Faculty of Science
ANTHER DEVELOPMENT IN THE STAMINATE FLOWER OF THE LODGEPOLE
PINE DWARF MISTLETOE, ARCEUTHOBIUM AMERICANUM
2013 |

KAITLYN CHIEMI BLEASDALE

B.Sc. Honours thesis

ANTHER DEVELOPMENT IN THE STAMINATE FLOWER OF THE
LODGEPOLE PINE DWARF MISTLETOE, ARCEUTHOBIUM AMERICANUM
by
KAITLYN CHIEMI BLEASDALE
A THESIS SUBMITTED IN PARTIAL FULFULLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF BACHELOR OF SCIENCE (HONS.)
in the
DEPARTMENT OF BIOLOGICAL SCIENCES
CELLULAR, MOLECULAR, and MICROBIAL BIOLOGY

We accept this thesis as conforming to the required standards:

Cynthia Ross Friedman (Ph.D.), Thesis Supervisor, Dept. Biological Sciences
Ron Smith (Ph.D.), Committee Member, Dept. Biological Sciences
Joanna Urban (M.Sc.), Committee Member, Dept. Biological Sciences

Dated this 1st day of May, 2013, in Kamloops, British Columbia, Canada

ABSTRACT

Dwarf mistletoes (genus Arceuthobium) are parasitic angiosperms that infect
commercially valuable conifer trees across North America, resulting in millions of
dollars of damage annually. The purpose of this study was to examine the
developmental processes taking place during anther development in the dwarf
mistletoe species, Arceuthobium americanum, and to describe the eventual threedimensional relationship between the anther and the columella (a region of sterile
tissue within the anther). Angiosperms typically have anther walls comprised of
several layers. However, the composition of the anther wall layers and aspects of
layer development varies from species to species. These processes have only been
superficially examined in the Arceuthobium genus. In order to obtain insight into
anther wall development in A. americanum, thin sections were cut, mounted and
stained on slides, and viewed with a light microscope. The data indicate that A.
americanum have three distinct anther wall layers including 1) the epidermis, 2) the
middle layer, and 3) the tapetum. Most angiosperms have a sub-epidermal layer
called the endothecium, but this work shows that A. americanum possesses an
“exothecium” similar to that of gymnosperms. The eventual three-dimensional
orientation of the anther with reference to the columella was described using a
combination of light and dissecting microscopy. Previously, it was thought that the
columella spanned the length of the anther to such an extent that the loculus (anther
chamber) was bisected; however, this study reveals that the columella does not span
the length of the anther in either direction, even though it does originate centrally.

Thesis Supervisor: Cynthia Ross Friedman, Ph.D.
ii

ACKNOWLEDGEMENTS
To the Natural Sciences and Engineering Research Council (NSERC) of Canada
(courtesy of a Discovery Grant awarded to Dr. Cynthia Ross Friedman), thank-you
for the financial assistance. Also, many thanks to TRU Undergraduate Research
Experience Award Program (UREAP) for the research scholarship. To Louis
Gosselin, the use of your microscope is greatly appreciated.
Cindy, I really appreciate the opportunity – I am forever grateful. Thank-you, also,
for your advice, patience, and support. The opportunity to do undergraduate
research has taught me a great deal and I’m glad that I was able to make the most of
my time here at TRU. Dr. Don, you said, “Do a project!” and I’m glad I did – thanks
for pushing me off the edge, it really made me believe in myself. Also, my gratitude
to committee members Dr. Ron Smith and Joanna Urban for taking time out of your
busy schedules to give me feedback.
Kurtis, thank-you for your support from start to finish – we are a great team! Fellow
research students, it would have been a lonely journey without you. Thank-you for
telling me I was crazy when I shared my doubts (whether I had good reason to be
doubtful or not) and all the practice presentations you watched. Our mutual
support and assistance has made us better than I ever thought we could be. And,
finally, to each and every friend and family member: you’ve all been so wonderfully
supportive, I could not have asked for anything more.

iii

TABLE OF CONTENTS

ABSTRACT

ii

ACKNOWLEDGMENTS

iii

TABLE OF CONTENTS

iv

LIST OF FIGURES

vi

LIST OF TABLES

viii

INTRODUCTION

1

DWARF MISTLETOES (GENUS ARCEUTHOBIUM)

1

OVERVIEW OF ANTHER DEVELOPMENT IN ANGIOSPERMS AND
ARCEUTHOBIUM SPP
3
THE COLUMELLA: A UNIQUE FEATURE OF THE ARCEUTHOBIUM
ANTHER
10
POLLEN DEVELOPMENT IN ANGIOSPERMS AND ARCEUTHOBIUM
SPP.
10
PURPOSE

MATERIALS AND METHODS

12

13

SITE DESCRIPTION AND SAMPLE COLLECTION

13

DISSECTION MICROSCOPY

14

LIGHT MICROSCOPY

15

Processing – Dehydration

15

Processing – Infiltration and Embedding

16

Sectioning

16

Staining and Viewing

17
iv

RESULTS

18

ANTHER DEVELOPMENT

18

MATURE ANTHER

20

DISCUSSION

27

ANTHER WALL DEVELOPMENT

27

MATURE ANTHER WALL

28

COLUMELLA

30

MATURE POLLEN

33

CONCLUSION

34

LITERATURE CITED

35

v

LIST OF FIGURES
Figure 1i. The terminal portion of the staminate shoot showing a lone mature
terminal bud and inferior lateral flower buds in opposite arrangement (image
adapted from Cohen 1968)
3
Figure 2i. Development of a typical angiosperm tetrasporangiate anther (i.e., having
four microsporangia). A. Very early in development anthers present with a
mass of homogeneous meristematic cells surrounded by the epidermis. B.
After a few periclinal divisions. C-E. Continued development leading to the
origin of the sporogenous tissue (image adapted from Maheshwari 1950) 4
Figure 3i. The four patterns of anther wall development (Davis 1966)

6

Figure 1. Dissecting micrograph of a fixed male flower. An open trimerous flower
with three sepals (s), each with a sessile anther (an) surrounding the tri-lobed
central cushion (arrow). Scale bar 0.5 mm
22
Figure 2. Dissecting micrograph of a fixed inflorescence with a single terminal floral
bud and paired lower lateral buds. Scale bar 2 mm
22
Figure 3. Light micrograph of a serial sectioned (CS) anther revealed a unilocular
anther. Scale bar 85 µm.
22
Figure 4. Light micrograph of an anther (LS). The connection between the base of
the anther and the perianth is two or three epidermal cells thick (arrows).
Scale bar 23 µm
22
Figure 5. Light micrograph of epidermal cells. Epidermal thickenings are visible at
the cell periphery (arrows). Scale bar 25 µm
22
Figure 6. Light micrograph (CS) of an anther with columella (arrow head) tissue
partially dividing the loculus. Also, evident are the tapetum (white arrow)
and the epidermis (ep). Scale bar 82 µm
24
Figure 7. Light micrograph (LS) of an anther with a partially penetrating columella
(arrow head) and another anther with a fully connecting columella (arrow).
Scale bar 160 µm
24
Figure 8. Dissecting micrograph of a dehisced anther with fully developed pollen
grains (pg) and the central columella (arrow). Scale bar 0.4 mm
24

