In vitro Culture of Immature Soybean Embryos

VARIEN R. TILTON and SANDRA H. RUSSELL

Agrigenetics Advanced Research Division, Cell Biology Section, 5649 East Buckeye Road,
Madison, WI 53716

Received November 10, 1983 . Accepted January 18, 1984

Summary
Immature embryos of soybean [Glycine max (1.) Men., cvs. Beeson, Chippewa, Swift, and
Tonica] ranging from globular to cotyledon stages were cultured in vitro. A variety of media
consisting of 24 combinations of basic media salts, growth regulators, and carbon sources was
tested in conjunction with initial exposure of cultures to light or darkness. The best combina-
tion tested was B5 medium supplemented with 0.1 mg .1- 1 indole-3-butyric acid (IBA) and an
intitial exposure to darkness for two weeks. The survival rate increased progressively with
embryo age and stage of development. Approximately 10 % of late globular-early heart stage
embryos survived to sexual maturity and produced seed whereas virtually 100 % of the cotyle-
don stage embryos produced plants which flowered and set fruit. Approximately 25 % of the
globular stage embryos were induced to initiate a hypophysis but they subsequently either
formed callus or died within a week. Although plants resulting from in vitro culture of imma-
ture embryos were diminutive, they otherwise were morphologically normal and produced
seed of normal size and weight, as did their progeny.
Key words: Embryo Culture; Glycine max; Soybeans.

Introduction
The in vitro culture of immature embryos has a number of basic and applied applica-
tions. In the former regard, in vitro culture systems allow for the study and fine control
of physiological, biochemical and environmental factors influencing embryogenesis.
In the latter regard, the most significant applied application is embryo rescue in sexual
crosses that otherwise result in embryo abortion and loss of the hybrid genome.
Embryo rescue is particularly important in interspecific and intrageneric crosses
used in breeding systems that are designed to greatly increase available genetic varia-
tion. Such crosses often are prevented from occurring naturally by anatomical, phy-
siological, or biochemical barriers. However, if the cross does occur, be it naturally
or by an in vitro pollinationlfertilization technique, the resulting embryo commonly
aborts because of endosperm failure.
In soybeans, attempts to make interspecific hybrids date back at least to the work
of Palmer (1965) and, most recently, include the work of Hymowitz and co-workers.
Hymowitz's work includes embryo rescue techniques (Newell and Hymowitz, 1982;
Ladizinskyet al., 1979) which, in soybeans, date back at least to the efforts of Chan
(Chan and Lin, 1967; Chan, 1969). Hymowitz's work also includes seed culture tech-

J. Plant Physiol. Vol. 115. pp. 191-200 (1984)

 192 VARIEN R. TILTON and SANDRA H. RUSSELL

niques (Newell and Hymowitz, 1983) as does the work by Hsu and Obendorf (1982).
Obendorf et al. (1983) successfully cultured seeds to maturity via in vitro culture of
whole fruits. Other publications on soybean embryo and seed culture and on
soybean embryo rescue that report varying levels of embryo and seed survival in-
clude Braverman (1975), Cutter and Bingham (1975), Thompson et al. (1977), and Va-
gera and Hanackova (1979). Although excised seeds have been cultured in vitro, true
ovule culture has not yet been accomplished with soybeans.
Objectives in the work reported here are two-fold. The first objective was to in-
crease the survival rate of older, but still immature (late heart to cotyledon stages;
5-21 days old; ;;:: 0.5 mm in length) soybean embryos cultured in vitro by developing
a protocol improved over those published in previously cited references. The second
objective was to culture younger (globular to young heart stages; 2-4 days old;
0.2-0.3 mm in length) embryos, with the ultimate goal of isolating and culturing
zygotes. The first goal has been accomplished and one can now be assured of virtually
100% germination and continued growth with older embryos. Significant advances
have been made toward the second goal, and continued progress in that work is at a
natural breakpoint. This breakpoint occurs between the globular and late-globular
stages, before all embryonic primordia are initiated. Although the survival rate with
very young heart stage embryos is low, one can culture to sexual maturity embryos
of all stages that have initiated primorida of all the major embryonic tissues.

