Memoirs, Horticultural Society of New York (1927)

Bureau of Plant Industry, U. S. Department of Agriculture

Sterile first-generation hybrids are not uncommon especially when very diverse species are crossed. It is, however, unusual to find such hybrids producing seeds apparently normal in appearance and of good viability. Such seeds are produced by many sterile F1 hybrids obtained by crossing the common sweet orange, Citrus sinensis, with the Chinese trifoliate orange, Poncirus trifoliata.

The author made the first accurately controlled cross-pollinations between these two species and secured one hybrid, named the Rusk citrange (1 and 2), by using pollen of Poncirus trifoliata on a flower (bagged and castrated) of Citrus sinensis. Only one hybrid was produced by this crossed fruit. At the same time another cross was made by applying pollen of the sweet orange to a properly bagged flower of the trifoliate orange and from a crossed fruit so secured a dozen hybrid seedlings developed, among them the Colman, Morton, Savage, Rustic, Etonia, Norton, Phelps and Sanford citranges. These citranges have been under observation for more than a quarter of a century and second generation plants grown from the seed have been studied more than twenty years.

Most of the citranges have never yet produced a fertile female gamete and consequently have not yielded any second generation seedlings, although most if not all of them do produce some fertile pollen which can be and has been used in making complex hybrids with other citrus fruits such as the citrangequats which were obtained by using pollen of citranges on the flowers of the oval and round kumquats, Fortunella margarita and F. japonica (3).

In the course of very extensive cross-pollination work by the author and his colleagues it was soon found that all of the citranges set seeds if pollinated and that these seeds were usually plump and viable, but the seedlings were, in most of the citranges, exactly like the mother parent and showed no trace of any influence of the male parent. In the case of the Rusk citrange, which seeds rather freely, it was found that the seedlings (often grown many hundreds at once under identical conditions in the greenhouses) were remarkably similar in appearance. One such lot of Rusks, more than 1,500 in number, was carefully studied and failed to show any noticeable variation; the whole bench full looked exactly like plants propagated vegetatively. Now that the Rusk citrange is being used on a considerable scale as a stock upon which to bud other citrous fruits, it is propagated both from seed and by cuttings and it is very difficult, if not impossible, to distinguish the cuttings from the seedlings when once the seedlings have gone past the early stage of germination. At the early stage, of course, they look very different from cuttings, which are set out with fully grown leaves. Many Rusk citrange seedlings have been fruited and in all such cases so far observed the fruit is exactly like that of the parent variety. As the Rusk is a very striking fruit, of deep red color when ripe, with an unmistakable aroma and flavor, it would be easy to detect any variation from the F1 in the seedlings. No such variations have been found, in spite of repeated search.

The Rusk citrange, and in fact most of the citranges, shows extraordinary vegetative vigor. Almost all of the citranges grow faster than the combined growth of both parent species. Several of them, and in particular the Rusk, have been found after long experience and tests to be excellent stocks upon which to graft the Satsuma and other oranges. It happens, therefore, in this case that the unusual and unexpected possibility of propagating generation after generation from seed without any perceptible variation is a very decided advantage in the practical utilization of these plants for stocks. Several of the citranges, and in particular the Savage, Morton and Rustic, all of which, like the Rusk, produce only F1 seedlings, are also being grown in the same way from seed for use as stock plants. Only two citranges, the Phelps and the Sanford, are known to produce regularly numerous true F2 seedlings.

The scientific explanation of this very curious and unexpected phenomenon is very simple indeed, as Strasburger (4) showed years ago the supernumerary embryos found in the seeds of most species of Citrus are produced by buds growing out from the nucellar tissue into an embryo sac. Naturally these buds reproduce the mother variety almost without variation, in the same way as do buds taken from ordinary vegetative branches. It is true that such nucellar buds do differ from ordinary buds in that the ontogeny is greatly changed and the young seedling developed from the nucellar bud has cotyledons and develops exactly like a normal seedling. This can perhaps be considered as a result of the powerful direct influence exerted on ontogeny by the embryo sac and its various organs. The changed ontogeny of such a nucellar bud can be compared somewhat remotely to the changed development of the egg or young larvae of the honey bee when the workers find that they have lost their queen and by enlarging the cell about the young larvae that would have developed into a worker cause it to develop into a queen bee by giving it special food and larger space for development. These and other relations observed in the development of nucellar bud embryos are valuable material for the study of endocrinology, until recently the monopoly of animal physiologists, but also beyond doubt of great importance in plant physiology and plant ontogeny as well.

It is known that supernumerary embryos derived from the mother plant occur in mangos and in other plants; it is probable that they may likewise prove interesting both from scientific standpoint and in practical plant culture.


  1. Webber, H. J., and Swingle, Walter T. New Citrus creations of the Department of Agriculture. U. S. Dept. Agric. Yearbook. 1904.
  2. Swingle, Walter T. Variation in first generation hybrids (imperfect dominance): its possible explanation through zygotaxis. IVe Conference Internat. Génetique, pp. 381-394. Paris, 1913.
  3. Swingle, Walter T., and Robinson, T. Ralph. Two important new types of Citrous hybrids for the home garden—Citrangequats and Limequats. Jour. Agric. Research 23: 229-238. 1923.
  4. Strasburger, E. Ueber Polyembryonie. Zeitschr. f. Naturwissen-sch. 12: 634-678. 1878.
  5. Swingle, W. T., Robinson, T. R., and Savage, E. M. New Citrus Hybrids. USDA Circular No. 181. 1931.