Genetic and Cytological Studies with Camellia and Related Genera

liI.i 111112.8 2 5 2 5 1.0 li.lI. ,~ 11111 . . 1.0 ~IIIII~ 11111 . ~ ~II~ 2.2 I~ ~III~ 2.2 Ibl . 1:.1 .;, I~ ~ I~ ~ ~ :i lri.\l ..~ ~~ ... " I.;. " 1.1 .,;,a..:. .... 1.1 I.,a.", I I

111I[L25 111111.4 111111.6 111111.25 111111.4 111111.6

MICROCOPY RESOLUTION TEST CHART MICROCOPY RESOLUTION TEST CHART NATIONAL aURlAU OF S[,\NDAIlDS·196J,·A NAIIONAl nUHlAU or STANDARDS·I%3·A Genetic and Cytological Studies With Camellia and Related Genera

By William L. Ackerman

Technical Bulletin No. 1427

Agricultural Research Service UNITED STATES DEPARTMENT OF AGRICULTURE ACKNOWLEDGMENTS

The major portion of the work shown here on interspecific and intergeneric hybridization and chromosome numbers waS done in partial fulfillment of the requirements for the Ph. D. degree at the University of Maryland, College Park. The dissertation research was accomp1ished under the guidance of Delbert T. Morgan, Professor of Botany.

Trade names are used in this publication solely for the purpose of providing specific information. Mention of a trade name does not constitute a guarantee or \varranty of the product by the U.S. Depart ment of Agriculture 01' an endorsement by the Department over other products not mentioned. III CONTENTS

Introduction ______-- Page1

~aterials and methods ______.--______3 Parental specie.s of Camellia and related genera______3 Experimental procedures ______-______4 HybrirUzation ______..•______4

~ethods of pollination • ____ . ______4 Embryo culture _. __ .. __ . __ ._ •. ___.. ______7

General culture ______~ ______• __ . ______..______8 Morphological comparisons of parents and hybrids _... ___ .. _ .______9 Determination of pollen abortion ___ ._. ___ . ______. _____ .. ____. 9 Cytological methods . _ ... _____ . ______. ____ .__ .. ___ .. _.. ____ .__ 9 Results . ______--______. _.______12 Interspecific hybridization _.-_-_ .______12 Morphological comparisont1 ______. ______.. ______12 D(;termination of pollen abortion __ .______27 Intergeneric hybridization ... __ • ______•___ ..______39

~orphological comparisons ___ . ___ .. ______39 Determination of pollen abortion ____ . ______47 Species compatibility relationships ___ • ______-- 48 Chromosome numbers __ .______. ______.______51 Chromosome numbers of parents. ______•.___• __ .______51 Chromosome numbers of interspecific hybrids ______• __ . ______55 Chromosome numbers of intergeneric hybrid~; • _____• ______.___ 64 Chromosome morphology .______. ____ . ______• ______. ______69 Discussion and conclusions . __ ..__ .. _____ . ______.______. 78 Summary ______• ______.______86 Literature cited ._ • _"_" .. __ ... _ .•___ . ______.______89 Appendix ______• ______.______93

Washington, D.C. Issued September 1971

For 8."11c by the SUPCl'intcndcnl or Documents, U.S. Govcrnment Printing Omec \Vnshin).!ton, D.C. 20402-Pl'icc 70 cents. . IV Genetic and Cytological Studies With Camellia and Rela'ted Genera

By WILL1AM L. ACKERMAN, research hm·ti(ntltm-ist, Plant Science Rcsea1·ch Division, Ag1·icultural Resea1·ch Service

INTRODUCTION

Interest in the genus Camellia was during the latter part of the L. has been largely concerned 18th century. with species of economic signifi During Victorian times, camel cance. In the Orient, camellias lias were grown as conservatory are important as ornamentals, for plants in both Europe and North the oil of the seeds of certain ern United States and as garden species, and as the source of tea. specimens in Southeastern United In both Europe and the United States. Popularity of camellias States, camellias are important waned near the turn of the cen only as ornamentals. Introduction tury, and not until after World of species of Camellia into Eu War II did they again gain prom rope from the Orient is generally inence as ornamentals. Pres accepted to have been about 1740 ently, the many cultivars of C. ja and was closely associated with ponica and C. sasanqua Thunberg the tea industry, based on C. si are grown rather widely as hardy nensis (L.) O. Kuntze. A report evergreens in southern and by Meyer (1959) 1 indicates a Pacific coastal regions and, to a much earlier introduction into limited extent, in greenhouses in Europe by Portuguese traders the North. C. hiemalis Nakai, C. who brought C. iaponica L. to reticulcda.. Lindley, C. sal7tenensis Oporto, Portugal, about 1550. In Stapt ex Bean, and C. ve?·nalis troduction into the United States (Makino) Makino are less frequently cultivated species CamelUa belongs to the family 1 References to Li';erature Cited, p. 89, are herein il1dic>!t

