas. Indice del area polar 0.4 grande.

CLAVE PARA LA SEPARACION DE GENEROS

1.- Polen aspidado, generalmente esferoidal o esférico con la ornamentacién estriada
POH CUA. cain .c sce cscedccecoraecdeacesecnssecocwsnnla stacenaiee amen Bursera
1.- Polen sin dspides, prolato, con la ornamentacién lisa o psilada. ............. Protium

CONCLUSIONES

Las especies del género Bursera del estado de Querétaro presentaron granos de
polen con caracteristicas muy homogéneas. Las aberturas se encuentran en las partes
altas de los 4spides y pueden ser tricolporadas o triporadas, pudiéndos apreciar polen
triporado en B. fagaroides var. fagaroides, B. lancifolia, B. schlechtendalii, y
tricolporado en B. galeottiana, B. morelensis, B. palmeri, y B. simaruba, la
ornamentacién en todos los casos es estriada reticulada y la forma generalmente es
esferoidal. Los colpos son difusos y mal definidos, pero la abertura central se aprecia
visible, generalmente en forma transversal eliptica y siempre se encuentran en la parte
alta de los aspides, rara vez circular.

El polen de Protium resulto ser muy diferente con las siguientes caracteristicas
palinolégicas, carece de aspides, la forma es prolata, tricolporado, con colpos muy

Palacios-Chavez et al.: Burseraceae pollen 65

bien definidos y con la ornamentacién lisa o psilada, no existen levantamientos
sexinosos en los lugares donde se encuentran las aberturas y la vista polar es angular
(angulotriaperturada). Las diferencias palinolégicas entre ambos géneros permiten la
separacion de cada taxon.

BIBLIOGRAFIA

Association des palynologues de langue francaise. 1974. Pollen et spores d’Afrique
tropicale. Centre de Etudes de Geographic Tropicale, Domaine. Universitaire de
Bordeaux. Talence. 283 pp.

Engler, A. 1913. Die verbreitung der afrikanischen Burseraceen in verhialnis zuihrer
systematischer Gliederung un die Einteilung der Gattung Commiphora. Bot.
Jahrb. Syst. 48:443-490.

Erdtman, G. 1943. An Introducction to Pollen Analysis. The Ronald Press
Company, New York, New York.: 233 pp.

Erdtman, G. 1966. Pollen Morphology and Plant Taxonomy (Angiosperms). Hafner
Publishing Co., New York, New York. 553 pp.

Faegri, K. & J. Iversen. 1975. Text Book of Pollen Analysis. Hafner Publishing
Co., New York, New York. 553 pp.

Gonzalez Q., L. 1969. Morfologia polinica: la flora del Valle del Mezquital,
Hidalgo. INAH. Paleoecologia 3:1-185.

Graham, A. 1976. Studies in neotropical paleobotany II]. The Miocene communities
of Veracruz, Mexico. Ann. Missouri Bot. Gard. 63(4):787-842.

Harley, M.M. & D.C. Daly. 1995. Burseraceae Kunth. Protieae March. em. Engl.
Wold Pollen and Spores Flora 20: 1-144.

Huang, T.C. 1972. Pollen Flora of Taiwan. Nat. Taiwan University Botanic
Department Press, Taipei, Taiwan. 276 pp.

Lobreau, C., D. Oltmann, & H. Straka. 1975. Palynology Madaggassica et
Mascerinica. Pollen et Spores 17(1):11-157.

Mitra, K., Mondal, & S. Saha. 1977. The pollen morphology of Burseraceae.
Grana 16:75-79.

Palacios Chavez, R. 1966. Morfologia de los granos de polen de arboles del estado
de Morelos. An. Esc. Nac. Cienc. Biol. Méx. 16:41-93.

Palacios Chavez, R. 1984. Morfologia de los granos de polen de las especies del
género Bursera. Biotica 9(2):153-182.

Palacios Chavez, R., D.L. Quiroz G., M. de la L. Arreguin S., & D. Ramos Z.
1986. Morfologia de los granos de polen del género Bursera (Burseraceae) del
Valle de México. Phytologia 62(1):58-61.

Palacios Chavez, R., B. Ludlow W., & R. Villanueva G. 1991. Flora palinoldégica
de la Reserva de la Bidsfera de Sian Ka’an, Quintana Roo, México. Centro de
Investigaciones de Quintana Roo, México. 321 pp.