vi

Figure 9. Diagrammatic representation of the eventual relationship between the
columella (c) and the anther (transverse section). The columella is partially
dividing the loculus with developing pollen grains (pg)
24
Figure 10. Light micrograph (CS) of maturing pollen grains in an anther with a
vacuolate epidermis (arrow), uninucleate and crushed middle layer (arrow
head), and an intact, uninucleate tapetal layer (t). Scale bar 40 µm
26
Figure 11. Light micrograph of young pollen grains (CS). The pollen aperture
points display a thinning exine and a thicker intine (arrows). The tapetum (t)
has begun programmed cell death and some lightly staining material (likely)
of tapetal origin can be seen in the loculus (arrow head). Scale bar 15 µm 26
Figure 12. Light micrograph of and anther (LS) with mature pollen grains. Both the
generative (gn) and vegetative nuclei (vn) are centrally located in the pollen
grain. The only remaining anther wall layer prior to dehiscence is the
epidermis (ep). Scale bar 18 µm
26

vii

LIST OF TABLES

Table 1. Sampling months, number of sample dates per month, and the number of
trees that were sampled for each sampling date
14

viii

INTRODUCTION

Dwarf Mistletoes (genus Arceuthobium)

Dwarf mistletoes (genus Arceuthobium, Santalaceae) are destructive parasites that are
of both economic and ecological importance to Canadian forests (Hawksworth and
Wiens 1996). The genus Arceuthobium was initially ascribed to the family Viscaceae
(Davis 1966), but that family has since lost its status, and its members have been
reassigned to the parasitic family Santalaceae (APG II 2003). The parasitic condition
in plants has arisen approximately eleven times, encompassing around 4,000 species
from 265 genera and twenty-two families, which roughly works out to represent 1%
of all angiosperms (Nickrent et al. 1998). The vast majority of parasitic angiosperms
do not elicit detectable negative effects upon their host; however, Arceuthobium spp.
do.
Roughly forty-four different species of dwarf mistletoes exist, dispersed as parasites
on members of either Pinaceae or Cupressaceae in Canada and the United States,
and almost all species of coniferous trees have an associated species-specific
Arceuthobium infection (Hawksworth and Wiens 1996; Jerome and Ford 2002). This
type of species-specific, parasite-host relationship is common to all domains of life.
In the genus Arceuthobium, for example, A. pusillum mainly infects Picea glauca (white
spruce) and Picea mariana (black spruce), while A. americanum primarily infects Pinus
contorta (lodgepole pine) and Pinus banksiana (jack pine) (Hawksworth and Wiens
1996).
Arceuthobium americanum, however, is capable of infecting up to fourteen different
conifers from the Pinaceae family, which is the most extensive range of infection of
any North American dwarf mistletoe species (Jerome and Ford 2002; Hawksworth
1

and Wiens 1996). This is problematic to foresters, as conifers infected with dwarf
mistletoes show an overall reduction in growth (Chhikara and Ross Friedman 2008;
Hawksworth and Wiens 1996), reproductive fitness, and wood quality, which
equates to extensive product loss and, by extension, monetary losses to the lumber
industry (Hawksworth and Wiens 1996). In 1982, for example, it was estimated that
dwarf mistletoe infections in Canada and the Western United States were
responsible for monetary losses upwards of $2.6 billion dollars, based on a steady
rate of infection (Drummond 1982).
Arceuthobium americanum is a small, dioecious, yellowish to olive green herb,
consisting of aerial shoots with whorled branching (Hawksworth and Wiens 1996),
connected to a network of cortical strands and radial sinkers that penetrate the
coniferous host (Calvin and Wilson 1996). The mistletoe invades the aboveground
parts of the host and is able to penetrate through the thin, chlorophyllous bark of
conifer branches, some as old as 60 years (Hawksworth 1954). Once infection
(seedling establishment) has occurred, shoots can take up to five years to emerge
from the bark. This shoot development is generally climate-dependent:
Arceuthobium infections in British Columbia produce shoots in the second and third
years after infection (Smith 1971), whereas infections in Alaska take three to six
years to develop shoots (Shaw and Loopstra 1991). The host usually shows signs of
swelling at the infection point (Hawksworth 1954), which can precede shoot
development by more than a year. Aerial shoot tips with developing inflorescences
consist of a single terminal floral bud (that is, the bud at the apex of the stem)
superior to two lateral buds (the floral buds found on the side of the stem), which
are opposite in arrangement (Figure 1i).

2

Figure 1i. The terminal
portion of the staminate
male) shoot showing a lone
mature terminal bud and
inferior lateral flower buds
in opposite arrangement
(image adapted from Cohen
1968).

Staminate (male) reproductive structures (flowers) consist of three sepals and no
petals: some older studies refer to the sepals as ‘tepals’ because it was previously
unclear as to whether these floral parts were developmentally sepals (i.e.,
comprising the outermost whorl) or petals developing from an inner region (Cohen
1968). Each sepal acts as a holding pad for a single sessile anther. Central to the
three sepals is a tri-lobed pad-like elevation of sterile tissue known as the central
cushion, which excretes extremely sugary nectar (Penfield et al. 1976). Dehiscence of
pollen from the anther is via a transverse slit in the anther’s epidermis, the anther’s
outermost layer (Gilbert and Punter 1990). Each male flower produces about 11,000
pollen grains per its three anthers (Penfield et al. 1976). In response to temperature
and humidity changes, each anther is capable of reversibly opening and closing
(Gilbert and Punter 1990).
Overview of Anther Development in Angiosperms and Arceuthobium spp.
In typical angiosperms (Maheshwari 1950), a very young anther consists of actively
dividing meristematic cells, which are surrounded by a single layer of epidermal
tissue (Figure 2i). Development of microsporangia (anther or pollen sacs) begins
3

with the differentiation of archesporial cells in the region below the epidermis (i.e.,
the hypodermal region, Figure 2iA). The archesporial cells divide by periclinal
division (divisions parallel to the surface, Figure 2iB) to give a subepidermal
primary sporogenous layer and a primary parietal layer. The former undergoes
periclinal and anticlinal divisions (divisions perpendicular to the surface) to form a
mass of sporogenous tissue within the microsporangium and tapetum. Once
divisions have ceased, the cells are referred to as microsporocytes. These
microsporocytes undergo meiosis and mitosis to form microspores (immature pollen
grains) and mature pollen grains, respectively.

Figure 2i. Development of a typical tetrasporangiate angiosperm anther (i.e., having
four microsporangia). A. Very early in development anthers present with a mass of
homogeneous meristematic cells surrounded by the epidermis. B. After a few
periclinal divisions. C-E. Continued development leading to the origin of the
sporogenous tissue. Legend: sp, sporogenous tissue; t, tapetum; m, middle layer;
end, endothecium; e, epidermis (image adapted from Maheshwari 1950).