MATERIALS AND METHODS

Parental Species of Gardens cooperative ornamental Camellia and Related exploration program. 3 J. L. Genera Creech, ARS, has collected 47 The Glenn Dale collection of camellia introductions during Camellia nnd species of related Lon g woo d Gardens-sponsored genera is composed of 'wild spe plant explorations in the Orient. cies and cultivated varieties im Plant breeders in the United ported during the last two dec States have con'~ributed other ades by the Plant Science Research species and varieties. Seventy-one Division, Agricultural Research introductions comprising 20 spe Service, U.S. Department of Agri cies of Camellia and four species culture, through plant explora of related genera were used as tions throughout much of the Ori parents. The sources of the par ent and, to a leRser extent, through exchange witb European botani "Longwood Gardens of Longwood cal gardens. A large number of Foundation, Inc., Kennett Square, Pa. these introductions have been the This cooperative exploration program has been responsible for 14 foreign result of the joint Agricultural plant explorations, including 8 to the Research Service-Longwood Orient since 1956. 4, TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE ents used are listed in table 1. tensive use as female parents. Voucher specimens of all intro Another factor that reduced the ductions included in the table, number of combinations was the except for C. taliensis and C. ten limitations in greenhouse space, 7tifiom, were submitted to the which curtailed the numbers of herbarium of the National Arbor parents and hybrid progenies etmn, Washington, D.C. that could be grown at one time, With the exception of outdoor M~thods of Pollination specimens of F'ranklinia alata maha and Ste7Va?'tia ovata, all of 'l'he blooming season of most the plants used as parents were species of Camellia grown under grown on greenhouse benches. greenhouse conditions at Glenn D;~ily temperature records have Dale extends from early October been maintained during the to late March, although a few blooming seaSon for the past 5 bloom sporadically through late years. The houses have been kept spring and summer. Flowers at night temperatures of 50 ° to were emasculated by cutting hori 56° F. during the blooming pe zontally around the flower bud riod. These temperatures are con with a razor blade to remove the sidered to be suitable for floral apex half of the petals and sta development. l~:l'\ns with attached anthers be fore dehiscence. To gain access to Experimental Procedures the pistil of semi double peony type flowers, we frequently had to Hybridization cut and remove considerably An effort was made to hybrid more of the petals. ize each species in as many dif Both fresh and stored pollen ferent combinations as possible, were used for making controlled including the reciprocal of each crosses. Where possible, fresh interspecific cross. However, sev poilen of the desired male parent eral factors reduced the number was collected and used the same of combinations from that first or tbe next day. In many cases, anticipated. Originally, there the flowering seasons of two spe were many varieties of Camellia ies to be hybridized did not over japon'ica, C. ?'usticana, and C. sas lap and pollen had to be collected anqua and only one or a very few in advance and stored for later plants of most of the other spe use, In these cases, the pollen was cies. Some of the more recent dried at room temperature for 24 acquisitions were represented hours and then stored in cotton only by small plants bearing lim stoppered glass via''3 over anhy ited numbers of blossoms. Al drous calcium chloride at 45° F. though these plants could be used Lengths of stOl"age of pollen var as male parents, limited produc ied from less than 1 week to sev tion of flowers precluded their ex eral months. Under the storage STUDIES WITH OAiHELLIA AND RELATED GENERA 5

TABLE 1.-Invent01·Y of species of Camellia and 'related genem used in hybr-idization studies

P.I. or B NO.1 Scientific name Source

132476.", . C. Iralema Hanee.•..•••••• ____ ._.••___ _ Kuling, China. 251534 .. C. granthamiana Sealy____ • _. ___ •____ . __ _ Victoria, Hongkong. B56!J95 •._. t..:. ; :emalis Nakai, 'Shishi-Gashira'______• A. Fendig, Brunswick, Ga. B56996..__ C. hiemalis 'Bill Wylam' ______•• Do. 229973 '. C. IIol!gkollg8nsis Seem. _.••__ •. _. ___ . _._ Victoria, Hongkong.