Palacios Chavez, R. & J. Rzedowski. 1993. Estudio palinolégico de las flores
fdsiles del Mioceno Inferior y principios del Mioceno Medio en la regién de
Pichucalco, Chiapas, México. Acta Botanica Mexicana 24: 1-96.

Punt, W., S. Blackmore, S. Nilsson, & A. Le Thomas. 1994. Glossary of pollen
and spore terminology. LPP Foundation, LPP Contributions. 1:1-71.

Roubik, W.D. & J. Moreno. 1991. Pollen and spores of Barro Colorado Island.
Ann. Missouri Botanical Garden. 268 pp.

66 PHYTOLOGIA July 1997 volume 83(1):58-66

Rzedowski, J. & R. Palacios Ch. 1985. La presencia de Commiphora (Burseraceae)
en México. Taxon 28(1,2/3):103-116.

Rzedowski, J. & F. Guevara-Féfer. 1992. Flora del Bajio y de regiones adyacentes.
Burseraceae. Instituto de Ecologia A.C. 3:1-46.

Phytologia (July 1997) 83(1):67-78.

MORFOLOGIA DE LAS ESPORAS DEL GENERO PTER/S PARA MEXICO

Maria de la Luz Arreguin-Sanchez, Rodolfo Palacios-Chavez, & David Leonor
Quiroz-Garcia*

Instituto Politécnico Nacional, Escuela Nacional de Ciencias Biolégicas, Plan de Ayala
y Carpio, Col. Santo Tomas, México, D.F. 11340 MEXICO

RESUMEN

Se estudia la morfologia de las esporas de catorce especies del género
Pteris presentes en México: Pteris altissima Poir., P. biaurita L., P. cretica L.,
P. erosa Mickel & Beitel, P. grandifolia L., P. longifolia L., P. muricata
Hook., P. muricella Fée, P. orizabae Mart. & Gal., P. paucinervata Fée, P.
podophylla Sw., P. pulchra Schlecht. & Cham., P. pungens Willd., y P.
quadriaurita Retzius. Se elaboro una clave palinolégica para separar las
especies de acuerdo con la morfologia de las esporas, se tomo en
consideracién la presencia o ausencia de collar ecuatorial, grosor de este y
ornamentacién. Se dan algunas interpretaciones tomando en consideracién la
morfologia de las esporas y se discute la posicién taxonémica de algunas
especies con base a los caracteres morfoldgicos encontrados y a los datos
bibliograficos.

PALABRAS CLAVE: Pteris, Pteridaceae, Pteridophyta, morfologia espora,
México

ABSTRACT

This paper deals with the spore morphology of fourteen species of Pteris
found in México: Pteris altissima Poir., P. biaurita L., P. cretica L., P. erosa
Mickel & Beitel, P. grandifolia L., P. longifolia L., P. muricata Hook., P.
muricella Fée, P. orizabae Mart. & Gal., P. paucinervata Fée, P. podophylla
Sw., P. pulchra Schlecht. & Cham., P. pungens Willd., and P. quadriaurita
Retzius. A key to taxa based on spore characters is included. The main
characters useful for identification are presence or absence of equatorial collar,
and its thickness and ornamentation. Some possible interpretations of the
spore morphology are given, and taxonomic position of some taxa is discussed
with regard to spore morphology and bibliographic data.

*Becarios de COFAA del IPN

67

68 PHYTOLOGIA July 1997 volume 83(1):67-78

KEY WORDS: Prteris, Pteridaceae, Pteridophyta, spore morphology, México

INTRODUCCION

Pteris es un género de aproximadamente 300 especies, 60 de ellas son americanas
(Mickel & Beitel 1988) y en México, a la fecha solo prosperan quince. El propédsito
del trabajo es el de conocer la morfologia de las esporas de Pteris y relacionarlas con
los estudios taxondémicos del género.

