全 文 ：广 西 植 物 Guihaia 30(5)j 584— 593 2010年 9月
獐牙菜小孢子发生及雄配子体发育
黄衡宇，龙 华，易婷婷，李 鹂
(湖南吉首大学 植物资源保护与利用湖南省高校重点实验室，湖南 吉首 416000)
摘 要：用石蜡切片法对獐牙菜小孢子发生及雄配子体发育过程进行首次观察研究 。主要结果如下：花药四
室，药壁发育为基本型；绒毡层异型起源，属于腺质型绒毡层，药室内具有的退化绒毡层核是早期该层细胞有
丝分裂凸入药室中央并原位退化形成的；中层细胞 2层 ；药室内壁同表皮同时宿存 ，细胞柱状伸长，纤维状加
厚。小孢子母细胞减数分裂为同时型 ，四分体排列方式主要为四面体形 ，少数为左右对称形和十字交叉形；成
熟花粉多为 2一细胞类型，偶见 3一细胞型、，具三萌发孔。
关键词：獐牙菜；小孢子；雄配子体
中图分类号 ：Q949，Q941 文献标识码：A 文章编号：1000-3142(2010)05—0584一10
Microsporogenesis and the development of male
gametophyte in Swertia bimaculata
HUANG Heng-Yu，LONG Hua，Ⅵ Ting-Ting，LI Li
(Hunan Province University Key Laboratory for Conservation and Utilization
of Plant Resources，Jishou University，Jishou 416000，Claim )
Abstract：Mierosporogenesis and the development of male gametophyte in Swertia bimaculata were studied by the
method of paraffin section for the first time in the present paper．The main results can be concluded as folows．An-
thers are tetrasporangiate，the development of anther walls conforms to the basic type and comprises of epidermis，~_XI-
dothecium，two middle layers and tapetum at mature stage．The tapetum has dual origin and belongs tO the Glandular
type The degenerating tapetum nuclei in the middle of anther locules are from the tapetum cells，which undergo mi—
tosis，then intrude into the anther locules and degenerate in situ at the early stage．Two middle layers are ephemeral
endothecium and epidermis persists and develops tO become fibrous-thickening．The eytokinesis of the microspore
mother cel in meiosis is of the simultaneous type．Most of the microspore tetrads are tetrahedral and there are stil a
few other types，such as isobilateral，dilatera1．Polen grain is mainly 2-celled type when shed，occasionaly 3-ceUed
type，and it has three apertures．
Key words：Sztertia bimaculata；raicrospore；male gametophyte
Swer~ia bimaculata，also named Da KU-Cao，Hei
Jie-Ku-Cao，Zou Dan-Cao in Chinese，iS an annum
herbage that belongs to Ser．Macultae T．N．Ho et S．
W．Liu of sect．Ophelia of Szvertia(Gentianaceae)．It
distributes mainly in Hunan，Yunnan，Guizhou，Tibet，
Hubei and Sichuan(Delectis Florae Reipublicae Popu一
1aris Sinicae，Agendae Academiae Sinicae Edita，1988)．
Most of species in Swertia are used as traditional med—
icine in China，Japan，India，and Nepal with a long his-
tory，for they are bitter and cold in property，and have
the effect of heat-clearing and detoxifying，cholagogic
and strengtheming，kiling insect．The whole plant can
Accepted date：2009-02—25 Reeived date：2010-01-27
Fouudafion item：Supported by the National Natural Science Foundation of China(3O360OO9)
Biography：HUANG Heng~Yu(1970一)，Jiangsu Wuxi，Male，Ph．D．，Associate professor，Major in Plant Developmental Biology，(E-mail)hhyhhy
@ 163．corn。
通讯作者(Author for correspondence，E-mail：lilyisu@126．com)
5期 黄衡宇等：獐牙菜小孢子发生及雄配子体发育 585
be used as medicine in curing digestive system disea8es
such as hepatits，choleeystitis，gastropathy and so on,
In recent years，it’s found that there are soree species
of this genus showing high bioactivity in anti-virus，de—
creasing blood glucose and relieving gastric ulcer，cu—
ring hepatitis and alopecia,thus，more and more people
pay attention to these species in Gentianaceae for the
great medica1 value they have(Chen et a1．，1998)．S．
bimaculata is widely used in curing jaundice hepatitis，
pneumonia，tonsilitis，gynecologic inflammation in ci—
vilian of Hunan．However，the wild resources decrease
greatly owing to coyoting and habitat destruction．
There is few research of basal biology in S．bimacula—
ta，excepting its seed germination(Long & Huang，
2008)，and the study on the aspect of generative biolo—
gY is blank．In order to protect the wild resources and
achieve the sustainable use of resources，we give the 1-e—
port on the microsporogenesis and the development of
male gametophyte of S．bimaculata and accumulate
embryological data of Swertia in order to ofer e~dence
for systematic taxonomy and protectiong biology． Fur—
thermore，it will provide basic data that is beneficia1 for
introduction and artificial cultivation of this genus．
1 Materials and Methods
Material for the present study was colected from
Wanba~shan Forestry Centre of Longshan County of
Xiangxi Tujia and Miao Autonomous，Hunan Prov-
ince，China(109。38 E，29。46 N，Alt：1396．2± 11 m)．
The voucher(Long hua177)is deposited in the Herbari—
um of Biological resources and Environmental Science
Department of Jishou University and identified by