B56272_.__ C. jnponicn L., 'Ville de Nantes'_ok. _____ • National Arboretum, Washington, D.C. 22G'L09 C. japonica No.6 ...... __ •• _... _. ______Angyo, Japan. 227063. C. ja1)OHica 'Bon-Shirotama'____ .••___ .._. Rokuji-Kai, Japan. 228024 C. ja1)onica "fsubaki'_ ..______.• Okayama-shi, Japan. 2P..)278 C.japonica. _ _.,._ ..._. ______• ______Aichi-ken, Japan.

231686 .• C. japonica 'Hasumi-shiro'______.. . __ .. Chiba, Japan. 231687 ... C. ja1)onica 'Komyotai' ___ ..._•.. ____ . _.• Do. 231689 • C. japonica 'Moshio' __ . . •.. _ . " ___ . __ _ Do. 231690. C. ja1)onica __ • •• _. _•• ______• __ •. __ _ Do.

231694 C. japonica 'Utamakura' _Ok' ______Do.

2316!l5 • C. japonicCL 'Yuki botan' ___ ._. ____ ._. __ _ Do. 231858 C. j'aponicCL 'Beni botan'_.._. _ ._. _____ .• __ Do. 231859 C. japonica 'Kanyo-tai' ___ .• ___ • _. ______Do. 238725 C. japonica 'Saudade de Martins Blanco'__ _ Porto, Portugal. 274530 C.japonicaNo.825 ..•.•.•••. __ •• _._ .• __ Kyushu, Japan.

274797 ... . C. japonica No. 870_. _. ___ .•______• _•. __ Do. 274799 ... . C.japonica No. 872. ______.... ______._ Do. 275054 C. japonica No. 913 .. __ ._._ .." •. __ •• __ _ Kita-Ibaraki, Japan. 275512 C.japoll1·ca D ____ ..._.. __ . ___ .• _...... Miyako, Iwate Prefecture, Japan. 309001._. C. japonica. 'Le Lys' ..._•. ____ • __ • ______. Botanic Garden, Nantes, France.

319283._ .• C. japonica K-20___ ....._____ .... _...... Cheju Island, Korea. 252062 .•. _ C. kissi WaIL.___ ...... ___ .. _.••.•..••• Katmandu, Nepal. 252064 .. C. kissi. _ _ ...... _...•.•..•" .. _ • _ • _ Do. 226756 . C. lulchuensis ~l'. Ito_ ...... _.•• _ Kunigami, Okinawa. 226704 .. C. miyagii (Koidz.) Makino & Nemoto____ _ Genka, Okinawa.

231057 .•.• C. miyagii... _•. . ___ ._._. ___ ._...... __ ._ Ryukyu, Okinawa. 162561 •.. C. oleifera Abel...... _. __ ... __ • '"_'" . _ Nanking, China. 235500 . C. olei/cra . _ • __ _ ..... _...... Oshima Island, Japan. B58296. __ _ C. 7lilardii Cohen-Stuart var. 7li/ardii. ...._ R. Cutter, Berkeley, Calif. B51685 ___ _ C. reticula/a Lind!. 'Crimson Robe' _ .•. _••. W. Gotelli, East Orange, N.J. See footnote at end of table. 6 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

TABLE l.-Invent01·y of species of Camellia and 'related genem u,sed in hyb'ridization st~tdies-Continued

P.I. or BNo.l Scientific name Source

B51686___ _ C. reliculala 'Great Shot Silk' ______Do. B51689 ___ _ C. reliculala 'Chang's Temple' ______• ____ _ Do. B51690___ _ C. reliculala 'Lurge Cornelian' ______Do. B51694 ___ _ C. reliculala 'Lion's Head'______Do. B58619 ___ _ C. rosaejlora HookeL ______Berkeley, Calif.

228181- __ _ C. ruslicuna Honda, 'Yoshida'______. ___ _ Niigata, Japan. 228188 ___ _ C. ruslicana 'Hatano' ______Do. 228190 ___ _ C. ruslicana 'Koshiji' ______.. __ _ Do. 233642 __.__ C. rUJticana B. White plena______Tokyo, Japan. 310320___ _ C. salici/olia Champ. ex .Benth. ______Royal Botanic Gardens, Kew, England.