4

Meanwhile, as the sporogenous cells are developing, cells of the primary parietal
layer divide periclinally (along with some anticlinal divisions) to form cell layers
found external to the microsporangium, which together comprise the anther wall
(Maheshwari 1950; Davis 1966). The layers that make up the four main constituents
of the mature angiosperm anther wall are (from outermost to innermost): the
epidermis, the endothecium (a layer with fibrous thickenings), the middle layer, and
the tapetum (Maheshwari 1950; Davis 1966; Dundas et al. 1982; Bedinger 1992).
Based on mitotic division patterns and the eventual differentiation of a secondary
parietal layer (if such a layer is develops), species of angiosperms can be categorized
into one of four patterns of anther wall development (Figure 3i) (Davis 1966). The
four types of anther wall development are basic, dicotyledonous,
monocotyledonous, and reduced.
Typically, angiosperms are tetrasporangiate (as in Figure 2i), becoming
bisporangiate at the time of dehiscence, owing to the breakdown of the cells in the
septum partitioning the two microsporangia of the same lobe. Dowding (1931)
determined that Arceuthobium spp., however, are actually uni- or monosporangiate
(i.e., developing only one microsporangium), although it remains unclear whether
or not the anther is monosporangiate from its inception (Dowding 1931) or is
initially bisporangiate and undergoes septum breakdown (Mauseth 1988).

5

Figure 3i. The four patterns of anther wall development in angiosperms (Davis
1966).
Most angiosperms develop an endothecial layer, which is commonly a single layer
of cells (Maheshwari 1950; Davis 1966; Vasil 1967; Dundas et al. 1982). In some
angiosperms, a bi- or even multilayered endothecium can arise via periclinal
6

divisions; however, a layered endothecium is only likely to occur in more primitive
angiosperm families (Tobe 1989). The cells that comprise a typical endothecium are
radially elongated, and fibrous bands of callose, which arise from the inner
tangential walls, run up and out, coming to an end near the outer wall of each cell
(Maheshwari 1950; Vasil 1967). These fibrous bands provide structure and support
to the anther and also function in anther dehiscence (Goldberg et al. 1993). Some
species do not have fibrous thickenings, and dehiscence takes place when cells at the
apex of the anther undergo programmed cell death (Maheshwari 1950).
In contrast to other angiosperms, neither A. oxycedri nor A. americanum are thought
to have an endothecium. Rather, the epidermis of these species’ anthers develops
fibrous wall thickenings, thus making the epidermis an ‘exothecium’ (Dowding
1931). The term ‘exothecium’ has been used to describe the epidermis alone, both
the epidermis and the endothecium, or the outermost layer of the anther wall
(Dowding 1931; Skene 1948; Bhatnagar and Moitra 1996). Although the term
exothecium has been used in a manner that suggests a broad range of meanings,
generally, an exothecium seems to be an all-encompassing term that describes the
active layer with fibrous thickenings in the absence of an endothecium, whereas the
endothecium is defined by its sub-epidermal location in addition to the presence of
fibrous thickenings (Skene 1948; Shukla et al. 1998).
Typically, the angiosperm middle layer only persists for a short time, as these cells
are unable to divide anticlinally and are eventually crushed (Davis 1966; Anjaneyulu
and Lakshminarayana 1989). Middle layers that do persist usually do so because
they have developed fibrous wall thickenings similar to the endothecial layer (Vasil
1966; Hardy et al. 2000). The number of middle layers ranges from zero to seven,
but commonly there are one to three layers (Maheshwari 1950). These middle layers
7

give structure and support to the anther and, if retained, can be involved in
dehiscence (Goldberg et al. 1993; Hardy et al 2000).
The tapetal layer of angiosperms is highly specialized and of considerable
developmental significance (Maheshwari 1950; Davis 1966; Scott et al. 2004). The
tapetum is the innermost anther wall layer, which supplies the nutrients and
necessary substances, such as macromolecules, e.g., lipids (Piffanelli et al. 1998), to
the reproductive cells in the anther chamber or loculus (Maheshwari 1950; Bedinger
1992; Goldberg et al. 1993). Previously, tapetal cells were thought to originate from
primary parietal cells (Maheshwari 1950), but some authors have reported
origination from more than one tissue type, such as the secondary parietal layer or
from connective tissue (Davis 1966; Anjaneyulu and Lakshminarayana 1989). In
most angiosperms, the tapetum is a single layer of cells with some known cases of
bi- or multilayer conditions (Davis 1966; Furness & Rudall 2001). Most often the
production of the tapetum (i.e., mitosis) is completed before the development of the
microsporocytes, but occasionally the tapetal cells undergo mitotic divisions during
the meiotic divisions of the encased reproductive cells.
Two main types of tapeta have been described, the secretory and the amoeboid
(Maheshwari 1950; Davis 1966). These two tapeta differ in their methods of
microspore nutrition; that is, all substances produced by the secretory tapetum reach
the developing pollen grains via the locular fluid (a medium infiltrating the space
between each developing pollen grain). In contrast, the locular fluid is greatly
reduced in volume or completely absent in amoeboid tapeta. Instead, the amoeboid
tapetal cell wall breaks down, similar to the secretory tapetum, but their protoplasts
fuse and a multinucleate plasmodium protrudes into the loculus (Furness and
Rudall 2001; Pacini et al 1985). Surrounding this protruding structure is the tapetal
8

membrane (Furness and Rudall 2001) or a culture sac (Pacini et al. 1985). In contrast
to the secretory type, the amoeboid type is a form of direct nutrition, as the
amoeboid tapetum directly surrounds the developing microsporocytes.
The most common tapetum is the secretory tapetum (Maheshwari 1950; Davis 1966;
Furness and Rudall 2001), which can also be referred to as glandular, parietal (Pacini
et al. 1985), or cellular (Greyson 1994). The amoeboid tapetum has also been
referred to as (peri) plamodial (Greyson 1994; Furness and Rudall 2001). A few
angiosperm families present with intermediate forms of tapeta, but these have not
been previously well characterized (Pacini et al. 1985); however, Furness and Rudall
(2001) introduced the term ‘invasive nonsynsytial’ as a possible description for the
intermediate type tapetum. The cells of a secretory tapetum act as nurse cells by
providing microsporocytes with nutrients such as polysaccharides and proteins as
well as saturated and unsaturated lipids (Maheshwari 1950; Davis 1966; Pacini et al.
1985; Clement et al. 1998; Piffanelli et al. 1998). These nutrients must pass through
the locular fluid, therefore indicating an indirect form of nutrition. Tapetal cells are
known to degenerate in order to free certain secretion products such as lipidic
components, which contribute to the formation of the pollen coat (Pacini 1990;
Piffanelli et al. 1998). They do so in a typical pattern, albeit via different mechanisms
and at different times (Pacini et al. 1985; Pacini and Franchi 1991).
Regardless of tapetal type, it is generally understood that the relationship between
the tapetum and the developing sporogenous tissue is of utmost importance (Davis
1966; Pacini et al. 1985; Bedinger 1992), which has been confirmed by recent
molecular genetic knockout experiments in Arabidopsis, where a number of
transcription factors were knocked out (Zhu et al. 2011). Bedinger (1992) also noted