ANTECEDENTES

Entre los trabajos que aportan datos sobre la morfologia de las esporas del género
Pteris tenemos la de Lugardon (1963) en cuyo estudio sobre las Pteridaceae de
Francia, describe e ilustra las esporas de Pteris cretica L. y Pteris longifolia L. Nayar
(1964), menciona algunos caracteres de las especies de Pteris e ilustra a Pteris
wallichiana var. magna. Lugardon (1974) estudia la infraestructura de la pared de
Pteris longifolia. Tryon & Tryon (1982), incluyen fotomicrografias de ocho especies
americanas del género con breves comentarios sobre la ormamentacién. Tryon &
Lugardon (1991) mencionan los caracteres generales de 70 especies que estudiaron, de
los cuales cuatro prosperan en México, ademés se incluyen fotomicrografias de diez y
ocho taxones. Arreguin-Sanchez et al. (1996) estudian las esporas de cinco especies
mexicanas del género Pteris para el estado de Querétaro.

METODOLOGIA

Se tomaron muestras de las esporas de catorce especies de Pteris depositadas en
los herbarios de la Escuela Nacional de Ciencias Biolégicas del Instituto Politécnico
Nacional (ENCB) y del Instituto de Biologia de la Universidad Nacional Auténoma de
México (MEXU). Las esporas fueron tratadas con la técnica de Pl4 Dalmau (1961).
Las esporas de Pteris chiapensis no fueron estudiadas por no encontrarse material de
esta especie en los herbarios mexicanos, al parecer es una especie escasa y endémica
de Chiapas cuyos ejemplares estan depositados en herbarios del extranjero.

Arreguin-Sanchez, et al.: Morphology of Pteris spores 69
RESULTADOS

DESCRIPCION DE LAS ESPORAS

Pteris altissima Poir. 3-4 Km al E de Santa Maria Chimalpa, Oaxaca, H. Hernandez

831 (ENCB). Lamina I, Figuras | a 3.

Espora trilete, vista proximal triangular de 30.4(32.7)35.5 por 32.0(34.0)35.5
wim. Vista lateral o ecuatorial ovada, de 30.4(33.1)34.6 por 20.3(23.4)26.2 um.
Exina de 6 tim de grosor. Nexina de 1 pm, sexina de | tum y perina de 44m de
espesor, rugulada, vista ecuatorial con collar de 10 jm de grosor. Brazos de la laesura
de 10 a 13 um de largo por 1 jim de ancho.

Pteris biauritaL. Pochutla, Oaxaca, J. Mickel & Leonard 5166 (ENCB). Lamina I,

Figuras 4 a 6.

Espora trilete, vista proximal triangular de 39.0(40.7)42.3 por 43.0(45.0)46.5
im. Vista lateral o ecuatorial ovada de 40.6(41.7)44.0 por 26.0(28.3)30.5 pm.
Exina de 5 pm de grosor. Nexina de 1 jum, sexina de 2 jim de espesor, tuberculada y
perina de 2 tm de espesor, psilada, vista ecuatorial con collar de 10 {1m de grosor.
Brazos de la laesura de 13 a 17 sm de largo por menos de | jm de ancho.

Pteris cretica L. Cerro Frio, Tlazapotla, Municipio Puente de Ixtla, Morelos, A.
Bonfil 27] (ENCB). Lamina I, Figuras 7 a 9.

Espora trilete, vista proximal triangular de 48.0(45.0)51.5 por 45(42)48 um.
Vista lateral 0 ecuatorial ovada de 45.5(41.0)54.0 por 30(33)40 im. Exina de 6 um

de grosor. Nexina de | [im, sexina de 1 tum y perina de 4 pm de espesor, rugulada en
vista proximal y reticulada en vista distal, sin collar ecuatorial. Brazos de la laesura de

15 a 19 um de largo por 2 a 3 tum de ancho.

Pteris erosa Mickel & Beitel. La Cuesta, Talpa de Allende, Jalisco, R. McVaugh
23376 (MEXU). Lamina I, Figuras 10 a 12.

Espora trilete, vista proximal triangular de 33.0(33.5)40.0 por 37.0(39.0)42.3
lim. Vista lateral o ecuatorial ovada de 30.4(32.2)33.8 por 32.1(35.7)39.7 tm.
Exina de 3.3 tum de grosor. Nexina menor de | um, sexina de | um y perina de 1.5

im de espesor, rugulada, vista ecuatorial con collar de 8 um de grosor. Brazos de la
laesura de 13.5 a 16.0 um de largo por mas o menos de | [im de ancho.