243862 ___ _ C. saluenensis Stapf in Bean, No. 6093 ___ _ Cornwall, England. 227624 ___ _ C. sasanqua Thumb. 'Kokinran'______Yamamato, Japan. 228025 ___ _ C. sasanqlla 'Sazanka' ______Kotohira-eho, Japan. 235568 ___ _ C. sasanqua______Rokko-san, Japan. 237854- __ _ C. sasanqua No. 79L______Yakushima, Japan.

277763 ___ _ C. sasanqua 'Narumi-gata' ______Winch~ster, England. 319284 ___ _ C. sasallqua______• ______. ____ _ Kurume City, Japan. 319285___ _ C. sasanqua 'Onishiki' ______• ___ _ Do. 235570 ___ _ C. sinensis (L.) Kuntze, 'Tama-midori' ___ _ National Tea Experiment Station, Kanaya, Japan. 235572 ___ _ C. sinensis Y-2______.._ Do.

235573 ___ _ C. sinensis Z-L______Do. 304404- __ _ C. sinensis No. L ______Oruziya, U.S.S.R. 304405" __ _ C. sinensis No. 2______Do. 31647L __ _ C. sinensis 'Beni-fuji' ______• ______Kanaya, Japan. 316472 ___ _ C. sinensis 'Beni-Homare' ______Do.

3Hi·!7:L __ _ C. sinensis 'Makinowara Wase' ______Do. 316476 ___ _ C. sinensis 'Yamatomidori' ______Do. 316477 __ __ C. sinensis ______Do. 316478 ___ _ C. sinensis ______• __ _. ______Do. B57043 ___ _ C. laliensis (W. W. Smith) Melchior ______C. Parks, Los Angeles, Calif.

B57048____ C. lenuijlora (Hay.) Cohen-Stuart. ______Do. 271683 ..._ C. tsaii HlL_ .... _ .• _". ______• ___ "_"_ Hillier & Sons, England. 69037_•_.• Franklinia alatamalla Bartr. ex Marsh. ___ .. S. Baxter, Philadelphia, Pa.

See footnote at end of table. STUDIES WITH GAMELUA AND RELATED GENERA 7

TABLE l.-lnventory of species of Camellia and related. genera u.sed in hybridization studies-Continued

P.I. or B No.1 Scientific name Source

104210____ Slewarlia ovala (Cav.) Weatherby ______Orleans, France. 230368____ Tulcheria spectabilis (Champ.) Dunn______Victoria, Hongkong. 229881. ___ 7'ulcheria virgata (Koidz.) NakaL ______Genka Haneji-son, Okinawa.