9

that morphological irregularities (e.g., fewer or absent mitochondria) in tapetal cells
have been observed in male sterile species prior to pollen abortion.
The Columella: A Unique Feature of the Arceuthobium Anther
Thoday and Johnson (1930) and Dowding (1931) reported the presence of a central
sterile column of tissue surrounded by sporogenous tissue in A. pusillum and A.
americanum, respectively. Dowding (1931) called this sterile column a ‘columella’.
The columella has been considered a unique feature to the genus Arceuthobium
(Maheshwari 1950). In other angiosperms (e.g., Ricinus communis) a central septum
has been reported, but the term ‘columella’ and the specific tissue it represents
seems to be reserved for the genus Arceuthobium. The only clear anatomical and/or
morphological difference between the septum and columella seems to be that the
cells of the septum degenerate prior to dehiscence and also that the septum fully
partitions the loculus. Dowding (1931), however, suggests that the protruding edges
of the columella are evolutionary remnants of a septum that might have partitioned
an bisporangiate anther of some ancestor. The three-dimensional positioning of the
columella has been the subject of much disagreement for many years, beginning in
1931 when Dowding described the columella as exhibiting a considerable amount of
orientational variation. The description of the relationship between the columella in
the mature anther has yet to be discerned with modern microscopy.
Pollen Development in Angiosperms and Arceuthobium spp.
Meiosis of microsporocytes in the anthers gives rise to haploid microspores or
immature pollen grains (Maheshwari 1950). A time after microsporocyte meiosis, an
asymmetric, mitotic division results in two nuclei within the microspore, referred to
as the generative and vegetative nuclei; at this point the pollen grain is said to be
10

mature. Generalized illustrations by Bedinger (1992) depicting pollen development
in maize suggest that these two nuclei migrate from a central location to a peripheral
one. A plasma membrane with limited cell wall material typically circumscribes the
generative nucleus as its own cell, while the vegetative nucleus remains in the
confines of the microspore (and can thus be considered a separate cell, the vegetative
cell). The vegetative cell forms the pollen tube, and a further mitotic division of the
generative nucleus forms two sperm cells, each with its own plasma membrane and
limited wall. The latter division can occur before or after germination so that each
pollen grain is either released at the two-celled stage (one vegetative cell and one
generative cell) or at the three-celled stage (one vegetative cell and two sperm cells).
In most angiosperms, pollen grains are shed at the two-celled stage (Shukla et al.
1998). When the pollen grain germinates, the pollen tube will grow out from an
aperture point at an extremely rapid rate (Bedinger 1992; Taylor and Helper 1997)
and if only a single generative cell is present, it will divide in the tube (Bedinger
1992; Taylor and Helper 1997; Shukla et al. 1998). The stage at which the pollen
grains of Arceuthobium are shed is not clear.
Pollen grains develop complex walls throughout meiosis and mitosis. The highly
resistant outer wall of the pollen grain, the exine, forms around haploid cells, but
includes material derived from the tapetum (as mentioned). The exine of mature
pollen grains is composed largely of an electron dense material called sporopollenin;
its presence is necessary for most species to produce a viable pollen tube, as it is
remarkably resistant to degradation (Bedinger 1992; Taylor and Helper 1997). The
inner wall of the pollen grain is the intine. In the angiosperm Nelumbo, the intine
forms rapidly during the earliest mature pollen stage and the mature pollen grain
has three apertures, the points where the pollen wall has thinned (Kreunen and
Osborn 1999). Hawksworth and Wiens (1972) defined the approximate average size
11

of Arceuthobium’s pollen grain as 23 µm in diameter; however, the details concerning
the development of the pollen grain remain unclear.
Purpose
The objectives of this study are to examine the developmental processes that take
place during anther development, to describe the eventual three-dimensional
relationship between the columella and whole anther, and to examine the mature
pollen in the dwarf mistletoe species, A. americanum using modern light and
dissecting microscopy techniques.

12

MATERIALS AND METHODS

Site Description and Sample Collection

Staminate Arceuthobium americanum plants with floral buds were collected by
Kathryn Pernitsky from March 2010 and to April 2011. Sampling was done in an
area adjacent to Stake Lake, 30 km south of Kamloops, BC (50° 31’ latitude and 120°
28’ longitude). Thus, this site was chosen because of its high degree of A.
americanum infection. The scheduling of sample collection ranged from twice per
week to once every two weeks, depending on the likelihood of major developmental
events occurring (according to guidance by Ross Friedman, pers. comm.). Samples
were collected from the same haphazardly- selected trees (marked with flagging
tape and numbered) over the sampling period in order to reduce the extent of
genotypic variation, and all samples collected were taken from eye level. From each
tree a minimum of ten male samples (aerial shoots) were removed each time, with
the average number of flowers per shoot around twenty-one. Samples were stored
in Karnovsky’s fixative (Karnovsky 1965) at the site and moved to a 4°C refrigerator
once in the lab. These samples were stored for approximately two years in 10 mL
glass vials sealed with plastic caps and made available for developmental analysis in
the spring of 2012. From the available sample dates approximately twenty-one were
used with the majority being from July, August, and September; although samples
from February, March, April, and June were used as well (Table 1). Data were from
A. americanum samples from four different trees.

13

Table 1. Sampling months, number of sample dates per month, and the number of trees that
were sampled for A. americanum for each sampling date.
Month

Number of sample dates

*Samples available from more than one
tree?

February
March
April
June
July

1
4
2
2
4

Not indicated
Yes, all four trees
Not indicated
Not indicated
Some (2 of 4) dates indicate all four
trees were sampled

August

5

Some (2 of 5) dates indicate all four
trees were sampled

September

3

Not indicated, but more than one glass
vial with samples per date

*Note: for each date that included samples from more than one tree (or more than
one glass vial per date), five (minimum) floral buds were dissected away from the
main stem of each plant per date per tree (vial) and added to a single glass vial for
infiltration. Of those, ten randomly selected buds were embedded in the
micromould for sectioning.

Dissection Microscopy

Samples were chosen for dissection based on the state of the flower. That is, only
partially opened flowers with closed anthers and plants with large buds
(presumably about to open, i.e., approaching anthesis) were dissected. Under the
fume hood, Karnovsky’s fixative was removed from glass vials containing whole
samples (entire aerial shoots with floral buds and stems), and replaced with
deionized, distilled water (DDW). On the bench, the whole samples were washed
with DDW three times with a minimum of 15 minutes between each change. On
average, five whole samples per date were viewed and dissected using a
14

combination of razor blades, scalpels, and forceps under a Nikon SMZ 1500
dissecting microscope. The dissecting microscope was equipped with a Nikon
Coolpix 995 camera for photographing the samples and a Toshiba television for live
viewing. Digital images were modified with Windows Live Photo Gallery.
Drawings of dissected samples were also made from this microscope.

Light Microscopy

All steps were performed at room temperature unless otherwise noted.
Processing - Dehydration

In a fume hood, fixed samples of individual staminate flower buds were dissected
from the stem using forceps, razor blades, and scalpels in various combinations with
the aid of a Kyowa Optical (model SDZ-PL) dissecting microscope and placed back
into the Karnovsky’s fixative prior to embedding, approximately an hour. Between
five and ten floral buds were dissected from each sampling date provided and if
more than one tree was sampled per date, at least five floral buds per tree were
dissected away from the main stem. In the fume hood, the dissected samples were
washed a minimum of three times with deionized, distilled water for a minimum of
15 minutes each. Following the three DDW washes, the samples were moved to the
bench and serially dehydrated via ethanol washes. All samples were immersed in
30%, 50%, 70%, and 90% ethanol for a minimum of 20 minutes each, as well as three
changes of 100% denatured ethanol (90% anhydrous ethyl alcohol, 5% methyl
alcohol, 5% isopropyl alcohol).