70 PHYTOLOGIA July 1997 volume 83(1):67-78

10

LAMINA I. Pteris altissima. 1.- Vista proximal mostrando la
cicatriz. 2.- Ornamentacién de la espora. 3.- Vista ecuatorial
mostando el collar. Pteris biaurita. 4.- Vista proximal mostrando
la cicatriz. 5.- Ormanentacién de la espora. 6.- Vista ecuatorial
mostrando el collar. Pteris cretica. 7.- Vista proximal mostrando

cicatriz. 8.- Vista distal mostrando grosor de la exina. 9.-
Ornamentaci6n de la vista distal. Pteris erosa. 10.- Vista
proximal mostrando cicatriz. 11.- Ornamentacién de la espora.

12.- Vista ecuatorial mostrando el collar. La linea negra que se
encuentra al lado de las fotomicrografias corresponde a 10 uy.

Arreguin-Sanchez, et al.: Morphology of Pteris spores ral

J a
f bg wa
g ; 2 4 e 4
Pi
? P
ye %,
—
r
15
® Wy,
e
EZ Me ae gy
~ ~« 17 y
18

LAMINA II. Pteris grandifolia. 13.- Vista proximal mostrando
cicatriz. 14.- Ornamentacién de la espora. 15.- Vista ecuatorial

mostrando el collar. Pteris longifolia. 16.- Vista proximal
mostrando cicatriz. 17.- Vista distal mostrando grosor de la
exina. 18.- Ornamentacién de la vista distal. Pteris muricata.

19.- Vista proximal mostrando cicatriz. 20.- Ornamentacién de la
espora. 21.-- Vista ecuatorial mostrando el collar. Pteris
muricella. 22.- Vista proximal mostrando cicatriz. 23.-
Ornamentaci6én de la espora. 24.- Vista ecuatorial mostrando el
collar. La linea negra que se encuentra al lado de las
fotomicrografias corresponde a 10 uz.

72 PHYTOLOGIA July 1997 volume 83(1):67-78

42

LAMINA IV.- Pteris pungens 37.- Vista proximal mostrando
cicatriz. 38.- Ornamentacién de la espora. 39.- Vista ecuatorial
mostrando el collar. Pteris quadriaurita 40.-Vista proximal
mostrando cicatriz. 41.- Ornamentacién de la espora. 42.- Vista
ecuatorial mostrando el collar. La linea negra que se encuentra
al lado de las fotomicrografias corresponde a 10 pu

Arreguin-Sanchez, et al.: Morphology of Pteris spores

73

LAMINA III. Pteris orizabae. 25.- Vista proximal mostrando
Cicatriz. 26.- Vista ecuatorial mostrando grosor de la exina.
27.- Vista ecuatorial mostrando el collar. Pteris paucinervata
28.- Vista proximal mostrando cicatriz. 29.- Ornamentaci6én de la

espora. 30.- Vista ecuatorial mostrando el collar. Pteris
Podophylla 31.- Vista proximal mostrando cicatriz. 32.-
Ornamentaci6n de la espora. 33.- Vista ecuatorial mostrando el

collar. Pteris pulchra 34.- Vista proximal mostrando cicatriz.
35.- Ornamentacidn de la espora. 36.- Vista ecuatorial mostrando
el collar. La linea negra que se encuentra al lado de las
fotomicrografias corresponde a 10 uz.

74 PHYTOLOGIA July 1997 volume 83(1):67-78

Pteris grandifolia L. 9 Km al sur de Palenque, Chiapas, R. Fernandez 2233 (ENCB).

Lamina II, Figuras 13 a 15.

Espora trilete, vista proximal triangular de 39.5(43.5)47.5 por 43.0(46.5)51.5
jim. Vista lateral o ecuatorial ovada de 38.0(39.0)40.5 por 45.5(47.0)50.0 um.
Exina de 7 jm de grosor. Nexina de 1 jum, sexina de 3 jm, tuberculada y perina de 3
jim de espesor, psilada, vista ecuatorial con collar de 12 [sm de grosor. Brazos de la
laesura de 28 a 32 um de largo por 2 a 3 yum de ancho.

Pteris longifolia L. Minatitlan, Municipio de Minatitlan, Colima, Lépez-Ferrari,
Espejo, & A. Flores 627 (ENCB). Lamina I, Figuras 16 a 18.

