Experimental studies have contributed, in recent years, to a clearer understanding of the extent to which sex expression in flowers can be modified by environmental, nutritive, and hormonal factors (323, 408), and to the formulation of model genetic control mechanisms for sex determination (324). One aspect of this interesting field is how far sex determination is centered in the developing floral primordium or is a consequence of interaction between the primordium and the plant body as a whole. Ability to culture excised floral buds to the point where their sexuality is expressed would permit of the study of the direct role within the bud of endogenous and exogenous hormonal factors.
Galun (250, 251) has made a study of the effects of auxin and gibberellic acid on sex expression in cucumber (Cucumis sativus). In this species there are monoecious varieties, which bear male and female flowers in proportions which vary in relation to environment and geno
type in a predictable way. Monoecious varieties can be grown to produce,
at least for a time, only male flowers. In addition, by suitable hybridizing, gynoecious plants (producing only female flowers) and hermaphrodite plants (producing bisexual flowers) can be obtained. Such stocks have now been used in experiments on the culture of excised floral buds in which the morphogenetic effect of IAA and gibberellic acid have been studied (252).
Contamination of the buds was eliminated by immersion of the stem apices for 4 minutes in 1 % hypochlorite, followed by dipping in ethanol and rinsing with distilled water. The buds excised at an early develop
mental stage (0.5-0.7 mm) grew considerably during the 20 days of culture into buds corresponding to the "advanced embryonic" stage (1.0 mm). The culture medium contained salts, sucrose, Β vitamins, tryptophan, casamino acids, and 1 5 % coconut milk. These studies showed that potentially male buds tended to develop ovaries when cultured in this way. Moreover this tendency to produce female flowers was en
hanced by early excision and by addition of IAA to the culture medium.
The promotion of ovary development by IAA was antagonized by gibberellic acid. One difficulty in interpreting this result was that stamen development in excised buds was significantly retarded compared with that occurring in attached buds, whereas the culture medium seemed to supply all that was needed for the development of ovaries. The de
velopment of potentially female and hermaphrodite buds was little affected by either IAA or gibberellic acid. Although all flowers are initiated as bisexual organs, there was no development of hermaphrodite flowers from buds expected to develop into unisexual flowers. A trigger mechanism operates in potentially unisexual flowers and auxin caused this trigger to operate in favor of ovary development. IAA and gibberellic acid failed to suppress stamen development in potentially hermaphrodite flowers; IAA at a concentration which converts male into female flowers did not cause female flowers to develop from potentially hermaphrodite buds. There is no trigger mechanism in the development of the her
maphrodite flowers.
Tepfer, Greyson, Karpoff, and Gerimonte (768) have also reported, so far only in abstract, that excised young floral buds of Aquilegia make extensive growth and development in a medium containing vitamins, coconut milk, IAA, gibberellic acid, and kinetin. Kinetin was essential although it had no distinctive morphogenetic effect. Omission of IAA led to abortion of carpel primordia. Gibberellic acid stimulated the elongation growth of all organs except the stamens. Normal petal ex
pansion was particularly fostered by the presence of gibberellic acid.
These preliminary studies strongly suggest that the culture of excised floral buds could be pursued further in order to assess the value of this
approach to the study of floral morphogenesis and of the chemical factors which are involved in sex expression.
Nitsch (541, 542) initiated studies on in vitro fruit development by culturing excised flowers under aseptic conditions. Flowers with pedicels sealed with liquid paraifin were sterilized by short immersions (3-10 minutes) in a decanted 5 % calcium hypochlorite solution and then washed with sterile water. The sealed ends of the pedicels were then cut and the flowers were transferred to the culture tubes. With liquid media the flowers were supported on filter paper by the technique of Heller (314) (Fig. 2 9 ) ; for solidified media 0.7-1.0% agar was in
corporated. Fruit development in a number of species was achieved using a simple medium containing inorganic salts and sucrose, always provided that the flowers had been pollinated two or more days before they were separated from the mother plant. Most such cultured fruits contained viable seed although the percentage of seed set was ab
normally low. With nitrogen supplied as nitrate, the sepals were centers of nitrate assimilation; in the absence of an external nitrogen source the nitrogenous reserves of the sepals were depleted. When excised before pollination and transferred to a simple medium, ovaries did not develop and no seed was present. However, with some species, such as tomato, parthenocarpic fruits could be developed, in culture, from unpollinated
FIG. 29. Development of tomato ovaries on a liquid synthetic medium. Flower trimmed down to ovary (middle). In culture small fruit develops (right). (Photo
graph supplied by J. P. Nitsch.)
flowers, either by incorporating auxin into the medium (0.1 mg of 2,4-dichlorophenoxyacetic acid [2,4-D] or 1.0 mg of 2-NOA per liter) or by treating the flowers directly with auxin (one drop of 100 mg of 2-NOA per liter), 24 hours before they were excised (543). Growth of such unpollinated tomato ovaries was further stimulated by incorporating the juice of both green and red tomatoes into the culture medium (545).
The activity of tomato juice appeared to be due to the presence of a natural cytokinin, the effects of which could not be reproduced by kinetin.
One feature of all the aseptically cultured fruits described by Nitsch (542) was their relatively small size. The ovaries of a Fragaria chiloensis X F. virginica cross and of Visum sativum cultured by De Capite (194), those of Tropaeolum cultured by Sachar and Kanta (636), and those of Linaria cultured by Sachar and Baldev (634) also gave diminutive fruits. In contrast, natural-sized or even larger than normal fruits, have been obtained from flower cultures of Iberis amara (462), Althaea rosea
(155), Allium cepa (295), Ranunculus sceleratus (635), and Anethum graveolens (359). Iberis amara flowers, explanted 1 day after pollination, yielded normal-sized fruits, or fruits only slightly larger than those occurring in nature, when both calyx and corolla were present and when the culture medium contained an appropriate mixture of vitamins (in-cluding calcium pantothenate) and IAA. Ovaries of Althaea rosea, ex-cised 3 days after pollination and cultured after removal of the calyx did not reach normal size; endosperm development only reached the free nuclear stage, and the embryos the heart-shaped stage. With the calyx attached, 8 5 % of the fruits reached at least normal size and contained viable seed. The beneficial effect of the calyx could not be reproduced by feeding indolebutyric acid, IAA, gibberellic acid, or kinetin. The studies with Allium, Ranunculus, and Anethum indicated that various combinations of the growth factors IAA, gibberellic acid, and kinetin promoted the enlargement of the fruit, but the most favorable treat-ments that promoted fruit size were also associated, in Allium and Ranunculus, with very low yields of viable seed. The fertility of the achenes of Ranunculus was enhanced by the use of casein hydrolyzate and coconut milk; in Anethum coconut milk enhanced, and gibberellic acid and kinetin reduced, the number of viable seeds which developed.
The advances in knowledge which have followed from the in vitro culture of fruits are clearly limited. However, such cultures have again demonstrated the importance of pollination as a stimulus to fruit growth and have indicated the involvement of the calyx in both ovary and seed maturation. Although ovary expansion can be stimulated by feeding auxin, gibberellin, and cytokinin, these growth regulators have in a
number of instances proved unfavorable to the early stage of ovule and embryo development in isolated ovaries.