Seedlings of C. taiwanica, Yuma citrange and some of the Ichang hybrids, for example, also show a low rate of polyembryony and severe cullage is needed in the nursery. Cultivars with a low nucellar rate should probably not be used as root stocks. However, it is possible that if the nucellar seedlings are of considerable root stock interest, clonal propagation by in vitro methods could be considered in the future. In varieties that produce a fairly high percentage of nucellar embryos, most of the seedlings can be recognized as representing the largest proportion of the progeny. They will have the same foliage and branching characters and will in general be of about the same size. They present a uniform and, commonly, easily recognizable type. Mixed with these will be a varying number of seedlings that show different characters, such as different branching and larger or smaller leaves, and that are usually less vigorous and of smaller size. However, in some instances occasional seedlings are more vigorous than the norm. Variations in trifoliate hybrids may be more difficult to detect because the trifoliate leaf character is dominant and all the hybrids exhibit the trifoliate leaf character. Here, in addition, one must look for size of leaves, leaf color, the tendency towards deciduousness, and other factors as well. These differences from the norm, in a large part, are the so-called “variant” seedlings which apparently are mainly produced from gametic embryos. It is these small, off-type seedlings or extremely vigorous seedlings which are variants and should be discarded upon removal from the seedbed,vertical garden indoor or later in the nursery row, in order to obtain uniform seedlings for use as root stocks. In spite of all precautions, some zygotic seedlings do escape culling and are budded and ultimately end up as orchard trees.

Some of these, but not all, may be culled out as budlings in the nursery row because of lack of uniformity as compared to the rest of the progeny. It is unfortunate that most of the people involved in the labor or nursery practices are not trained or qualified to recognize variants. They probably do a better job on the elimination of “goose necks” and “bench roots,” which in reality may not be as important. Seedling progeny of some of the monoembryonic cultivars such as Algerian tangerine and a number of shaddock cultivars, although all genetically different, have been remarkably uniform in the seedbed and nursery. When budded to standard scion varieties and planted in root stock experiments at the Citrus Experiment Station, Riverside, most of the trees at 33 years of age were as uniform in size, appearance, and yield as those budded on nucellar seedlings of other varieties. Certain inherent characteristics persist in the progeny, however. Some seedlings of Algerian tangerine, shaddock, Poorman’s orange and others continue to show a variable percentage of the seedlings with a bark disorder. In testing numerous shaddocks as root stocks for tristeza resistance, the fact that every stock was genetically different did not influence the reaction to tristeza. All were susceptible. Many of the citrus relatives which are monoembryonic also display a high degree of uniformity when grown from seeds. Thus Clausena, Murraya, Triphasia, Citropsis, and other genera will, although possessing one zygotic embryo, produce progeny which appear phenotypically identical to the mother plant. In many of the world’s citrus areas, the importance of strain selection within scion varieties is well recognized and put into practice by the growers and nurserymen. This is perhaps more true in Japan than any other citrus area. Japan’s large Satsuma industry is divided into three fruit categories according to their maturity. The Wase group are early, the Nakates are mid-season, and the Futsu, or common group, are late. Within the Wase group are Okitsu, Miko, Miyagawa and dozens of others.

The other two groups may have a few less strains. Little differences in fruit yield, time of fruit maturity, hardiness, or local preference, etc., are of considerable importance to the grower and the industry without changing the variety. Spain has many strains of Clementines also selected for size, color, seediness, time of maturity, etc. Italy has made similar selections within the Willow Leaf mandarin. Strain selection within a variety has also been extensively practiced in California. Within the Washington navel variety are the Parent, Frost, Newhall, Tulegold, Bonanza, Thomson Imperial , Atwood, Fischer, Dream, Lane Late, etc. Even a wider range of strains may be found within the Eureka and Lisbon lemon varieties. These strain differences are for the growers’ preferential choice; the navels are all sold collectively as navels, and all the lemon strains are sold as lemons. Strain selection, or even varietal selection, while recognized within root stocks, is still generally not practiced, although there could be many advantages in doing so. Thus, for root stock purposes, the citrus industry has just nonchalantly considered a sweet orange a sweet orange, a sour orange a sour orange, or a Rough lemon a Rough lemon. On the other hand, root stocks like Troyer citrange, Sampson tangelo, Cleopatra mandarin, or Rangpur lime, are more specific; they imply one specific cultivar with no recognized variations, or strains, at least at the present time. A grower choosing one of these cultivars for root stocks doesn’t indicate to the nurseryman he wants citrange, tangelo, or mandarin as stock, because it is now a well known fact that there are other citranges, other tangelos, and other mandarins and, unless he specifies, he may get something else. Root stock trials at the Citrus Experiment Station, Riverside, indicate that there are performance differences between different strains of sweet oranges, sour orange, grapefruit, etc., that justify a clonal selection within that species to take advantage of greater yields, variations in tree size, better gummosis resistance, or better nematode resistance , and see root stock yield later in this manuscript. In these trials, and across several scion varieties, the performance of the Rubidoux sour and Brazilian sour was definitely superior to that on African sour and to a lesser extent on Paraguay sour.