Espora trilete, vista proximal triangular de 42(47)51 por 40(45)50 um. Vista
lateral o ecuatorial circular de 41.5(44.5)47.5 tum de diametro. Exina de 6 um de

grosor. Nexina de 1 jim, sexina de | um y perina de 4 ym de espesor, vista proximal
rugulada y distal reticulada, en vista ecuatorial sin collar. Brazos de la laesura de 11 a

15 sm de largo por mas o menos | pm de ancho.

Pteris muricata Hook. Talquién, Municipio de Unién Juarez, Chiapas, E. Ventura &

E. Lopez 3315 (ENCB). Lamina Il, Figuras 19 a 21.

Espora trilete, vista proximal triangular de 33.0(35.3)37.2 por 35.5(37.3)39.0
im. Vista lateral o ecuatorial ovada de 28.0(29.7)31.3 tm de largo por
37.2(39.2)41.4 um de ancho. Exina de 3.3 tum de grosor. Nexina menor de 1 um,
sexina de 1 jm y perina de 1.3 tm de espesor, psilada, vista ecuatorial con collar de

11 pm de grosor. Brazos de la laesura de 10 a 13 pum de largo por 1.5 a 2.0 um de
ancho.

Pteris muricella Fée. Laguna Ocotal, Municipio Ocotsingo, Chiapas, D.E. Breedlove

33076 (ENCB). Lamina II, Figuras 22 a 24.

Espora trilete, vista proximal triangular de 28.7(30.4)32.0 por 33.0(34.5)36.3
fim. Vista lateral o ecuatorial ovada de 23.6(25.0)26.2 jm de largo por
34.6(37.0)39.0 um de ancho. Exina de 3.3 tum de grosor. Nexina menor de | um y
sexina de 1.5 jm de espesor, tuberculada, perina de 1 1m, vista ecuatorial con collar

de 10 tum de grosor. Brazos de la laesura de 8 a 12 um de largo por 1.0 a 1.5 tm de
ancho.

Arreguin-Sanchez, et al.: Morphology of Pteris spores 75

Pteris orizabae Mart. & Gal. Chipoca, cerca de Otongo, Municipio Tlalchinol,
Hidalgo, J. Rzedowski 32479 (ENCB). Lamina III, Figuras 25 a 27.

Espora trilete, vista proximal triangular de 27(29)36 por 31(38)36 um. Vista
lateral o ecuatorial ovada de 28(30)31 jum de largo por 28(35)43 tm de ancho. Exina
de 3.5 tm de grosor. Nexina de 1 im, sexina de | tm y perina de 1.5 tum de
espesor, rugulada, vista ecuatorial con collar de 18 sm de grosor. Brazos de la laesura
de 8 a 11 um de largo por 1.5 a 2.0 tm de ancho.

Pteris paucinervata Fée. Km 10 camino Omiltemi, Guerrero, F.G. Lorea 117]

(ENCB). Lamina II, Figuras 28 a 30.

Espora trilete, vista proximal triangular de 37.0(44.4)50.7 por 43.0(48.4)54.0
Him. Vista lateral o ecuatorial ovada de 41.4(42.3)43.0 [um de largo por
28.0(31.8)34.6 tim de ancho. Exina de 2.5 tum de grosor. Nexina menor de 1 um y
sexina de 1.7 [1m de espesor, escabrosa, parecen unirse las escabras en forma irregular
unas con otras, con collar ecuatorial de 9 im de grosor. Brazos de la laesura de 15 a
17 xm de largo por | pm de ancho.

Pteris podophylla Sw. Dos Lagos, Municipio de la Trinitaria, Chiapas, D.E.
Breedlove 56487 (ENCB). Lamina III, Figuras 31 a 33.

Espora trilete, vista proximal triangular de 31(33)38 por 33.0(36.3)40.6 pm.
Vista lateral o ecuatorial de 24.5(26.4)29.6 um de largo por 36.3(39.0)42.3 tum de
ancho. Exina de 2.5 tum de grosor. Nexina menor de | pm y sexina de 1.6 tm de
espesor, psilada, vista ecuatorial con collar de 9 um de grosor. Brazos de la laesura de
11 a 13 pm de largo por 1.0 a 1.5 um de ancho.