Materials and Methods

Material was collected from soybean [Glycine max (L.) Merr., cvs. Beeson, Chippewa, Swift,
and Tonica] plants grown in the Agrigenetics Advanced Research Division greenhouse at Ma-
dison, WI. Plants were grown at 22/18 °C and under aI4/10h photoperiod using high pressure
sodium lamps to supplement natural lighting when required.
Young fruits were surface sterilized for 1 min in reagent grade ethanol and subsequently
rinsed twice in sterile water, 5 min/rinse. A dissecting microscope was used to aid in excising
the young embryos after the immature seeds were removed from the ovary. To excise globular
to young cotyledonary stage embryos an incision was made through the two integuments and,
depending on age, nucellar tissue and embryo sac wall were cut as well (Fig. 1). Two techniques

Fig. 1: Diagramatic representation of embryo size and position in relation to rest of seed at time
of excision.

were used to expose young embryos for removal; either the tip of the scalpel was used to lift the
embryo or tweezers holding the seed were squeezed gently to eject the embryo through the in-
cision. With older embryos, an incision was made through the young seed coat which was then
peeled off to expose the immature embryo for removal. Embryos were placed in liquid or on

J. Plant PhysioL VoL 115. pp. 191-200 (1984)

 Soybean embryo culture 193

Table 1: Quantitative List of Media Components Tested.

Basic Medium Salts: Neal's HLH (N)
Murashige and Skoog (MS)
Gamborg's B5 (B5)
Beasley and Ting (BT)
8p (8p)
Carbon Sources: sucrose (0.025-6.0 %)
glucose (1.8-6.84 %)
fructose (0.025-3.6 %)
ribose (0.025 %)
xylose (0.025 %)
mannose (0.025 %)
rhamnose (0.025 %)
cellobiose (0.025 %)
sorbitol (0.025 %)
mannitol (0.025-10.0%)
Growth Regulators: BA (0.5-2.5 mg ,1- 1)
lAA (1.0-15.0 mg ,1- 1)")
NAA (0.47-1.0 mg .1- 1)
2,4-D (0.05-0.2 mg .1- 1)
Zeatin (0.5 mg ,1- 1)
IBA (0.001-1.0mg ,1- 1)")
Kinetin (0.21 mg ,1- 1)
Other Additives: casein hydrolysate (0.5-5.0 g ,1- 1)
total homogenate from fresh tomatoes b )
") filter sterilized and added after medium is autoclaved.
b) substituted equal volume for 50 % of water in medium.

solidified medium in petri dishes and then cultured at 28°C in darkness for 1-2 weeks.
Embryos were oriented so the hypocotyl pointed downward.
Twenty four media that consisted of different combinations of basic medium salts [Neal's
HLH (N), (Neal, 1981); Murashige and Skoog (MS), (Murashige and Skoog, 1962); Gamborg's
B-5 (B5), (Gamborg et al., 1968); Beasley and Ting (BT), (Beasley and Ting, 1973); and Sp me-
dium (Sp), (Kao and Michayluk, 1975)], carbon sources [sucrose, glucose, fructose, ribose,
xylose, mannose, rhamnose, cellobiose, sorbitol, and mannitol], growth regulators [N6-benzyl-
adenine (BA), indole-3-acetic acid (IAA), naphthalene-I-acetic acid (NAA), 2,4-dichlorophe-
noxyacetic acid (2,4-D), zeatin, indole-3-butyric acid (IBA), and kinetin], and other additives
[casein hydrolysate and total homogenate from fresh tomatoes*)] were tested (Table 1).
After germination, germlings were exposed to a 16/S h photoperiod. As the resulting
seedlings outgrew their petri dishes, they were transferred to solid medium in Magenta boxes
(Magenta Corporation, Chicago, IL) to allow greater shoot growth. When these seedlings had
well-established root and shoot systems, they were transplanted to potting mix and placed in a
Wardian case in the greenhouse for approximately 2 weeks to harden. Plants subsequently were
transferred to a greenhouse bench.

*) Ripe tomatoes were pureed in a blender, filtered through cheesecloth, autoclaved, and re-
filtered.

J. Plant Physiol. Vol. 115. pp. 191-200 {1984}

 194 VARlEN R. TILTON and SANDRA H. RUSSELL

To compare plants resulting from embryo culture with conventionally grown controls, we
used average seed weight, and progeny size, morphology and fecundity as test characters. One
lot of 25 seeds was collected and weighed from each of 7 plants selected at random from plants
obtained by in vitro embryo culture. Three seeds from each lot were sown and the resultant
plants allowed to achieve maximum size and sexual maturity. For controls, we collected 25
seeds each from randomly selected plants grown conventionally from seed of 4 soybean cul-
tivars (G. max cvs. Beeson, Clark, Swift, and Tonica) and determined the average seed weight.