1 P.I. refers to the accession number assigned to foreign introductions by thePlant Sci ence Research Division, Agricultural Research Service, U.S. Department of Agriculture. B numbers are code numbers assigned to importations under postentry quarantine permit by the U.S. Plant Introduction Station, Glenn Dale, Md. conditions used, pollen viability Embryo Culture had decreased somewhat at the Seed capsules matured in late end of 2 months but not enough summer and early fall. Interspe to seriously affect hybridizing re cific and intergeneric crosses fre sults. Fresh pollen was most eas quently resulted in highly defec ily transferred by removing sta tive seeds. Many of these seeds mens from the male parent with were small and deformed and forceps and using the anther as contained incompletely developed the applicator. Stored pollen was embryos with disorganized tissue applied with a small camel's hair development. Most of these seeds brush. did not germinate when the rou Most controlled crosses were tine moist sphagnum peat method made in greenhouses screened was used for the germination of again:;t insects. Therefore, emas seeds of Camellia. Embryo cul culated flowers did not have to be ture methods were investigated protected from undesired pollina and found more satisfactory and tion. The flowers of Franklinia were adopted as standard proce and Stewa.1·tia, grown outdoors, dure for germinating all seeds were bagged with nylon netting from interspecific and interge for protection from pollination by neric crosses. insects. Approximately 2 months The seeds were prepared for after pollination, successful devel embryo culture by carefully re opment of capsules and failures moving both the._outer and inner were recOl'ded. All capsules were seedcoats with a sharp knife. The individually enclosed in bags of outer seedcoat normally was very nylon netting to prevent acciden hard, and frequently a small area tal loss. at the side of the seed had to be 8 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE scraped with the knife blade held transfer room could be con at an oblique angle until penetra structed especially for embryo tion was mad~. The size and loca culture. tion of the embryo could then be determined. In a normal seed, the The culture bottles were then embryo and associated endosperm placed under fluorescent lights in fill the entire space within the a room with a near constant tem seedcoat. In seeds of wide perature of 68° F. The cultured crosses, however, the embryos embryos were maintained under and endosperm occupy only 1/10 these conditions until they had to 1/2 of the space. After the adequate root development and outer seedcoat was removed, the the first set of true leaves was at least partly unfolded. r~st of the seed was placed in water for approximately 2 hours. The inner seedcoat softened and, General Culture in most cases, was not difficult to After the seedlings were re remove. The embryos and asso moved from culture bottles, they ciated endosperm were then were transplanted to 2-inch pots transferred under aseptic condi of milled sphagnum. The pots were tions to the sterilizing solution set on gra1Tel in a greenhouse and gently agitated for 5 minutes bench and subjected to automatic before being transferred with mist for 1 minute each hour. sterilized forceps to culture bot After 2 weeks, the pots were tles. plunged in peat moss in an open Several sterilizing solutions greenhouse bench until growth were tried, but the most success required transfer to larger pots. ful was 5 percent sodium hypo The seedlings were then contin chlorite (Clorox). A drop of ued in open gravel benches. Dur Tween 20 in 100 cc. of Clorox ing the first several seasons, the solution reduced surface tension plants were transplanted to steri and promoted wetting with the lized soil. Growth was not always sterilant. satisfactory and other cultural methods were investigated. Tests The nutrient agar used was showed that superior growth re that described by Tukey (1934) sulted from continuing the plants and later adapted for embryo cul in milled sphagnum moss, ferti ture of Camellia by Lammerts lized once a month with a (1950). Transfers of the embryos 20-20-20 liquid fertilizer. The were made during the first two night period from September to seasons in a small homemade, March was broken with incandes glass-topped chamber, fitted with cent light from 10 :00 p.m. until hand holes. Later, a larger, rigid 2 :00 a.m. to help accelerate vege plastic chamber was used until a tative growth of young plants. STUDIES WITH CAMELLIA AND RELATED GENERA 9

Morphological Comparisons of parents after the pollen was Parents and Hybrids stained with acetocarmine. A I-millimeter square grid was Morphological observations scribed with a diamond-pointed were made for evidence of inter pencil on microscope slides to fa specific and intergeneric hybrid cilitate making pollen counts. Pol ity. Early observations of vegeta len from parental species and the tive characters disclosed that hybrids was stained with a 2 ..per some were strongly dominant in cent acetocarmine solution. The the progenies. These characters slides were allowed to stand at could thus serve as useful mar room temperature for approxi kers for hybridity when the par mately 10 minutes to allow for ticular species was used as a male stain penetration before observa parent. Taxonomic descriptions tion at a magnification of 125. were recorded as presented by Pollen abortion counts were based Sealy (1958) for the species used on 1,000 grains for each test if in successful crosses. The taxo both normal (filled) and aborted nomic characters of the parental grains were observed. If, how species were compared with Sea ever no normal grains were ob ly's descriptions and summarized in the first 500 examined, in appendix table 21. We then serv~d that hybrid was classified as com compared pairs of parental spe pletely male sterile. cies for 'differences in distinct, eas Two classes of pollen, stainable ily identifiable characters and ob and unstainable, were clearly evi served these specific characters dent. The stainable grains were among the Fl hybrid populations. smooth-surfaced in contrast to F \ plants possessing one or more the shrunken, wrinkled, and un vegetative or floral characters of stainable aborted pollen. In most the male parent and F\ plants in cases, shape alone would have termediate in appearance between been sufficient to distinguish the the two parents were considered two classes of pollen, but the as valid hybrids. staining facilitated classification.