Zhang Dai—gui of Jishou University．
Flower buds and flowers at different development
stages were colected and fixed in the modified FAA
(5O 9／6 alcohol：glacial acetic acid ：formaldehyde=89
：6 ：5)． These materials were washed with water
and then stained in Ehrfich’s hematoxyfin together．
After that，they were dehydrated in ten grades of aleo—
hol，transparentized with five grades of xylene and em—
bedded in paraffin． These materials were serially see—
tioned by microtome(Microm)at the thickness of 5—
8~rn． Sections were mounted on slides in Neutra1 bal—
sam，observed and photographed with microscope(Lei—
ca DM2000)．
2 Observation and Results
s．bimaculata is a pentamerous flower(Plate I：
1)．In the early phase of budding，pefianth and stami—
hate primordium were formed from flower primordium
in sequence．The basal part and terminal part of stami—
hate primordium are differentiated to form filaments
and anthers separately． Each anther has four micro—
sporangium，and at the early stage of anthers，the sur—
face of anther is composed of a layer of flat epidermal
cells，inside whieh there are a mass of meristematic
cells wi th sim lar morpha．Thereafter，the anther be—
come quadripartite ，for the tissue of four patches dif—
ferentiate faster．ArchesporiaI ceils differentiate below
the epidermis in each patch，having larger size and den—
ser cytoplasm(Plate I：2)．These cells divide periclinal—
ly to produce primary parietal cells outward and prima—
ry sporogenous cells inward(Plate I：3)．
2．1 The formation and differentiation of the anther wal
Two layers(outer layer and inner layer)were dif—
ferentiated in the periclinal division of primary parietal
cells(Plate I：4)，forming．three layers together with ep—
idermis(Plate I：5)．W hereafter，through the division of
the two layers，anther wall has fully differentiated into
five layers(Plate I：6，7，8)when secondary sporogenous
cells appear(Plate I：9)：epidermis，endothecium，two
middle layers and one tapeturn． It indicates that both
of the outer and inner layers derive from primary parle—
tal ceils，and through further division they differentiate
into endothecium and middle layer，and middle layer
and tapetum respectively．The middle layer is make up
of the cells from the outer and inner layers．According
to Davis(1966)’s compartmentalization．the develop—
ment of anther wall conforms to the basic type
．
Tapetum is the innermost layer in anther wall，and
it is ma inly formed from the inner primary parietal cells
and partly from the connective cells．mapetum ceils are
form ed at phase of early sporogenous cels，during the
process of microspore mother cell development，its cels
586 广 西 植 物 3O卷
■
2
Plate l 1．Transection of a young anther，showing five anthers×200；2．Male archesporium(十)×1000；3．Microspore archesporial cels divide per—
iclinaly to forln primary sporogenous cels(PS)and primary parietal ceUs(PP)X 1000；4．Primary parietal cel dividing(十)x 1000；5．0uter and the
inner layers from primary parietal cel dividingX 1000；6．0uter layer of the primary Wall dividirig(十)X 1000；7．Inner Iayer of the primary wal dirid—
ing(t)×1000；8．Outer andinnerlayer ofthe primarywal divi~ng(十)all~1000；9．Fivelayers antherwal duringthe periodof secondary spomgi-
nous cell×1000；10．Tapetum cels during microspore mother cells(T)×1000；11．“Trabeculae”and“Placentoid”from the tapetumX400．ANT,An—
tipodal cells；EP．Epidermis；EN．Endothelium；M Middle 1ayer；PL．“Placentoid”originating from the tapetum cells；PP．Primary parietal cell；IS．
P mary sporogenous cell；SS．Secondary sporogenous cel；T．Tapetum ．
and cell nuclei augment，protoplasm manifold gradually
and undergo endogenetic mitosis and ami tosis．The ta—
petal cells become the most complete development，fea
tured the glandular cell，at phase of microsporocytes
M eanwhile，the size of tapetal cells are simi lar to mi —
crosporocytes but are much larger than those of other
layers and present close squareness(Plate I：10)．In the
course of development，some tapetal cells undergo peri—
clinal division to produce some cells protruding in the
anther chamber，which form so-called“quasi：placenta”
or“cross-grid”(Plate I：11；Plate II：12，13)．It is ob—
served that the majorities of tapetal cells are uninude—
ate and the minorities of tapetal cells are binucleate or
muhinucleate。Tapetal cells start tO separate from each
5期 黄衡宇等：獐牙菜小孢子发生及雄配子体发育 587
other at the end phase of microsporocytes(Plate II：