Results
Notwithstanding the fact embryo development is dynamic and one stage grades
into the next, embryo development was divided into five principal stages: zygote,
globular, heart, cotyledon, and mature. Embryos were categorized on the basis of
their size and morphology (Fig.2) and intermediate stages indicated by the prefixes
young or early, and late. Embryo development and fruit size were correlated suffi-
ciently to allow ovary length to be used as a criterion for selecting fruits that contain
the three principal stages of embryo development of interest in this report, viz.,
globular, heart, and cotyledon (Table 2; Fig. 3).
Best growth in vitro resulted when embryos were oriented vertically. This was par-
ticularly true with younger embryos just developing cotyledons because cotyledons
become swollen and puffy when in contact with the agar surface. The swelling of
only one cotyledon often resulted in an asymmetry that younger embryos could not
compensate for. In extreme cases with late heart or very early cotyledon stage

2 3

Fig. 2: Photomicrograph of mature, cotyledon, heart, and globular soybean embryos.

Bar = lmm.
Fig. 3: Soybean fruits which contain embryos corresponding in size and stage to those in Fig. 2.
Bar=5mm.

J. Plant Physiol. Vol. 115. pp. 191-200 (1984)

 Soybean embryo culture 195

Table 2: Fruit Lengtha ) vs. Embryo Stage, Size, and Age.

Fruit Length (mm) Embryo Stage Embryo Size (mm) Embryo Age
20 Globular 0.2 2 days
15-30 Heart 0.2-0.5 3-5 days
30 Cotyledonary 0.5 5-21 days
a) Variation in fruit length is correlated in part with number of developing seeds.

100
90
~ 80
';:
::>
15
70
::::;;
60
"0
::>
>< 50
Q)
(f)
40
.2
30
0
>
.~ 20
:::>
(f)
10
~
0

0
Globular Late Heart Late Cotyledon
Globular- Heart-
Early Early
Heart Cotyledon

Fig. 4: Survival rate to sexual maturity vs. embryo stage at excision for Tonica embryos on B5
medium supplemented with 0.1 mg .1- 1 IBA.

embryos, the swollen cotyledon dwarfed its mate by several fold, the smaller cotyle-
don did not develop, and the embryo eventually died.
Embryos developed beyond the heart stage at the time of excision germinated pre-
cociously when placed into culture on most of the media tested. The cotyledons
spread apart within a few hours of excision and, after about 12 hours, the plumule
was exposed. At this point, the embryos looked like miniatures of conventionally
grown germlings in the first stages of germination. This initial reaction occurred in
virtually 100 % of the embryos with at least some cotyledon development. However,
further growth and sustained ontogeny of heart stage and younger embryos was
highly dependent on the cultivar, culture medium and culture regimen.
We obtained our best results for germination and sustained development of heart
stage and older embryos by placing them on solidified (0.8 % wlv agar) B5 medium
(Gamborg et al. 1968) supplemented with 0.1 mg .1- 1 IBA, and then placing them in
darkness for 2 weeks at 28°C. After the dark incubation period, during which time
embryos germinated and established basic root and shoot systems, cultures were placed
under lights and exposed to 45,000 Ix during a 16/8 h photoperiod, also at 28°C.

J. Plant Physiol. Vol. 115. pp. 191-200 (1984)

196 VARIEN R. TILTON and SANDRA H. RUSSELL

Fig. 5: Diminuative but sexually mature plants from in vitro culture of late heart stage embryos
(large arrows) accompanied by progeny of similar plants. Note fruits on in vitro cultured plants
(small arrows).