15

Processing – Infiltration and Embedding

Infiltration began by replacing the discarded 100% ethanol from the final wash with
a 1:1 solution of LR White Resin and 100% ethanol. The 1:1 solution was allowed to
infiltrate the samples for a minimum of 12 hours on an Ames Aliquot Mixer,
although longer infiltration times (24 hours) were most successful. Next, the same
infiltration procedure was followed with a 3:1 solution of LR White and 100%
ethanol. Finally, the samples were infiltrated with three changes of 100% LR White,
each on the rotator for at least 12 hours - although allowing the samples to remain in
the resin for three days between each change was optimal. To polymerize and
embed, each sample was extracted from the 100% LR White and positioned at the
base of a 10 capsule micromould with the samples orientated so that the tip of the
bud was either facing down (oriented for cross sectioning) or on its side (oriented for
longitudinal sectioning). Each micromould was filled to the top with a mixture of
LR White and LR White Accelerator (10 mL: 1 drop, respectively) and allowed to
harden for minimum one hour. Samples tended to reorient with the addition of the
polymerization mixture to the micromould and in order to counter the change of
orientation, a small wooden probe was used to reorientate any notably skewed
samples.
Sectioning

In preparation for sectioning, each block with an embedded sample was removed
from each micromould with a razor blade. Using a razorblade, the blocks were then
trimmed to a trapezoidal blockface under a dissecting microscope prior to sectioning
with a microtome. Glass knives were made using a LKB Bromma 7800 Knifemaker,
equipped with photographic tape water troughs (“boats”), and sealed with wax.
16

The glass knives were used with a Sorvall MT2-B Ultra microtome to make thin
sections (about 3-5 µm in thickness) that were removed from the boats with a
flattened wooden probe and floated on single drops of DDW on Superfrost Plus
Gold Slides. In a fume hood, the slides were placed onto a GCA/Precision Scientific
slide warmer set to 55°C and covered with a petri dish lined with cloth and
moistened with xylenes (to aid the flattening/dewrinkling of each section).
Staining and Viewing

Sections were stained with crystal violet for observation using light microscopy.
The crystal violet solution was comprised of 10 g crystal violet dissolved in 100 mL
95% ethanol and 4 g ammonium oxalate dissolved in 400 mL of DDW. The crystal
violet (10 g/0.5 L) stain was added to the slides with a syringe equipped with a 0.45
µm nylon filter (about one drop per section) and allowed to sit for thirty seconds to
one minute, after which they were rinsed with DDW. Rinsed slides were then
placed back on the slide warmer to dry.
An Olympus CX2ILED compound light microscope was used to survey the stained
sections allowing the creation of preliminary drawings. On average, approximately
forty sections per sample were made with a minimum of five samples per sample
date, totaling approximately 200 sections per sample date (minimum). A Nikon
Eclipse E400 microscope equipped with a Nikon Coolpix P6000 camera was used to
photograph selected sections. Digital images were modified with Windows Live
Photo Gallery.

17

RESULTS

Anther Development
The perianth (non-reproductive segments) of the male flower of A. americanum is
typically comprised of three sepals, and no petals, with each sepal bearing a stalkless (sessile) anther (Figure 1). At the center of the flower is a tri-lobed central
cushion of tissue whose lobes alternate with the perianth segments. Shoot tips of the
developing inflorescences have a single terminal bud with two inferior lateral buds
arranged in opposite orientation (Figure 2). Taller stems likely have lateral buds
below those inferior to the terminal bud, which are decussate (perpendicular) to the
first set of paired lateral buds. Most male flowers of A .americanum have three
perianth segments (i.e., are trimerous); however, some terminal and lateral buds
observed had four perianth structures (and four anthers).
Serial cross-sections of immature floral buds (oblique sections of the anthers)
revealed unilocular anthers encased by three cell wall layers (Figure 3). The outer
most layer is the epidermis, the middle layer is the sub-epidermal layer, and the
innermost layer is the tapetum. These layers are generally all uniseriate. There is a
point at the base of the anther’s connection to the perianth where the epidermal
layer becomes two or three cells thick (Figure 4). Also, the epidermal layer develops
thickenings in the periphery of the epidermal cells near to the time of dehiscence
(Figure 5).
Serial sections of both developing and mature floral buds, along with mature anther
dissections, revealed that the anther of A. americanum is penetrated by a central
column of sterile tissue or columella (Figures 6, 7, 8, and 9). In many of the
developing floral bud cross sections, the anthers appeared to be oddly penetrated by
18

the columella (Figure 6). The extent to which the columella seemed to penetrate the
anther depended upon the minor degrees of rotation that the samples may have
undergone during polymerization (block formation). Longitudinal sections of
developing floral buds revealed a similar pattern of penetration into the anther by
the columella, although in these sections, the penetration occurred in the lateral
plane (Figure 7). In cases where serial cross sectioning of the developing and mature
anthers were achieved, these showed that the columella did not seem to span the
transverse axes of the anther in either direction; in addition, the columella did not
seem to directly contact the apex of the loculus in the vertical axis (opposite to where
the anther connects to the perianth at the base of the anther). Instead, a few cells,
likely tapetal cells, bridged the gap between the columella and the apex of the
loculus.
In the latter phases of development, the epidermal layer contains vacuoles, and
although crushed, the middle layer has visible, prominent nuclei (Figure 10). The
tapetal layer does not appear to have begun programmed cell death and is also
uninucleate. Tapetal cells in the later stages of development appeared irregular in
shape. Just prior to and during tapetal degeneration, light-staining extraneous
material, likely of tapetal origin, appears in the loculus. The darker staining, thicker
“spots” in the loculus are the spiny exines of pollen grains (as seen by serial
sectioning). Upon closer examination of the microspore cell walls, a thickened intine
below each of the three apertures as well as a thinned, spiny exine can be observed
(Figure 11). The intine and the exine appear to remain uniform in thickness at all
other locations on the pollen grain surface

19

Mature Anther
Dissection of the mature anther confirmed that the columella does not span the
horizontal length of the anther (Figure 8); however, the columella structure did seem
to run the length of the vertical axis (although examination of the serial sections
suggested cells of a different type may be bridging a small gap) to such an extent
that suggests a connection between the anther wall and the columella (represented
in Figure 9). The connection likely causes a central groove in the anther’s outer
surface.
The mature anther wall consists of only one layer of functional cells - the epidermal
layer, which has developed thickenings and is, therefore, more appropriately named
the exothecium (Figure 12). At this time (anther maturity), mature pollen grains are
present in the anther’s loculus, the tapetum has completely degenerated, and the
remnants of the middle layer appear as a thin covering on the inner side of the
epidermal (exothecial) layer (Figure 12). The mature pollen grains contain two
nuclei, the generative nucleus and the vegetative nucleus. No obvious plasma
membrane or wall circumscribed the generative nucleus, but such a tenuous
boundary might not be detectable by light microscopy. The generative nucleus is
round in shape and the vegetative nucleus is fairly elongate. Both nuclei are central
to the pollen grain and in relatively close proximity to one another. It is at this stage
that the pollen will be shed.

20

Figure 1. Dissecting micrograph of a fixed male flower. An open trimerous flower
with three sepals (s), each with a sessile anther (an) surrounding the tri-lobed
central cushion (arrow). Scale bar 0.5 mm.
Figure 2. Dissecting micrograph of a fixed inflorescence with a single terminal floral
bud and paired lower lateral buds. Scale bar 2 mm.
Figure 3. Light micrograph of a serial sectioned (CS) anther revealed a unilocular
anther. Scale bar 85 µm.
Figure 4. Light micrograph of an anther (LS). The connection between the base of
the anther and the perianth is two or three epidermal cells thick (arrows).
Scale bar 23 µm.
Figure 5. Light micrograph of epidermal cells near to the time of dehiscence.
Epidermal thickenings are visible at the cell periphery (arrows). Scale bar 25
µm.