1 4)，degenerate obviously and disaggregate intensively
at their original site in the time of tetrads．The degen—
erated nuclei of tapetal ceils in the anther chamber are
formed by the degenerated layer cells which enter it at
early time(Plate II：1 5)but not formed by circumfer—
ence tapetal ceils which enter it after degeneration．It is
not found that amalgamation occurs and periplasmodi—
um appear when tapetal cels disaggregate from large
numbers of sections．The tapeta1 cells are single til
disappearing．At intermediate phase of uninucleate mi—
crospore，the tapetal cells disappear completely(Plate
II：16)．Therefore，the tapetum is similar to the Glan—
dular type．
The middle layers accomplishes diferentiation
during secondary sporogenous ceils，it included two
layers of cells with smaller size，bigger nucleus and
shaped in flat rectangle(Plate I：9)．W hen the meio—
phase I of mi crosporocytes is over，the middle layers
cells degenerate quickly and left one layer only(Plate
II：14)，then，it gets degenerated completely at the early
stage of the formation of 2-celed pollen grain．
The endothecium is form ed at the stage of second—
ary sporogenous ceils while cells arrange close and
present strip shape(Plate I：9)．The endothecial calls
are most developed when microsporocytes are formed
(Hate I：10)．After that，the endothecium does not de—
velop ahy more and is pulled as flat and long with the
expansion of anther(Plate II：14，15)． At the early
stage of the form ation of 2一celled polen grain，the en—
dothecial cells nearby the connective tissue become vac—
uolization to present rectangle shape while there is only
sere vestige left outside til1 shedding(Plate II：17)．
The epidermis is formed the earliest and only in—
dudes single-layer cels that arrange closely．W hat’s
more，the exterior of epidermal ceils often possess
horny layer which plays a role in protection． Before
long，the epidermal cells augment，the radial wal and
tanlgential wall elongate to present in the shape of
quadrangle or rectangle(Plate I：10)．At time of tet—
rads，the epidermal cells reach the highest level of de—
velopment and become fibrous_thickening at the inside
tangentia1 wall(Plate II：15)．At shedding，the connec—
tive tissue between the two anthers chambers at the
sanle side disappears，resulting in the connectivity of
two anther chambers(Plate II：18)．It is observed that
the epiderm al cells in the joint of two anther cham bers
are not thickened and become the focus of force which
is the place where anther wall dehisce and the mature
pollens emit．
2．2 M icrosporogenesis
The primary sporogenous cells derived from ar—
chesporial cells undergo mitosis resulting in secondary
sporogenous cells．Comparing with those surrounding
wal cells，the secondary sporogenous ceils have larger
size，higher nuclear-cytoplasmic ratio,dense cytoplasm，
and deeper color．Moreover，they arrange closely，pres—
ent polygon but don’t have obvious vacuoles(Plate II：
9)． The polygonal secondary sporogenous cels turn
into circular microspore mother cels with development
(Plate II：19，20)．
Microsporogenesis of S．bimaculata is derived
from the meiotic divisions of mi crosporocytes like other
angiosperms．In the process of meiosis，the structure of
chromosome has a series of variations as other plants
and the meiosis is not synchronized．The process is de—
scribed in detail as follows：Prophase I(Plate II：20)：
According to numerous observations，prophase I can be
observed in different period of materials，indicating this
period lasts longer time．
Metaphase I(Plate II：21)：The dumpy chromo—
somes are arranged on the equatorial plane in the center
of mi crosporocytes with promi nent spindle fibers after
the prophase I is over．
Anaphase I(Plate II：22)：Chromosomes are sepa—
rated from each other and move to the two poles．
Spindle fibers are observed connecting wi th the two
groups of chromosomes．
Telophase I，prophase II(Plate II-23)：The nude—
ar membrane appears again and new nucleus is form ed
gradually after the two groups of chromosomes arrive
to the two poles．But rio cell wall is formed and a bi—
nucleate cell appears without the period of dyad．When
the meiophase I is over，cells can not divide for a while
and chromosomes do not disappear completely．But in
a few mi nutes，those ceils keep on dividing and enter
588 广 西 植 物 3O卷
PlateI 12，13．Tapetum cels protrudingthe anther chamberinthetimeofmicrosporemother celsX400．14．Ttapetum ceilsinthe stageofmicro—