Of the 4 cultivars tested, the highest and most consistent survival rates were ob-
tained with the cultivar Tonica (Fig. 4), followed respectively by Swift, Beeson, and
Chippewa. Greenhouse conditions, and the physiological and developmental state of
the donor plants were also significant factors. The best embryos were obtained from
young, healthy, vigorously flowering plants grown under carefully monitored
greenhouse conditions.
Using this protocol and Tonica embryos, approximately 10% of the late globular-
early heart stage embryos grew to sexually mature plants and set seed. The survival
rate increased dramatically to 66 % with heart stage embryos and continued upward
to 75 % with late heart-early cotyledon stage embryos. With older cotyledon stage
embryos a survival rate of virtually 100 % was obtained.
Some globular stage embryos subsisted under this regime without showing any ex-
ternal signs of change for 5-7 days before they became moribund and died. In the li-
mited trials to date with liquid culture in Cuprak dishes (Costar, Cambridge, MA,
USA) Kao and Michayluk's (1975) 8p medium and our own modification of 8p (8p
with 3 % sucrose and 10 % mannitol instead of 6.84 % glucose), 75 % of the globular
embryos either formed callus or became moribund and died. The remaining 25 % of
the globular embryos, however, did show limited morphogenetic development and
did form a hypophysis, but they too eventually formed callus or died.
Plants grown from in vitro cultured immature embryos were diminutive relative to
plants grown from mature embryos excised and grown in vitro or from mature seed.
The plants grown from immature embryos reflected the size of the embryo at the
time it was excised, progressively smaller plants resulting from progressively smaller
embryos. These diminutive plants had shorter and fewer internodes and had smaller
leaves, but they were otherwise morphologically normal and sexually viable (Fig. 5).

J. Plant Physiol. Vol. 115. pp. 191-200 (1984)

 Soybean embryo culture 197

Table3: Seed Weights.

--------------------------------------
I. Seed From In Vitro Cultured Immature Embryos*)
Plant No. Average/Seed (gm) Average/25 seeds (gm)
A 0.2064 5.16
C 0.2017 5.04
G 0.1884 4.71
J 0.2107 5.27
K 0.1853 4.63
L 0.2262 5.66
M 0.2093 5.23
(0.204±0.014) (5.10±0.35)

II. Control Seed

Cultivar Average/Seed (gm) Average/25 seeds (gm)
Beeson 0.2353 5.88
Clark 0.1778 4.46
Swift 0.1727 4.32
Tonica 0.2132 5.33
(0.200±0.030) (5.00±0.74)
*) For Tonica embryos on B5 medium with 2% sucrose and 0.1 mg. I-I lEA.

However, because plants derived from immature embryos had fewer internodes, they
had fewer flowers and thus produced fewer seeds per plant than did controls. Also,
the especially small plants tended to have a greater number of 1-2 seeded fruits.
In spite of their diminutive stature, plants grown from immature embryos pro-
duced seed of average weight and size relative to conventionally grown greenhouse
plants (Table3). All seeds collected and sown from these diminutive plants ger-
minated and grew into fertile plants of normal size that yielded seeds of average
weight and size relative to controls.

Discussion
The protocol reported here provides reasonable assurance for survival and growth
of immature soybean embryos cultured in vitro, especially for the cultivar Tonica.
For example, virtually 100 % of the embryos placed into culture as older cotyledon
stage embryos and as many as 75 % of the late heartearly cotyledon stage embryos
were grown to sexual maturity. This is an improvement over the protocol published
by Vagera and Hanackova (1979) in which they had a maximum survival rate of 88 %
with embryos from mature seed and 42 % with embryos from 3-4 mm seeds (late
heart-early cotyledon stages).
A protocol with the capacity for high embryo survival is particularly significant in
rescuing wide-cross hybrids between Glycine max and its wild relatives. The need for
such a protocol is readily evident when one considers that the best attempt to date to
recover wide cross Glycine max hybrid embryos in semino without using embryo cul-