Of course, this was before the incidence of Tristeza.The CRC #343 grapefruit provided better performance results than Duncan, McCarty, Camulos, and several others. The Koethen sweet orange in the Riverside trials appeared preferable to other sweet oranges. In Ventura County, the Olivelands sweet has frequently been preferred. The established performance of root stock cultivars warrants their perpetuation as root stock seed sources and their choice over untested and unproven sources which a nurseryman thinks may be just as good. In California, many citrus nurserymen have, or have had, their own individual seed source trees. Many had been chosen because of past good orchard performance in a specific area, or statewide. Some were the nurseryman’s selection,vertical plant rack others were obtained from the Citrus Research Center at Riverside. In either case, the nurseryman could offer the same clonal selection of root stock year after year with some confidence in its continued good performance. The immediate and rapid acceptance and success of Troyer citrange as a root stock in California almost had disastrous effects. With the increased plantings of young orchards, Troyer seed was in such demand that many nurserymen chose to grow seedlings for future seed sources. Prices were prohibitive for that time , as seed sold in most instances at a rate of ten cents each, or fifty cents per fruit, and three hundred dollars a standard field box . The author observed some of the fruit seedling sources and was able to identify a number of off type trees. The fact that some of the trees were off type didn’t seem to make any difference to the seed supplier. Fruit from the off type trees was harvested along with the fruit from the normal trees, either unknowingly or willingly , and sold as Troyers. While it is one thing to be able to identify any off type trees which can and should be discarded, another situation arises with off types in which there are no discernable visual differences from the mother tree. These may not be hybrid differences, but rather somatic variations which may be just as critical. The author will cite several examples, and keep in mind that this happened before the use of isozymes and other techniques were available for identification purposes. In the 1948 tristeza plantings at Baldwin Park, California , two sources of Bessie sweet orange were used as root stocks and half of the trees per root stock inoculated with tristeza. One of the sources of Bessie was an old source from the Citrus Variety Collection, Riverside. It served as the mother seed source for Bessie seedlings grown in 1924 for 1927 root stock plantings at Riverside . From out of these many seedlings, Dr. Webber had selected seedling #47 as being the largest and healthiest seedling of them all. This should have been the tip off. Seedling #47 was assigned CRC #1693 and also was placed in the Citrus Variety Collection, Riverside. In the Baldwin Park plantings, trees on the Bessie seedling CRC #1693 showed a reaction to tristeza and the trees on the Bessie #245 did not. Obviously the two seed sources were different. Many times my colleagues and I examined the two accessions and could not on leaf and fruit characteristics establish any differences. However, the largest and most vigorous seedling out of a progeny should never have been selected as true to type, and was probably either a gametic or somatic variant. Webber in 1924 also grew a fairly large number of sour orange seedlings for root stock purposes. Out of a population of 389 nursery seedlings, he selected 43 he considered as variants.

He propagated these variants and brought them into fruiting, and also budded the variants to a selected source of Washington navel. The author only wants to make reference to one of these variants. Just prior to my arrival in Riverside in 1946, one of the variants out of the 43 was selected for inclusion in the 1948 citrus root stock-tristeza trials at Baldwin Park . It was merely listed as a “sour orange variant.” The author was always interested in the correct identification of the root stock selections used in his experiments. Examination of this sour orange variant revealed leaf and fruit characters which appeared identical to Rough lemon. Showing foliage and fruit to my colleagues, including Dr. L. D. Batchelor, Dr. H. B. Frost, Dr. E. R. Parker and others, we could not differentiate between the variant and verified sources of Rough lemon. We removed the variant from the sour orange category and placed it in the lemon category, where we thought it more appropriately belonged. Upon being inoculated with tristeza, it proved to be equally, if not more, susceptible than those trees on sour orange. The standard Rough lemons showed no effect of the inoculation. Obviously, the variant was indeed a sour orange variant. How often does this happen? A number of the accessions in the Citrus Variety Collection date back to Dr. Webber’s “largest seedling” in the seedling nursery population. Many other accessions are more recent seedlings. Other seedling accessions may be more true to type than the Bessie, but do we know? Most of the accessions in the variety collection at Lindcove Field Station are seedlings, some of the accessions in the foundation block also. This is adequate cause for concern and alarm. Are they identical to the parent sources? A few years ago, Dr. John Carpenter of the USDA Date and Citrus Station at Indio, California, shocked a group of citrus nurserymen he was speaking to at Riverside, when he said many of you do not have Swingle citrumelo CPB 4475; what you have is a seedling source of Swingle citrumelo, and they are not the same. In recent years there has been considerable interest in C-32 and C-35 citranges . As information was developing and some credence given to the future use of these two root-stocks, some nurserymen obtained seed from Riverside and planted seedlings for future seed sources. The all important question remains, are these seedling sources identical to the original sources at Riverside, and to those tested at South Coast Field Station for tristeza resistance ?