spore miosisX1000；15．Degenerating of the tapetum at orl‘ginal sites，noting the remaining nuclei of the tapetum in the mi ddle of the anther chamber
×400；16．Anther walI in the stage of uninucleate microsporeX400；17．Anther wall during the period of pollinating×200；18．Anther Wall in the
polinating periodX 100；19，20．Microspore mother celIs×1000；21．Microsporocyte at metaphase I of meiosisX 1000；22．Microsporocyte at anaphase
I of meiosisX 1000；23．IVlicrosporocyte at telophase I of meiosisX 1000；2．4．№crosporocyte at metaphase I of meiosisX 1000．PL “Placentoidorigi—
nating from the tapetum cells； Tapetum ．
into prophase II．In the short period，the nuclear mem—
brane iS obviOUS and some thick chromosomes twist to
a m ass．
Metaphase II(Plate II：24)：Nuclear membrane
disappears；Stumpy chromosomes are arranged in the
center of cells；spindle fibers are obvious in the center
of cells but not obvious in the pole of cells．
Anaphase II(Plate III：25)：Each group of chro—
mosomes in the two daughter nuclei separates from
each other and moves to the two poles．Spindle fibers
are observed connecting with each two groups Df chro—
mosomes，presenting four umbrellas in shape．
Telophase II(Plate III：26)：Chromosomes disap—
pear gradually after they get to the tWO poles． Fur—
5期 黄衡宇等：獐牙菜小孢子发生及雄配子体发育 589
thermore，nucleolus and nuclear membrane appear；
four new nuclei circular are formed；these nuclei are
separated from each other by the phragmoplast which
appears among them and produce the cell walls，thus
four new cells are formed．
The cytokinesis of the microspore mother cell in
meiosis is of the simultaneous type according to plates
of the meiosis process． Those tetrads are surrounded
together by the callose．W hat s more．there is also cal—
lose among the microspores in the same tetrad．In that
case,microsporocytes，tetrads or microspores are exis-
ted lonely． The functions of the callose surrounding
the tetrads are similar to a“molecular sieve”，which al—
low a great deal of nourishment pass while prevent
some macromolecular substances pass，controlling the
communications of substances between cells．Most of
microspore tetrads are tetrahedral(Plate III：27)and
smal part of microspore tetrads are dilateral(Plate III：
28)or isobilateral(Plate III：29)．
2．3 M ale gametophyte
At the late phase of rnicrospore tetrads，cells sepa—
rate from each other and the expansion of tetrads made
the conlrnon callose thinner．Finally。the callosal dis—
solved；these four microspores separated respectively
and were released into anther chamber that permeated
with secretion of tapetum．The microspores have syn—
thesized their own cytoderm when they are in coIilrnon
callose．The shape of newly released microspores re—
mains，and the interface and the free-face of it are plane
and spherical respectively(Plate III：30)．This period
lasts for longer time．Since then，the microspores grad—
ually grow into round-shaping，with a centering nucle—
us，dense protoplasm，thin cytoderm ，and an unclear
germi nal aperture． W ith the development of micro—
spore，the wall of which is becoming visible and germi —
nal aperture is also becoming obvious(Plate Ill：31)．
Like most angiosperm s，the microspore nucleus of
S．bimaculata will experience a series of movements．
The newly released mi erospore has a central nucleus
and dense protoplasm．After a period of development，
many smal vacuoles appear in the protoplasm (Plate
III：32)，they gather to become a large vacuole that lo—
cates in the center of the cel，and protoplasm then dis—
tributes in circumambience of cytoderm，and nuc1eus
moves randomly from center to one side of the waII
(Plate HI：33)．Before long，mi erospore enters mi tosis
prophase near cytoderm．There are mainly two direc—
tions of mitosis：one is centripetal division(Plate III：
34)，the other is along mural division(Plate III：35)．
Furtherm ore，the division of microspores in the same
chamber is not simultaneous． The microspore divides
into two nuclei with unequal size．Then the two nuclei
undergo cytokinesis to produce a small generative cell
which only has a little protoplasm and a large vegeta—
tive cell which contains large vacuole and most proto—
plasm of form er mi crospore．This may be named as the
early phase of two-celled pollen．
After a period of development，the combination of
pollen tone up，protoplasm continuously increased，the