J. Plant Physiol. Vol. 115. pp. 191-200 {1984}

 198 VARIEN R. TILTON and SANDRA H. RUSSELL

ture, had a survival rate of 4.7% (7 plants were recovered from 149 hybrid embryos
placed into culture in semino 19-28 days after pollination) (Newell and Hymowitz,
1982). Application of the embryo culture protocol reported here as an embryo rescue
technique may help to increase the recovery rate from future wide cross attempts,
and possibly from somatic embryos derived from protoplast culture as well (Gam-
borg et aI., 1983).
Also of note with the protocol reported here is reduction in the time interval be-
tween pollination and sexual maturity of progeny. Vigorous plants grown from coty-
ledon stage embryos can mature rapidly enough to be flowering and set fruit in
10-12 weeks after pollination, and they can be transplanted to potting mix at that
time or, in some cases, even earlier. Less vigorous plants may enter the seminific stage
of their life cycle at about 12 weeks, but they usually require an additional 1-2 weeks
in culture before transplanting to potting mix. The principal factor in determining
readiness for transplanting is a full and healthy root system.
In the orderly sequence of natural events in vivo, the root system is always the first
organ system to resume development at the onset of germination. This usually is evi-
denced by protrusion of the radicle through the micropyle. However, the first ex-
ternal evidence of germination by immature soybean embryos in vitro is the cotyle-
dons spreading apart, but this soon is followed by hypocotyl elongation and radicle
development. These initial two events in germination, viz. hypocotyl elongation and,
in particular, development of the radicle into a vigorous root system, are essential to
seedling survival for three key reasons: 1) to anchor the plant; 2) to absorb water and
nutrients; and 3) to synthesize hormones, especially cytokinins.
Although the first two reasons are rather self-evident and are not of particular sig-
nificance for in vitro systems, the synthesis of hormones in their proper sequence and
balance is vital to normal plant development. For this reason, we included IBA in our
culture medium to enhance the initial root development response and thereby
emulate natural ontogeny. The rationale to stimulate root growth was to enhance
endogeneous cytokinin synthesis which occurs in root apicies as Short and Torrey
(1972) demonstrated. In turn, cytokinins promote RNA and protein synthesis
(Wareing, 1977) and, in conjunction with gibberellic acid (GA), also supplied by the
roots, subsequently influence the rate of leaf growth (Atkin et aI., 1973).
In maize germlings, Tilton (1981) noted that the epicotyl grew very slowly when-
ever the hypocotyl was removed from the embryo axis or was retarded severely in its
growth. However, in individuals which later produced adventitious roots, the
growth rate of epicotyls increased subsequent to root development. We found a simi-
lar correlation with germinating immature soybean embryos; those germlings with
vigorous radicle and root growth develop vigorous shoot systems.
Maize (Smith and van Staden, 1978) and Acer (Pinfield and Stobart, 1972) germlings
initially are dependent on endosperm-derived cytokinins. Cytokinin glycosides are
transported from maize endosperm to the embryo axis for at least the first three days
following germination. The majority goes to the radicle, but as the radicle develops it

J. Plant Physiol. Vol. 115. pp. 191-200 {1984}

 Soybean embryo culture 199

becomes the primary source of cytokinin (Smith and van Staden, 1978). Smith and
van Staden (1979) noted they could substitute partially for endosperm deprivation by
exogenous application of cytokinins, especially cytokinin glycosides.
As with immature soybean embryos reported here, immature maize embryos also
gave rise to diminutive plants (LaRue, 1936). And, additionally, immature maize
embryos deprived of endosperm in vitro give rise to smaller plants than do embryos
cultured with endosperm (Andronescu, 1919). However, as just noted, cytokinins
can partially substitue for the absence of endosperm (Smith and van Staden, 1979). In
contrast to maize, we did not notice any compensatory activity with the cytokinins
we tested on soybean (Table 1). But, our tests to date do not exhaust all possibilities
and the mechanism underlying the diminutive growth habit resulting from in vitro
culture of immature soybean embryos remains an enigma.
That a diminutive growth habit does not influence seed size and weight is an inter-
esting but not surprising phenomenon. The transition from vegetative to reproduc-
tive growth commonly occurs in response to many physiological stresses. It follows
from this that the stresses immature embryos are subjected to during excision and in
vitro culture are sufficent to cause the resulting plants to enter their reproductive
cycle precociously. With all resources channeled toward seminific activity it stands to
reason that the seeds are normal in size, especially since cotyledons function as the
nutrient storage tissue and thereby account for the major volume of soybean seeds.
And, as one would expect, progeny of in vitro cultured embryos are fully normal in
all aspects of size, morphology and fecundity.

Acknowledgements
We thank Liz Blomquist for seed weight data and, with Jane Hawley, for tending our plants.
We also thank Cynthia Koehler for typing this manuscript.

References
ANDRONESCU, D. I.: Germination and further development of the embryo of Zea mays sep-
arated from the endosperm. Amer. J. Bot. 6, 443-453 {1919}.
ATKIN, R. K., G. E. BARTON, and D. K. ROBINSON: Effect of root growing temperature on
growth substances in xylem exudate of Zea mays. J. Exp. Bot. 24, 457-453 (1973).
BEASLEY, C. A. and I. P. TING: The effects of plant growth substances on in vitro fiber develop-
ment from fertilized cotton ovules. Amer. J. Bot. 60, 130--139 (1973).
BRAVERMAN, S. W.: Aseptic culture of soybean and peanut embryonic axes to improve phytosa-
nitation of plant introductions. Seed Sci. Techno!. 3, 725-729 (1975).
CHAN, K. L.: Methods of overcoming cross imcompatibility and hybrid sterility in Genus
Glycine. J. Agric. Assn. China N. S. 6, 16-24 (1969).
CHAN. K. L. and F. M. LIN: Studies on the method of embryo culture in the cultivated and wild
forms of soybean. J . Taiwan Agric. Res. 16, 8-14 (1967).
CUTTER, G. L. and E. T. BINGHAM: Soybean embryo culture studies. Soybean Genet. Newslett.
2, 52-53 (1975).
GAMBORG, O. L., R. A. MILLER, and K. OJIMA: Nutrient requirements of suspension cultures of
soybean root cells. Exp. Cell Res. 50, 151-158 {1968}.