21

22

Figure 6. Light micrograph (CS) of an anther with columella (arrow head) tissue
partially dividing the loculus. Also, evident are the tapetum (white arrow)
and the epidermis (ep). Scale bar 82 µm.
Figure 7. Light micrograph (LS) of an anther with a partially penetrating columella
(arrow head) and a fully connecting columella (arrow). Scale bar 160 µm.
Figure 8. Dissecting micrograph of a dehisced anther with fully developed pollen
grains (pg) and the central columella (arrow). Scale bar 0.4 mm.
Figure 9. Diagrammatic representation of the eventual relationship between the
columella (c) and the anther (longitudinal section). The columella is partially
dividing the loculus with developing pollen grains (pg).

23

24

Figure 10. Light micrograph (CS) of maturing pollen grains in an anther with a
vacuolate epidermis (arrow), uninucleate and crushed middle layer (arrow
head), and an intact, uninucleate tapetal layer (t). Scale bar 40 µm.
Figure 11. Light micrograph of young pollen grains (CS). The pollen aperture
points display a thinning exine and a thicker intine (arrows). The tapetum (t)
has begun programmed cell death and some lightly staining material (likely)
of tapetal origin can be seen in the loculus (arrow head). Scale bar 15 µm.
Figure 12. Light micrograph of and anther (LS) with mature pollen grains. Both the
generative (gn) and vegetative nuclei (vn) are centrally located in the pollen
grain. The only remaining anther wall layer prior to dehiscence is the
epidermis (ep). Scale bar 18 µm.

25

26

DISCUSSION

Anther Wall Development

The characteristic of the middle layer’s eventual crushed state and the tapetal layer’s
degeneration in A. americanum is not unique to the species; many angiosperms
anther walls undergo a similar process. Davis (1966) noted that the middle layers of
angiosperms lack the ability to divide anticlinally. This disability handicaps the
middle layer’s ability to adjust to the increasing size of the microsporangium as the
microsporogenous tissue divides within.
Typically, the tapetal layer of angiosperms is composed of a single cell layer, as in A.
americanum; however, in some species, the tapetum is comprised of two cell layers
resulting from periclinal divisions (Davis 1966). Serial sectioning of the A.
americanum anther revealed a uninucleate tapetum, an observation that is in
agreement with Cohen (1968), although other genera closely related to Arceuthobium
within the family Santalaceae (ascribed to the former family Viscaceae) have bi- or
tetranucleate tapetal cells (Davis 1966). Throughout development, but more
prominently seen in the later stages of development, the tapetal cells of A.
americanum exhibited odd cellular organization and some irregularities in cellular
shape. That is, the cells were not defined like those of the epidermis; rather, they
appeared more elongate and fluid, like sickle-cell erythrocytes, sometimes
overlapping just at the edges. This observation seemed independent from the plane
of sectioning. Tapetal disorganization seems to occur as the result of orbicule (small
sporopollenin bodies associated with secretory tapeta) detachment, secretion, and
support substance secretion (Shukla et al. 1998; Furness and Rudall 2001). Also, the
tapetal cells remain associated with the anther wall until their degeneration;
27

therefore, the tapetum of A. americanum seems to exhibit the characteristics of a
secretory tapetum.
Davis (1966) stated that members of the family Santalaceae follow a dicotyledonous
type of anther wall ontogeny, whereas the former family Viscaceae fits the
monocotyledonous type of anther wall development (Figure 3i). However, with the
absence of an endothecium neither, neither of these two descriptions of anther
development completely fits Arceuthobium. Cohen (1968) describes the developing
microsporangium as a modified form of dicotyledonous or reduced type. I believe
that the pattern of A. americanum’s anther wall ontogeny is better ascribed to a
modified dicotyledonous type, as Davis (1966) definitively stated that the reduced
type does not form a middle layer. However, an argument can be made for
ascribing A. americanum to a modified form of the reduced type of anther wall
ontogeny, as A. americanum does not seem to develop a secondary parietal layer, a
condition for the dicotyledonous type. Therefore, it may be necessary to create a
new category (type) of anther wall development to best describe Arceuthobium.
Mature Anther Wall
An unusual aspect of A. americanum’s anther wall is the development of thickenings
in the epidermal layer along with the absence of an endothecium. Normally, the
endothecial layer is present directly below the epidermis and functions in normal,
non-explosive dehiscence (Maheshwari 1950). Other angiosperms that lack an
endothecium, but have epidermal thickenings at maturity (that is, have an
exothecium) are those belonging to the family Ericaceae, specifically from the genera
Kalmia and Phyllodoce (Skene 1948; Davis 1966; Kron et al. 2002). The development
of thickenings in the epidermal cells of the microsporangium is also present in all

28

gymnosperms, except Ginkgo (Bhatnagar and Moitra 1996). Interestingly, Ginkgo
develops its thickenings in a sub-epidermal layer, similar to the typical angiosperm.
The epidermal layer of the sporangia in Leptosporangiate ferns has a specialized
spine of thickened cells (the annulus) intermixed with vacuolated cells filled with
water, comprising what is known as a “plant catapult” (Noblin et al. 2012). When
the cells between the annulus lose water to the environment, there is a decrease in
the cell volume. As a result, the tension created is focused on the aperture point and
the sporangia open very rapidly. At this stage, many of the spores remain in the
head of the plant catapult, and it is not until the head springs back that the spores
are sent flying. Because of some structural similarities between the fern sporangia
and the epidermal cells of the A. americanum anther (that is, vacuolated cells with
thickenings), there may be some mechanistic cross-over. Perhaps as the moisture
held in epidermal cells of A. americanum evaporate, the cell volume would decrease
and tension would build on the connection between the anther wall and columella.
Finally, when the tension becomes greater, the anther wall would burst open along a
central groove (transverse slit) that corresponds to the columella’s position beneath.
When the anther opens in this manner, some pollen would be sent into the air,
although the majority would remain in the anther to be dispersed via the wind or by
insects. Some angiosperms that do not develop an endothecium dehisce via the
dissolution of certain cells in the anther wall, which could cause further weakening
in the structural integrity of the anther wall; this could allow the anther burst open
with more ease (Maheshwari 1950). However, I did not note any evidence of anther
wall cellular dissolution in mature anthers with mature pollen grains.