Pteris pulchra Schlecht. & Cham. Santiago Atzalaén, Municipio Atzalan, Veracruz, F.
Ventura 19494 (ENCB). Lamina III, Figuras 34 a 36.

Espora trilete, vista proximal triangular de 31.3(33.6)36.3 por 33.0(35.4)38.0
im. Vista lateral o ecuatorial ovada de 35.5(37.0)39.7 tm de largo por
22.8(24.7)27.0 um de ancho. Exina de 3.3 ym de grosor. Nexina menor de | um y
sexina de 1.0 jm de espesor, rugulada, perina de 1.5 tm, psilada, vista ecuatorial con
collar de 13 tum de grosor. Brazos de la laesura de 8.5 a 13.0 um de largo por 1.5 a
2.0 um de ancho.

76 PHYTOLOGIA July 1997 volume 83(1):67-78

Pteris pungens Willd. 7 Km al N de Santa Maria Chimalapa, Oaxaca, H. Hernandez
686 (ENCB). Lamina IV, Figuras 37 a 39.

Espora trilete, vista proximal triangular de 39.0(41.5)44.8 por 37.2(40.0)43.0
jim. Vista lateral o ecuatorial ovada de 37.0(39.6)42.3 pm de largo por
25.3(28.0)30.4 tm de ancho. Exina de 6 im de grosor. Nexina de 1 lim y sexina de
2.5 pum de espesor, tuberculada, perina de 2.5 tm de grosor, psilada, vista ecuatorial

con collar de 11 jm de grosor. Brazos de la laesura de 15 a 18 jm de largo por 2 um
de ancho.

Pteris quadriaurita Retzius. La Barranca, Municipio de Teocelo, Veracruz, F. Ventura
18217 (ENCB). Lamina IV, Figuras 40 a 42.

Espora trilete, vista proximal triangular de 52.4(54.3)55.8 por 56.6(58.6)61.0
yim. Vista lateral o ecuatorial ovada de 52.4(56.7)61.0 jum de largo por
30.4(32.4)33.8 pum de ancho. Exina de 8 tm de grosor. Nexina de 1 pm, sexina de
3.5 um de espesor, tuberculada, perina de 3.5 jum de grosor, psilada, vista ecuatorial
con collar de 11 jm de grosor. Brazos de la laesura de 17 a 20 tum de largo por 2.5 a
3.0 pm de ancho.

Con base a la morfologia de las esporas podemos separar las especies de Pteris en
los siguentes grupos tomando en consideracién la presencia 0 ausencia de collar
equatorial, ormamentacién y grosor del collar ecuatorial.

1.- Ausencia de collar ecuatorial; vista proximal rugulada, distal reticulada...............
Ky Shgdhpuleinistuara shi aciels sine se Oeiaagateinatces «to veseals si teen ac ae ere ee P. cretica; P. longifolia

L.- Presencia de collar ecuatorial.............0000..c2..00000e2e0scheleilonnae eee 2
2.- Exina tuberculaday 45 sieviesinss ssiass.s cre sieaseaie nnd os.cahsiortnns tients a an ate

Bs eidachibsicaien P. biaurita; P. grandifolia; P. muricella; P. pungens; P. quadriaurita

2.- Exina rugulada, psilada, o escabrosa...-:.......22::.-0«=+ses-eeesesseaseeemeeeeens 3
3.- Exina rugullada. .. .....52.:.00000e0c000sessseieseoasyaseen saint nea ieee ae 4

4.- Grosor del collar ecuatorial menor de 15 pm....... P. altissima; P. erosa

4.- Grosor del collar ecuatorial 18 Pm.................seceeeeeee noes P. orizabae

3.- Exina psilada 0 escabrosa. ......<..-0000ss«0-2<c0-00<0s0 0s sneer 5

3.- Exitia psilada. 722...5..002a8t P. muricata; P. podophylla; P. pulchra

5;- Exina escabrosasic 25; « <ie taiasxhe- -2. coq aiemepe ap ae P. paucinervata

CONCLUSIONES

Pteris presenta esporas triletes, generalmente con collar ecuatorial, vista proximal
triangular, y distal ovada, en algunas especies la vista proximal y distal con diferente

Arreguin-Sanchez, et al.: Morphology of Pteris spores q1

ornamentacién como P. cretica y P. longifolia; con exina rugulada como P. altissima,
P. erosa, y P. orizabae; exina psilada como P. podophylla, P. muricata, y P. pulchra;
escabrosa como P. paucinervata; y tuberculada como P. biaurita, P. grandifolia, y P.
muricella.