large vacuole disappears，while nucleus fills out and
moves from fringe to center with protoplasm increas—
ing．Here,it can be seen that vegetative cell and gener—
ative cel are seperated by visible cytoderm (Plate III：
36)．Anon,one side of generative cel disengages from
polen wall and extrudes internaly．Finally，generative
cell is divorced．In this course，the walI of generative
cell dissolves gradually，and its nucleus augments
ceaselessly．At last，generative cell is free in the proto—
plasm of vegetative cel1． Meanwhile，generative cel
shows sphericity and its wall disappears completely
(Plate III：37)．Before long，generative nucleus lay a—
board vegetative nucleus． Thus，the development of
two-celled pollen has completed(Plate III：38)． The
generative cel divides to produce two sperms in the
pollen tube． However，it’s also observed that several
generative nuclei of mature polens divides to form
three-celled pollen before shedding(Plate III：39)．
For S．bimaculata，pollen grains germinated before
shedding．Most of pollen grains are uniporous aper-
ture，but it can also be observed biporose or triporate
aperture(Plate III：40)．However，only one pollen tube
can grow up，others stop growing in the mi dway．
2．4 M ale abortion
A 1ot of male abortions are observed from many
sections．Abortion is undertaken in some of those mi—
crospores when staminate primordium is just form ed
59O 广 西 植 物 3O卷
{
37
Plate m 25．Microsporocyte at anaphaseⅡof meiosisX 1000；26．Microsporocyte at telophase 1I of meiosisX 1000；27．Tetrahedral microspore tet—
rads×1000；28．IsobilaterO1 microsporetetrads× 1000；29．DilaterO1microsporetetrads×1000；3O．Microspore hadiUStfotrnedfrom microsporetet—
radsX 1000；31．NeMy released microspore with three aperturesX 1000；32．Some small vacuoles appear，entering the middie of mi crosporeX 1000；
33．Vacuolate period of uninucleate microspo re× 1000；34．Unirillcle-ate microspore WaS radially dividing× 1000：35．Uninudeate microspore was
breadthwise dividing×1000；36．The early 2-celled pollen，showing ob~ous cel wo11×1000；37．Generative cell aggrandizing and moving central×
】000；38．Mature 2-celled polen and one aperture bourgeoning(十)×1000；39．Mature 3-celledpolen and one aperture bourgeoning(十)×1000；40．
Three apertures bourgeoning(+)×1000．
(Plate IV：41)，while some take place at the phase of
n1icrosporocytes(Plate IV：42，43)． Abortion iS also
observed in field observations，interestingly，flowers
developed on the top of axial shoot and every ramifica—
tion are normal(Plate IV：44，45)，While that under axi-
al shoot and ramification are always abortive(Plate IV ：
46—51)，this has never been seen in other genera of
Oentianaceae．
3 Discussion
In S． bimaculata，anthers are tetrasporangiate．
The cytokinesis of the mi erospore mother cel in meio—
sis is of the simultaneous type．Most of the mi crospore
【=]
5期 黄衡宇等 ：獐牙菜小孢子发生及雄配子体发育 591
1 — I
Plate IV 41．Aborted stan1enX100；42，43
． Aborted microsporocyteX 1000；44，45
． Natu f】。wers；46—51．FJ。wers th aborted starnen
．
tetrads are tetrahedral and there are stilj a few oth—
er types，such as isobilateral，dilatera1
． Pollen
grains are mainly 2一celled type when shedding，OC—
casionally 3一celled type． In investigations of more
than 2000 angiosperms by Brewbaker，it was ob—
served that pollen grains of 7 0％ of those plants
were 2一celled pollen while al1 of species that deve1一
op primarily(Brewbaker，1 9 6 7)were 2～celled pol—
len．Thus，2-celled pollen was considered as prima—
ry characters and 3-celled pollen iS derived from 2一
celled pollen(Yuan et n ．，1991)． In S
． 6 7 ac Z口～
ta，pollen was always 2-celled which indicates that
there are some primary characters existed
． In com—
paring with 3一celled pollen，the livingness of 2一
celled pollen is obvious stronger． Through compa—
ring with 3-celled pollen，the followings may be the
reasons for why the livingness of the 2一celled pol—
lens are stronger than that of 3-celled：① 3-celled
pollen consume more stored nourishments in the
course of producing two sperms during the last mi—
592 广 西 植 物 3O卷
tosis；②Generally，the ektexine of 3-celled pollen is
thinner and pigment is fewer，so its resistivity is
lower；③3一celled pollen is in an active state of me—
tabolizability while 2-celled pollen is in a state of
dormancy．Consequently，their perdurable ability is
different．
The phenomenon that abortions are undertak—
an in a large numbers of male has never been ob—