29

Columella
A unique aspect of A. americanum’s anther is the sterile tissue called the columella,
which penetrates the loculus of each anther. The columella has been the subject of
much disagreement amongst researchers for nearly one hundred years, although
there have not been any recent studies (within the last twenty to thirty years) that
have attempted to characterize the nature of Arceuthobium’s columella with modern
microscopy. Previously, the columella of A. americanum was described as extending
the total width of the anther, i.e., along one of the transverse axes, in either one
direction or both directions (Dowding 1931) or as being a central column with a
toroid of sporogenous tissue surrounding it (Dowding 1931; Cohen 1968). The
confusion is understandable, as serial cross sections of the buds often generate
oblique sections through the anther. In this study, such sections that just began to
enter the anthers obliquely revealed unilocular anthers, each anther having a
relatively uniform oval to kidney shape that seemingly lacked a columella.
However, as the sectioning continued, columella tissue began to become apparent at
the base of the anther, almost as though it penetrates the anther from the connection
point of the anther to the perianth. Further sectioning into the anther revealed a
connection between the anther and the outer epidermal tissue, seeming to partition
the locule in two or give the anther a three-dimensional toroid-like structural design;
both observations are actually incorrect as continued sectioning revealed, once
again, a unilocular anther.
Longitudinal sectioning of the buds was also used in an attempt to further
understand the overall three-dimensional arrangement. Due to the trimerous nature
of the male flower, longitudinal sections usually only captured two anthers at a
time, and were difficult to interpret, as entrance into each of the anthers was never
30

completely symmetrical. Therefore, typical longitudinal sections would consist of
two anthers, one with a fairly symmetrical circular shape lacking columella tissue,
and another oval shaped anther with the loculus partitioned in two by columella
tissue. This sectional morphology is explained by partial entrance into one anther
(the circular anther), i.e., not deep enough into the anther to see evidence of the
columella. The anther being partitioned in two appears as such due to the depth of
the sectioning, i.e., in deep enough to note the presence of the columella partitioning
the anther.
After dissecting numerous anthers, and in conjunction with the light microscopy
data, it was concluded that the columella does not span the width (transverse axes)
of an A. americanum anther at maturity. Dowding (1931) claimed to see a
considerable amount of variation with regard to columella orientation in the mature
unilocular anther (although the columella positions she observed were only ever
parallel to the transverse plane or at right angles to the transverse plane), but in this
study, all dissected samples were relatively homogeneous (parallel to the transverse
plane). However, I do believe that Dowding (1931) only examined serial sections
and did not include whole dissection in her study; therefore, the variation she
observed could have been due to the two dimensional nature of her work. Even in
this study, I found it challenging to make three dimensional conclusions from my
two dimensional sections alone.
The conclusions drawn by Dowding (1931) and Cohen (1968) are not dissimilar to
what can be seen during serial sectioning. However, when serial sectioning of
varying orientations is completed in combination with whole specimen dissection,
as performed here, the description becomes clear; the eventual three-dimensional
shape of the mature anther can be described as being penetrated, from the base of
31

the anther, by the columella, which extends up the vertical axis, near to the total
height of the anther (likely making a full connection via an extended tapetum) and
only extends about half the transverse width of the anther in one direction. The
anther wall attaches to the columella, which results in a central groove in the outer
anther surface, contributing to a cushion-like appearance. Dowding (1931) noted
that the connections between the anther wall and columella were of tapetal origin.
Because of this connection, the columella may function in pollen dehiscence or as a
means of maintaining the structural integrity of the mature anther such that the
anther wall does not cave in at maturity.
The discrepancies between my observations and those of Dowding (1931) could
have been due to the age of the samples being dissected, as Dowding (1931) looked
at all ages, whereas in this study, dissected samples were in the latter stages of the
anthers’ development; therefore, it is possible that certain aspects of the columella
abort such that the eventual anther is quite different from the early anther. Mauseth
(1988) argues that the Arceuthobium loculus is initially bisporangiate or even
tetrasporangiate (and becomes unilocular at dehiscence due to the breakdown of the
dividing septum). However, I have only observed unilocular anthers in agreement
with Dowding (1931), Cohen (1968), and Takhtajan (2009).
The cellular origin of the columella, then, because a question worthy of future
pursuit. It is likely that some early repeated periclinal divisions of the primary
parietal layer reach into the loculus, forming the columella. Or perhaps, since the
tapetum may form from connective tissues (Davis 1966; Anjaneyulu and
Lakshminarayana 1989), the same is true for the columella, which does seem to
connect to the tapetum. More work is needed.

32

Whole specimens collected on April 16th had partially opened flowers (but fully
closed anthers), and upon dissection, the anthers burst open. The samples were
likely dehydrating as a result of the heat from the microscope lights; therefore, the
columella and the anther wall would have been simultaneously drying out, creating
more and more tension until finally the anther burst open sending a small amount of
pollen up into the air. Prior to the start of dehiscence, the anther of Ricinus communis
consists of a central septum attached to the stomium, i.e., the point at which rapture
occurs in a pollen sac to release the pollen (Bianchini and Pacini, 1996); this
arrangement is similar to the pre-dehiscence anther of A. americanum. The opening
of the R. communis anther begins with the separation of the stomium from the
septum. A similar process may also occur in A. americanum, with the columella
being considered homologous to the septum. From this stage, the anther of R.
communis opens as the stomium cells separate, but the pollen is not yet released.
Rather, the loculus walls first separate and turn inside-out, completely exposing the
pollen, which remains anchored via the pollenkit (adhesive material that surrounds
pollen grains). The pollen is then explosively ejected from the anther when the
loculus rapidly returns to its original position. The anther of A. americanum does not
seem to rapidly return to its starting position, to disperse pollen; however, it can
close in response to low temperatures and increased humidity, to protect and
conserve the pollen (Gilbert and Punter, 1990). This unique method of controlling
pollen dispersal restricts the exposure of A. americanum pollen to conditions that
favor dispersal such as warm, dry, windy days.
Mature Pollen
The mature pollen of A. americanum has three aperture points, two nuclei (a
generative and vegetative nucleus), and a spiny exine. No obvious newly-formed
33

cellular boundary surrounded the generative nucleus, but such a boundary
comprised of a plasma membrane and perhaps some cellulosic wall material is likely
present; transmission electron microscopy would probably reveal that the generative
nucleus is circumscribed in its own cell. Because the generative nucleus does not
seem to undergo a second mitotic division before the pollen is shed, each mature
pollen grain is probably released at the two-celled stage. This characteristic is
similar to that of most angiosperms (Shukla et al. 1998).
Conclusion
There are notable deviations regarding the anther of A. americanum when compared
with typical angiosperms: the endothecium never develops; the epidermis develops
fibrous thickenings, called the exothecium; and a central column of sterile tissue
remains present in the mature anther, called the columella. These deviations might
be explained by an overall evolutionary reduction in Arceuthobium. Similar to most
other angiosperms, A. americanum’s tapetum is likely the secretory type and pollen
grains are probably shed at the two-celled stage. This work contributes to the
growing body of information contributing to the understanding of A. americanum,
the genus Arceuthobium, and angiosperms as a whole.