Autores como Tryon & Tryon (1982) dividen al género en cinco grupos: Grupo
Pteris deflexa donde se incluyen P. altissima, P. muricata, P. muricella, y P.
podophylla;, Grupo Pteris quadriaurita con P. biaurita, P. pungens, y P. quadriaurita;
Grupo Preris longifolia con P. longifolia; Grupo Prteris cretica con P. cretica; y grupo
Pteris haenkeana con P. grandifolia, P. orizabae, P. paucinervata, P. pulchra, P.
pungens, y P. erosa no son considerados en ninguno de estos grupos, citados por
Tryon & Tryon (1982).

Al comparar las divisiones taxonémicas con las observaciones palinolégicas
observamos que en el Grupo de Pteris deflexa, P. altissima presenta exina rugulada,
P. muricata y P. podophylla psilada y P. muricella tuberculada. Del grupo Preris
quadriaurita, P. biaurita, P. pungens, y P. quadriaurita presentan esporas muy
semejantes con omamentacién tuberculada. Los grupos Pteris cretica y Pteris
longifolia las especies que se estudiaron en este trabajo presentan esporas muy
semejantes con ausencia de collar ecuatorial, vista proximal rugulada y distal
reticulada, por lo que palinolégicamente no se justifica la separacién en dos grupos
taxonémicos diferentes y por Ultimo, el grupo de Pteris haenkeana presentan
omamentaci6n muy diversa, por ejemplo, P. grandifolia y P. pungens con
omamentaci6n tuberculada, P. orizabae y P. erosa rugulada, P. paucinervata
escabrosa, y P. pulchra psilada. Por lo anterior, se considera que la separacién
taxonémica no corresponde con la morfologia de las esporas.

Tryon & Tryon (1982) indican que una caracteristica de las esporas de Pteris es la
presencia de un collar equatorial, sin embargo, Tryon & Lugardon (1991) indican que
generalmente las esporas de Pteris presentan este collar, quizds este cambio de
apreciaci6n se deba a que nueve afios después se han estudiado mas especies de este
género y se ha observado que pueden existir especies en las cuales no este presente
este collar.

Mickel & Beital (1988) sefialan que Preris erosa y P. orizabae son especies muy
cercanas y a nivel palinolégico se encontré similitud entre ambos taxa, ambos con
ormamentaci6n rugulada, solo que en la primera la omamentacién de la exina mucho
mas conspicua que la segunda. También indican que varios ejemplares de P. pungens
son confundidos con P. paucinervata, sin embargo, estas dos especies se separan
facilmente por sus esporas, la primera con esporas con ornamentaci6n tuberculada y la
segunda con ornamentaci6n escabrosa.

BIBLIOGRAFIA

Arreguin-Sanchez, M.L., R. Fernandez-Nava, R. Palacios-Chavez, & D.L. Quiroz-
Garcia. 1996. Morfologia de las esporas de pteridofitas isosporeas del Estado de
Querétaro, México. Polibotanica 2:10-60.

78 PHY TOLOGIA July 1997 volume 83(1):67-78

Lugardon, B. 1963. Les spores des pteridacées de France. Pollen et Spores
5(2):325-336.

Lugardon, B. 1974. La struture fine de l’exospore et de la périspore des filicinées
isosporées. Pollen et Spores 16(2):161-226.

Mickel, J. & J.M. Beitel. 1988. Pteridophyte flora of Oaxaca, Mexico. Mem. N.Y.
Bot. Gard. 46:1-568.

Nayar, K.B. 1964. Palynology of modern pteridophytes. In: Advances in
Palynology. National Botanic Gardens, Lucknow, India. 6:102-141.

Pl4 Dalmau, J.M. 1961. Polen. Talleres Graficos Gerona. 526 pp.

Tryon, R.M. & A.F. Tryon. 1982. Ferns and Allied Plants with Special Reference to
Tropical America. Springer-Verlag, New York, New York. 857 pp.