served in the other species of Gentianaceae．But it
was reported in Stachyuraceae by some authors．
Tang et a1．(1983)observed that there are at least
six species of bisexual flowers in Stachyuraceae had
no pollen．It is said that it is bisexual flower but
actually differentiated into dioecious． W ei et a1．
(2001)investigated on Stachyurus himalaicus，
through further observations and discussions about
the phenomenon，he proved that those flowers were
functiona1 unisexua1．And he called them“pseudo—
hermaphrodite flower” or“functional unisexual
flower．Synchr0nously，W ei et a1．(2002)indicated
that development of ovule was later than that of
staminate in functional male flower．W hen pollens
in the anther are mature and begin to shed，the o—
vule has developed to the phase of megaspore
mother cells． As a result，ovule can not undergo
further development and pollinate，it will disappear
with the abscission of flower． W hereas，in fune—
tional female flower，the staminate has degenerated
when micr0sp0rocytes enter into prophase of meio—
sis and can not go on developing． According to
field observations，although there are large num—
bers of“functional female flower”existed in S．bi—
maculata but the evidence that could prove those
flowers are differentiated into dioecism is not e-
nough．The phenomenon can be explained by re—
dundancy theory of ecology，it is said that each or—
san of plant is composed of many members that
perform the same function，that means composition
of organs are redundant，such as some vegetative
organs，stem，leaf and root，as well as those genera—
tive organs，stamina，pistil and flowers． Cutting
down redundance appropriately can help plant dis—
tribute and use those limited nourishment and en—
ergy to improve utilization rate of limited energy
(Li，2006)．As to living beings，for resources are
limited，they increase functional inputs on one cer-
tain aspect is inevitable at the cost of reducing the
other functions(Li，2OO6)according to strategy the—
ory of life history of living beings．As a result，re—
ducing the inputs on nourishment and energy dur—
ing the development of stamina is to ensure the de—
velopment of pistil．
In 1984，Reghuvanshi and Singh(Johri et a1．，
1 992)reported that in Capsicum，male abortion was
frequently appeared in nine species．The formation
of abortive pollen is caused by abnormal function of
tapetum ，disaggregation of middle layer，degenera—
tion at a relatively early stage or abnormal augment
of tapetum，enlargement of cell nucleus，vaeuoliza—
tion and radial elongate of cells resulting in abnor～
mal meiosis and formation of 2一or 4一nucleate mi—
crospores．At last，nuclei degenerate resulting in a～
bortion．In China，there are two kinds of mecha-
nisms to explain it．One said it is caused by abnor～
mal meiosis of microsporocytes，because of the
highly vaculated tapetal cells which extrude micro～
spores resulting in abnormal meiosis and the for～
mation of natural mierospores tetrads．Thus abor～
tion occurs before the formation of mierospores tet—-
rads(Wu et a1．，1988；Geng et a1．，1994)．Howev—
er，the other said the meiosis of microsporocytes
and the formation of microspores tetrads were all
normal，abortion occurs at the phase of male game—
tophyte．Abnormal variation begins when nucleus
is at peripheral position，meanwhile，tapetal cells
take radial enlargement abnormally and extrude
pollen grains which make mierospores are disag-
gregated before forming 2-celled pollen resulting in
abortion(Wang et a1．，2004)．According to obser—
vations from sections，we agree with the former，
via．it is because of abnormal function of tapetum。
disaggregation of middle layers，degeneration at a
relatively early stage or abnormal augment of tape—
turn，enlargement of cell nucleus，highly vacuoliza—
tion and radial elongation of cells．
It is observed that there are always some
5期 黄衡宇等：獐牙菜小孢子发生及雄配子体发育 593
dichophysis that are similar to mi．crospores arches—
porial cells on morpha in microsporangium from in—
vestigations on microspor0genesis of Gentianaceae，
Scrophu1ariaceae，Labiatae，Bignoniaeeae and Arali—
aceae，and the origin，evolvement，type of those
dichophysis are in debate．(；uerin(1926)considered
that those dichophysis were derived from archespo—
rial cells according to the observations that
dichophysis in microsporangium of Gentianaceae
separated the anther into many small lumina or ac—
cureulated to a mass．Ralph(1949)also found 1arge