34

LITERATURE CITED
Angiosperm Phylogeny Group. (2003). An update of the Angiosperm Phylogeny
Group classification for the orders and families of flowering plants: APG II.
Botanical Journal of the Linnean Society 141: 399-463.
Anjaneyulu, C., and Lakshminarayana, K. 1989. Microsporogenesis and the
development of male gametophyte in two species of Diospyros L. Taiwania, 34:
187-91.
Bedinger, P. 1992. The Remarkable Biology of Pollen. The Plant Cell 4: 879-887.
Bhatnagar, S.P., and Moitra, A. 1996. Gymnosperms. Daryaganj, New Delhi: New
Age International (P) Limited, Publishers.
Bianchini, M., and Pacini, E. 1996. Explosive Anther Dehiscence in Ricinus communis
L. Involves Cell Wall Modifications and Relative Humidity. International
Journal of Plant Sciences 157: 739-45.
Calvin, C.L., and Wilson, C.A. 1996. Endophytic system. In Dwarf Mistletoes:
Biology, Pathology, and Systematics. Edited by F.G. Hawksworth and D.
Wiens. Agriculture Handbook 709, USDA Forest Service, Washington, D.C.
113-122.
Chhikara, A., and Ross Friedman, C.M. 2008. The effects of male and female
Arceuthobium americanum (lodgepole pine dwarf mistletoe) infection on the
relative positioning of vascular bundles, starch distribution, and starch
content in Pinus contorta var. latifolia (lodgepole pine) needles. Botany 86:
539–543.
Clément, C., Laporte, P., Audran, J.C. 1998. The loculus content and tapetum during
pollen development in Lilium. Sexual Plant Reproduction 11: 94–106.
Cohen, L.I. 1968. Development of the staminate flower in the Dwarf Mistletoes,
Arceuthobium. American Journal of Botany 55: 187-193.
Davis, G.L. 1966. Systematic Embryology of the Angiosperms. New York: John Wiley
& Sons.
Dowding, E.S. 1931. Floral morphology of Arceuthobium americanum. Botanical
Gazette 91: 42-54.

35

Drummond, D.B. 1982. Timber loss estimates for the coniferous forests of the United
States due to dwarf mistletoes. USDA Forest Service, Fort Collins, Colorado.
Rep. 83-2.
Dundas, I.S., Saxena, K.B., Byth, D.E. 1982. Pollen mother cell and anther wall
development in a photoperiod-insensitive male-sterile mutant of pigeon pea
(Cajanus cajan (L.) Millsp.). Euphytica 31:371–375.
Furness, C.A., and Rudall, P.J. 2001. The tapetum in basal angiosperms: early
diversity. International Journal of Plant Sciences 162: 375-392.
Gilbert, J., and Punter, D. 1990. Release and dispersal of pollen from dwarf mistletoe
on jack pine in Manitoba in relation to microclimate. Canadian Journal of
Forest Research 20: 267-73.
Greyson, R.I. 1994. The Development of Flowers. New York: Oxford University
Press.
Goldberg, R.B., Beals, T.P., and Sanders, P.M. 1993. Anther development: basic
principles and practical applications. Plant Cell 5: 1217-1229.
Hardy, C.R., Stevenson, D.W., and Kiss, H.G. 2000. Development of gametophytes,
flower, and floral vasculature in Dichorisandra thyrsiflora (Commelinaceae).
American Journal of Botany 87: 1228-1239
Hawksworth, F.G. 1954. Observations on the age of Lodgepole Pine tissues
susceptible to infection by Arceuthobium americanum. Phytopathology 44: 522.
Hawksworth, F.G., and Wiens, D. 1972. Biology and Classification of Dwarf Mistletoes
(Arceuthobium). Agriculture Handbook 401, USDA Forest Service, Washington,
D.C.
Hawksworth, F.G., and Wiens, D. 1996. Dwarf Mistletoes: Biology, Pathology, and
Systematics. United States Department of Agriculture Forest Service.
Washington.
Jerome, C.A., and Ford, B.A. 2002. The discovery of three genetic races of the dwarf
mistletoe Arceuthobium americanum (Viscaceae) provides insight into the
evolution of parasitic angiosperms. Molecular Ecology 11: 387-405.
Karnovsky, M.J. 1965. A formaldehyde-glutaraldehyde fixative of high osmolarity
for use in electron microscopy. Journal of Cell Biology 27: 137A.
36

Kreunen, S.S., and Osborn, J.M. 1998. Pollen and anther development in Nelumbo
(Nelumbonaceae). American Journal of Botany 86: 1662-76.
Kron, K.A., Judd, W.S., Stevens, P.F., Crayn, D.M., Anderberg, A.A., Gadek, P.A.,
Quinn, C.J., and Luteyn, J.L. Phylogenetic Classification of Ericaceae:
Molecular and Morphological Evidence. The Botanical Review 68: 335-423.
Maheshwari, P. 1950. An introduction to the embryology of angiosperms. New
York: McGraw-Hill.
Mauseth, J.D. 1988. Plant Anatomy. Menlo Park, California: The
Benjamin/Cummings Publishing Company.
Noblin, X., Rojas, N.O., Westbrook, J., Llorens, C., Argentina, M., and Dumais, J. The
Fern Sporangium: A Unique Catapult. Science 335: 1322.
Nickrent, D.L., Garcia, M.A., Martin, M.P., and Mathiasen, R.L. 2004. A phylogeny
of all species of Arceuthobium (Viscaceae) using nuclear and chloroplast DNA
sequences. American Journal of Botany 91: 125-38.
Pacini, E., Franchi, G.G., and Hesse M. 1985. The tapetum: Its form, function, and
possible phylogeny in Embryophyta. Plant Systems and Evolution 149: 155–
185.
Pacini, E. 1990. Tapetum and microspore function. In: S. Blackmore and R.B. Knox
(eds.), Microspores: Evolution and Ontogeny, pp. 213-237. London: Academic
Press.
Pacini, E., and Franchi, G.G. 1991. Diversification and evolution of the tapetum. In: S.
Blackmore and S. Barnes (eds.), Pollen and Spores: Patterns of Diversification,
pp. 181-86. Verlag, Berlin: Springer.
Penfield, F.B., Stevens, R.E., and Hawksworth, F.G. 1976. Pollination ecology of three
Rocky Mountain dwarf mistletoes. Forest Science 22: 473-84.
Piffanelli, P., Ross, J.H.E., and Murphy D.J. 1998. Biogenesis and function of the
lipidic structures of pollen grains. Sexual Plant Reproduction 11: 65–80.
Scott, R.J., Spleman, M., and Dickinson, H.G. 2004. Stamen structure and function.
Plant Cell 16: S40-S60.

37

Shaw, C.G., III, and Loopstra, E.M. 1991. Development of dwarf mistletoe infections
on inoculated western hemlock trees in southeast Alaska. Northwest Science
65: 48-52.
Shukla, A.K., Vijayaraghavan, M.R., and Chaudhry, B. 1998. Biology of Pollen, (1st
edition). Darya Ganj, New Delhi: APH Publishing Corporation.
Skene, M. 1948. Biology of Flowering Plants. Darya Ganj, New Delhi: Discovery
Publishing House.
Smith, R.B. 1971. Development of dwarf mistletoe (Arceuthobium) infections on
western hemlock, shore pine, and western larch. Canadian Journal of Forest
Research 1: 35-42.
Takhtajan, A. 2009. Flowering Plants, (2nd edition). Netherlands: Springer.
Taylor L.P., and Hepler P.K. 1997. Pollen Germination and Tube Growth. Annual
Review of Plant Physiology & Plant Molecular Biology 48: 461.
Thoday, D., and Johnson, E.T. 1930. On Arceuthobium pusillum, Peck. II. Flowers and
fruit. Annals of Botany 44: 813-24.
Tobe, H. 1989. The embryology of angiosperms: Its broad application to the
systematic and evolutionary study. The Botanical Magazine of Tokyo 102:
351–367.
Vasil, I.K. 1967. Physiology and Cytology of Anther Development. Biological
Reviews 42: 327–366.
Zhu, J., Lou, Y., Xu, X., and Yang, Z-N. 2011. A Genetic Pathway for Tapetum
Development and Function in Arabidopsis. Journal of Integrative Plant
Biology 53: 892–900.

38