J. Plant Physiol. Vol. 115. pp. 191-200 (1984)

 200 VARIEN R. TILTON and SANDRA H. RUSSELL

GAMBORG, O. L., B. P. DAVIS, and R. W. STAHLHUT: Cell division and differentiation in cell cul-
tures of Glycine species and leaf tissue of soybean. PI. Cell Rep. 2, 213-215 (1983).
Hsu, F. C. and R. L. OBENDORF: Compositional analysis of in vitro matured soybean seeds. PI.
Sci. Lett. 27, 129-135 (1982).
IUo, K. N. and M. R. MICHAYLUK: Nutritional requirements for growth of Vida hajastana cells
and protoplasts at a very low population density in liquid media. Planta 126,105-110 (1975).
LADIZINSKY, G., C. A. NEWELL, and T. HYMOWITZ: Wide crosses in soybeans: prospects and li-
mitations. Euphytica 28, 421-423 (1979).
LARUE, C. D.: The growth of plant embryos in culture. Bull. Torrey Bot. Club 63, 365-382
(1936).
MURASHlGE, T. and F. SKOOG: A revised medium for rapid growth and bioassays with tobacco
tissue cultures. PhysioI. Plant. 15,473-497 (1962).
NEAL, c.: In vitro culture of embryos from compatible and incompatible crosses of Lycoper-
suon. MS Thesis, Cornell Univ., Ithaca, N. Y., USA, 1981.
NEWELL, C. A. and T. HYMOWITZ: Successful wide hybridization between the soybean and a
wild perennial relative, G. tomentella Hayata. Crop Sci. 22, 1062-1065 (1982).
- - Hybridization in the genus Glycine subgenus Glycine Willd. (Leguminosae, Papilio-
noideae). Amer. J. Bot. 70, 334-348 (1983).
OBENDORF, R. L., G. T. RYTKo, and M. C. BYRNE: Soya bean seed growth and maturation by in
vitro pod culture. Ann. Bot. 51, 217-227 (1983).
PALMER, R. G.: Interspecific hybridization in the Genus Glycine. MS Thesis, Univ. Illinois, Ur-
bana, n., USA, 1965.
PINFIELD, N. J. and A. K. STOBART: Hormonal regulation of germination and early seedling de-
velopment inAcer pseudoplatanus. Plant a 104, 134-145 (1972).
SHORT, K. C. andJ. G. TORREY: Cytokinins in seedling roots of pea. Plant PhysioI. 49, 155-160
(1972).
SMITH, A. R. andJ. VAN STADEN: Changes in endogenous cytokinin levels in kernels of Zea mays
L. during imbibition and germination. J. Exp. Bot. 29, 1067-1075 (1978).
- - Cytokinins in excised embryos and endosperm of Zea mays L. grown under aseptic con-
ditions. Z. PflanzenphysioI. 93, 95-103 (1979).
THOMPSON, J. F., J. T. MADISON, and A. M. E. MUENSTER: In vitro culture of immature cotyle-
dons of soya bean [Glycine max (L.) Merr.]. Ann. Bot. 41, 29-39 (1977).
TILTON, V. R.: The influence of individual embryonic tissues on the morphology and develop-
ment of Zea mays (Poaceae) germlings. Amer. J. Bot. 68, 980-993 (1981).
V AGERA, J. and H. HANACKOVA: Embryo culture of soybean-culture technique and possibilities
of technical application. Rostl. Vyroba 25,349-360 (1979). (Engl. sum.)
WAREING, P. F.: Growth substances and integration in the whole plant. In: JENNINGS, D. H.
(Ed.): Integration of Activity in Higher Plants. Symp. Soc. Exp. BioI. 30, 337-365 (1977).

J. Plant Physiol. Vol. 115. pp. 191-200 (1984)