numbers of dichophysis in some species of Swertia
L．，and considered that those diehophysis were de—
rived from parenchyma cells of connective and sim—
ilared to primary parietal cells on function and ori—
gin．Steffen et a1．(1958)reported the dichophysis
in Gentiana L． again and supported the former．
Steffen et a1．(1958)separated it into two kinds ac—
cording to morpha：(1)Ptacentoide which accumu—
lates in angle of anther wall is similar to placenta；
(2)Trabeculae which separates the anther chamber
into many small lumina is in the center of sporangi—
um and cells elongate longitudinally．Eames(1 9 6 1)
believed that the dichophysis was not derived from
archesporial cells，and it needed further observa—
tions and confirmations．Bhoj wani(1 9 7 9)and Echi—
lin(1971)indicated that，either inferred from origin
or function，the structure was of the characteristic
of tapetum． It was considered as sporogenous tis—
sue in the past possibly because tapetal cells and
sporogenous cells were similar on morpha at the
early stage observed under optical microscope．Rao
Pt口Z．(1 983)reported that the structure was de—
rived from sporogenous tissue in Cansora． Zhu et
d Z．(1989)and Li et a1．(1994)reported that in Gen—
tiana L．separately and considered it was derived
from primary parietal cells and parenchyma ceils of
connective．W e are trying to explain the phenome—
non as follows according to early reports and our
observations on movements of microspores and ta—
petum． Firstly，the dichophysis in microsporangi—
um is of characteristic of tapetum without doubting
and may be of dual origin which is derived from
primary parietal ceils and connective tissues．Sec—
ondly，both of the tapetal cells that from primary
parietal cells and connective tissues undergo abnor—
mal divisions to form Placentoide and Trabeculae．
There is only plaeentoide in Swerti& L．(Ralph，
1949)，while in Gentiana L．(Guerin，1958；Steffen
et a1．，1958；Zhu et a1．，1989；Li et a1．，1994)and
Gentianopsis Ma．(Liu et a1．，1997)，both types are
existed．As to the type of the tapetum discussed a—
bove，the previous authors(Guerin，1926；Ralph,
1949；Steffen et a1．，1958)had not discussed．Zhu
et a1．(1989)considered that was of glandular type，
while Rao et a1．(1983)and Li et a1．(1994)thought
that was of amoeboid type．0ur observations indi—
cates that the type we discussed is different from
typical amoeboid type，for that cell wall of tapetum
is destroyed earlier resulting in protoplasmic mass
intruding into the anther chamber to form proto—
plasm surrounding the microspores．It is also dif—
ferent from typical glandular type causing that the
cells of tapetum differentiate only to the inner wall
of anther but do not intrUde into the anther cham-
ber to form placentoide or trabeculae． However，
according to the process of degeneration of tapetal
cels in S．bimaculata，all of those cells degenerate
at their original site and protoplasm do not move．
Thus the tapetum is more close to glandular type．
The structure，function and degeneration of the ta—
petum need further investigations． Especially，the
observations under electron microscope can helpful
not only to the understanding of the origin，differ—
entiation and type of the tapetum ，but also to the
acknowledgement of development and function of
tapetum in reproductive biology of angiosperm．
References：
Bhojwani S，Bhatnagar S．1979．The embryology of angiosperms
rM]．New Delhi：vikas Pub1．House
Brewbaker JL．1967．The elistribution and phylogenatie signifi—
cance of binucleate and trinucleate pollen grains in the angio—
sperms[J3．AmerJ Bot，59(9)：1 068—1 083
Chen JC，Yu W ，Cai DY．1 998．The survey of study and applica
tion in medieinal plants Swertia[J]．Chin J Integrated Trad
Western Medicine on Liver Diseases，8(Supp1)：223—226
Davis Gwend L 1966．Systematic Embryology of the Angiosperms
(下转第 690页 Continue on page 690)
690 广 西 植 物 3O卷
用，且对其不定芽质量也有一定的促进作用 ；但是长
期使用 6-BA会产生玻璃化现象。低浓度的 NAA
适宜于白花天 目地黄不定芽的生根。
本试验成功建立 白花天 目地黄愈伤组织再生体
系，为其在庭院观赏花卉绿化应用提供了可能性；同
时在保存地黄属优良种质资源和改良地黄药材品质
等方面具有重大意义。
参考文献：
Chen MY(陈敏艳)，Liang Zs(梁宗锁)，Wang JZ(王吉之)，et a1．
2004．Tissue culture and plantlet regeneration of Rehmannia
glutinosa(Yg黄组 织培养及植株 再生的 研究 )[J]．Acta Bot
Boreal—Occident Sin(西北植物学报)，24(6)：1 083—1 089
Cui YY，Hahn FJ，Toyoki Kozai，et a1．2000．Number of air ex—
changes，sucrose concentration，photosynthetic photon flux，and
diferences in photoperiod and dark period temperatures affect