Tryon, A.F. & B. Lugardon. 1990. Spores of the Pteridophyta. Springer-Verlag,
New York, New York. 648 pp.

Phytologia (July 1997) 83(1):80.

Issue

82(1)
82(2)
82(3)
82(4)
82(5)
82(6)

PUBLICATION DATES, VOLUME 82

Cover date

January 1997
February 1997
March 1997
April 1997
May 1997
June 1997

79

Publication date

30 September 1997
28 November 1997
19 December 1997
23 January 1998
30 March 1998

6 April 1998

Phytologia (July 1997) 83(1):79.

INDEX TO REVIEWERS, VOLUME 82

The editor expresses his most sincere appreciation to the following individuals.
These are persons who have reviewed papers that were submitted for publication in
volume 82 of Phytologia. Without the willingness and diligence of these reviewers,
the task of the editor would be much more difficult, and the quality of the papers
published would be lessened. To each of you, I offer my most sincere thanks.

Michael J. Warnock, Editor

Allain, L. Kiger, R.
Barneby, R.C. Kline, G.
Brown, L.E. Kral, R.
Carlquist, S. Lellinger, D.B.
Cochrane, T. Lipscomb, B.
Dehgan, B. Moore, D.
Delevoryas, T. Nesom, G.L.
Dutton, B.E. Riba, R.
Elsen, D. Stacey, L.
Evans, R.E. Strother, J.
Evans, R.J. Thomas, R.D.
Foti, T. Thorne, R.F.
Fox, II, W.E. Turner, B.L.
Ganders, F. Turner, G.
Gandhi, K.N. Ware, S.
Garciai\e)- Warnock, M.J.
Hammel, B.E. Wilken, D.
Holst, B. Williams, J.
Hyatt, P. Zanoni, T.A.
Jones, G.D.

Judd, W.

80

eae De ee

New York Botanical Garden

LOOMOL AMELIA

3 5185 00288 4383

a
~~
yeaa

pice
J

Information for Authors

iad
, oe
Articles from botanical systematics and ecology’, iacludinve iM
biographical sketches, critical reviews, and summaries of literature |
will be considered for publication in PHYTOLOGIA. Manuscripts may)
be submitted either on computer diskette, or as clean typescript. —
Diskettes will be returned to authors after action has been taken oH”
the manuscript. Diskettes may be 5.25” or 3:5” and may be winkten tie as
any IBM or Macintosh compatible format. Typescript manuscripts
should be single spaced and will be read into the computer using a-
scanner. The scanner will read standard type fonts but will not read
dot matrix print. Manuscripts submitted in dot matrix print cannot ©
be accepted. Use underscore (not italics) for scientific names.
- Language of manuscripts may be either English or Spanish. rae a

will be reduced to fit within limits of text pages. Therefore, figures 4
should be submitted with internal scales. Legends for. figures should —
be included in figures whenever possible. Each manuscript ‘should —
have an abstract and key word list. Specimen citations should be
consistent throughout the manuscript. Serial titles should be cited ©
with standard abbreviations: References cited only. as part of —
nomenclatural summaries should not appear in Literature. Cited:
Nomenclatural work should include one paragraph per bens 4
and must provide proper. (as defined by the current International ~ sr
Code of Botanical Ronee eee) citation of sources of Spiess: and 4
combinations. | “Rena ice oe

A
is

cs

Authors should arrange for two whickens in the pe ees” field’ to. i
review the manuscript before submission. Copies of reviews. should
be forwarded to the editor with the manuscript. Nae ey will not.
be published without review. 4

Cost of publication is currently $13 00 US per. sae des papillon
without reprints. Publication with: 100 reprints ‘is. Preece: for:
$18.00 US per page, 200 reprints for $21.50 US per page. Page
charges are due with manuscript and no paper will be published &
before payment is received in full. Reprints must be ordered. and ©
paid for in advance. Page charges will be determined on the basis of
a typeset page. Title page should include title, authors(s) name(s) My
and address(es). No extra charge is made for line drawings. provided
they conform to limitations of size as Beebe: for normal Texts 9

Halftones require an extra charge,o ge at 10096.
Enlargement or reductions cost an aia bad be page: Ty a ns : ae
JUN 2 1608

~NEW YORK
BOTANICAL GARD eN