growth of Rehmannia glutinosa plantlets in vitro plant cell[J]．
丁issueOrgan Culture，62：219—226
Seon JH，Cui YY，T0yol【i Kozai，et．d．2000．Influence of in vitro
growt h conditons on photosynthetic competence and survival
rate of Rehmannia glutinosa plan tlets during acclire atization pe—-
riod Plant Cel[J]．TissueOrgan Culture，61：135 一142
Li GY(李根有)，Xia GH(夏国华)，Ma DD(马丹丹)．2009．Tow
new forms of Rehmannia ckingii Li from Zhejiang(浙江天目地
黄 2新变 型)[J]．Acta Bot Boreal-Occident Sin(西北植物学
报)，29(1)：193—194
Liang J(梁佼)，Cheng LJ(程丽娟)，Chen Y(陈莹)，et a1．2007．
Tissue culture and establishment of cloml propagation of Digita-
lis purpurea variant(毛地黄优 良变异植株组织培养及无性系的
建立)[J]．ShandongAgric Sci(山东农业科学)，6：11—13
Shao cY(邵春月)，Gao SL(高山林)，Chen F(陈峰)，eta1．2008．
Virus—free culture and rapid-propagation of Rehmarmia glutinosa
Libosch(怀地黄分生组织培养脱病毒及快速繁殖技术 的研
究)[J]．Pharm Biotech(药物生物技术)，lS(4)：258～261
Wen xs(温学森)，Huo DL(霍 德兰)，Yang SL(杨世林)，et a1．
2002．Multiplication of virus-free seedlings of Rehmannia gluti-
”om CV．85—5 in vitro(地黄优 良品种“85—5”脱毒苗的快速繁殖
研究)[J]．Ch锄Trad Herb Drugs(中草药)，33(5)：452-455
Xie xH(解晓红)，Wu ZX(武宗信)，Feng WL(冯文龙)。2003．
Rehmannia stem-apex-propagation techniques and problems(地
黄茎尖快繁技术及其问题探讨)口]．J Shanxi Agric Sd(山
西农业科学)，31(3)：66—68
Yah K(囝坤)，Zhao N(赵楠)，Li HQ(李宏庆)．2007．Systematic
relationships among Rehmannia(Scrophulariaeeae)species(地黄
属种问亲缘关系研究)[J]。Acta Bot Boreal—Occident Sin(西
北植物学报)，27(6)：1 112—1 120
Zhao N(赵楠)，Wu w(武雯)，Yan K(国坤)，et a1．2007．Effect
of different plant growth regulators on tissue culture and plantlet
regeneration of Rehmannia chingii(不同植物生长调节物质对
天目地黄组织培养及植 株再生的影 响)[J]．J Henan Agric
Univ(河南农业大学学报)，41(1)：33-37
(上接第 593页 Continue from page 593)
[M]．New York：John Wiley
Delectis Florae Reipublicae Popularis Sinicae，Agendae Aeademiae
Sinicae Edita．1988．Flora Reipublicae Sinicae Tomus 62．rM]．
Bering：Science Press，385～386
Eames AJ．1961．Morphology of the Angiosperms[M]．New
York：McGrAw-Hil Book Company Inc。
Echilin P．1971．The role of tapetum during microsporogeness of
angiospermsi,M]．London：Buterworth：In Heslop Hardson，J．
(e也)．Pollen development and physiology
Guefin P．1926．Le development de I’anthere chez Ies Gentianaee—
ae[J]．Bull Soc Bot，73(ser5)：5—18
Geng SS，Wang ZY，Jiang JZ，et a1．1994．Cytological studies on
of the male sterile lines of pepperiJ]．Acta
Hort Sin，21(2)：165—169
Johri BM，Ambegaokar KB，Srivastava PS． 1 992． Comparative
Embryology of Angiosperms I-M]．Berlin：springer—Verlag
Li DY．2006．Redundancy theory and its application to ecology
l-J]．J Nantong Univ(Nat Sci)，5(1)：5o一54
Li}U，Wang YZ．1994．Emeryological studies in Gentiana?HgLCYO—
phylla[J]．Acta Bot Boreal-Ckcident Sin，14(4)：243—248
Liu JQ，He TN．1997．Embryology of Gentianopsis paludosa口]．
Acta Biol Plateau Sin，13：31— 41
Long H ，Huang HY．2008．The seed germination of Swertia．Bi—
maculatai,J]．Bul Bot Res，28(3)：347-352
Ralph MW．1949．On the embryology of Szoertia carolinensisIJ]。
Bull Torrey Bot，76(6)：430-439
Rao KS，Chinnappa CC．1 983．Studies in Gentianaceae．Microspo‘
rangium and pollen[J]．Can J Bot，61：324-336
Steffen K，Landmann W． 1958．Entwieklungsges ehichtliehe und
cytologische untersuchungen am balken tapetum yon Gentiana
cruciata and Impatiens glanduliferaiJ]．Planta，50：423-460
Tang YC，Cao YL，Xi YZ，et al。1983．Systematic studies on chi—
nese staehyuraceae(1)一phytogeographical，cytological，palynologi-
calU]．ActaPhytotax Sin，21(3)：236-253
Wu HM。Yu JM，Zhao HL。eta1．1988．Cytological observation on
the male-sterile and maintainer lines of Capsicum frutescens vat．
1ongrum Baileyi,J]．Jiangsu J Agric Sci，4(2)：35—38
Wang SB，Luo XD，Dai LF，et a1．2004．Meiotic observation and
male gametes development in cytoplasmic male sterile of pepper
(Capsicum annuum)[J]．Acta Hort Sin，31(6)：807-810
W_cj ZX，Yang ZH．2001．Growth．development and 50me biological
phenomena of Stachyurus himalaicus under different envimnmen-
tal conditionsi,J]．Chin J Appl Environ Bio，7(4)：3l5—3z0
Wei ZX，Jin O2，Yang SX，et a1．2002．The development of male
and female gametophytes of Stachyurus himalaicus and its Sys-
tematic enlightenmenti,J]．Acta Bot Yunnan，24(6)：733—742
Yuan RJ，Li P，Zhen X J． 1991． Mega-
mega-microgametogenesis and changes in Polysaecharides in C／i-
tonia ude,sisi,J]．Acta Bot Yunnan，13(3)：197-302
Zhu XH，Shen JH。1989．Microsporogenesis and megasporogenesis．
and the development of the male and female gametophytes in G —
tiananu~nshurica[J]．NatSci J Harbin Normal Univ，5：63—73