Determination of Pollen Cytological Methods Abortion Chromosome studies were Many of the hybrids were de made on both hybrids and species. veloped from rather wide inter Root tips were collected from ma specific or intergeneric crosses. ture plants, from cuttings rooted Thus, reduced fertility in many under intermittent mist, and of the hybrids was anticipated. from seedlings shortly after re Accordingly, percentages of filled moval from embryo culture. and unfilled pollen were deter After collection, root tips were mined for the hybrids and the placed immediately in vials of 10 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE tapwater. A prefixation treat were made permanent, using the m~nt of 0.2-percent aqueous solu method of Zeilinga and Kroon tion of colchicine for 2 hours at (1965) . room temperature (68 ° F.) was Seasonal variations in mitotic used to shorten and spread the activity in the root tips were ob chromosomes. Prefixation for 3 served during 5 consecutive hours was used on polyploid years. The lowest degree of mi clones to facilitate counting, but totic activity was during the sum longer treatment caused sticki mer months, especially when the ness and clumping of chromo night temperature in the green somes. A saturated aqueous solu house was above 68° F. A gradual tion of paradichlorobenzene for 3 increase in activity began in early hours at 45° F. was a good alter autumn and reached a peak dur nate prefixation treatment but ing December and January. This not as consistent in results as the peak was followed by a gradual 0.2-percent solution of colchicine. decrease toward spring and Root tips were fixed for 6 to 8 reached a low point again in sum hours at room temperature in mer. Accordingly, after variation three parts of 95 percent ethyl al in mitotic activity became evi cohol and one part of glacial dent, most root tips were col acetic acid and then stored at lected during the winter months. 32°F. Acetocarmine proved supe rior to acetoorcein and the Feul The use of chromosome mor gen technique for staining the root phology as a method for distin tips and therefore was used rou guishing species and hybrids was tinely. Before the preparation of investigated. Chromosomes were acetocarmine smears, the root measured by taking photomicro tips were softened in one normal graphs, projecting the negative hydrochloric acid solution for 5 at a magnification of 2300 X, and minutes at 68° F. to obtain a then tracing the enlarged chro good spread of cells. This was fol mosomes on paper. Also, glossy lowed by transfer to 70 percent prints were made at this same ethyl alcohol for the same time at magnification for comparative the same temperature (Darling purposes. Chromosome configura ton and La Cour, 1960). Good tions presented problems for spreads were indispensable be accurate measurements. Soft cop cause of the high number of chro per wire was used to duplicate mosomes eoncentrated in rather each chromosome configuration. small cells. Photomicrographs of Chromosome length and centro good temporary slides were taken mere position were marked on the at 1525 X. Some of these slides wire with a sharp knife. The wire STUDIES WITH CAMELLIA AND RELATED GENERA 11 was then straightened and meas chromosome. Chromosomes were ured with a vernier caliper. classified as subterminal, subme Few cells of anyone species dian, and median where the F% studied showed all chromosomes values fell within 0 to 30, 30 to in n position where they could be 45, and 45 to 50, respectively. followed accurately. A tentative Bammi (1965) and Settle (1967) karyotype, showing chromosomes used long arm to short arm ind in sequence of descending order ices. Kawano and others (1967) of total chromosomal length, was and Murin (1962) used short arm first constructed for each species to long arm indices. Ourecky from the best cell available. The (1966) used double indices of chromosomes of other cells were long arm to total length and short likewise placed in order of chro arm to total length. Other investi mosomal length as closely as pos gators, including Nakajima sible and then individual chromo (1963), Jagathesan and Sreeniva somes were matched with those san (1967), and Mallick and of the karyotype. This method Sharma (1966), used alphabetical was found convenient and re designations based on narrative duced the possibility of having descriptions. The resulting classi the same chromosome included fications appear arbitrary at best. twice in the karyotype to the ex For example, Bammi (1965) clas clusion of some other chromo sified a chromosome as submedian some. Individual chromosomes with a long arm to short arm from six cells were selected to index of 3.7. If converted to the construct the final karyotype for F % used by Shindo and Kame each species. moto (1963), the value .is 21, Methods of classification of which is well within their classi chromosomes as median, subme fication for subterminal chromo dian or subterminal on the basis somes. of centromere position were in For the purposes of this study, vestigated. A general lack of uni long arm to short arm indices formity of methods described in were used and chromosomes clas the literature results in consider sified as median, submedian, and able overlapping between the var subterminal where index values ious types of chromosomes. were 1.00 to 1.20, 1.21 to 2.40, Shindo and Kamemoto (1963) and 2.41 upward, respectively. and Kaneko (1966) used mean This method of classification re F% values derived by taking the sulted in separations very closely percentages of the short arm comparable with the F 70 values length over the total length of a of Shindo and Kamemoto (1963). 12 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE RESUlTS

Interspecific Poor seed is very common among Hybridization interspecific and intergeneric crosses because incompatibility Sufficient blooms were availa factors frequently exist where ble on 16 Camellia species for use large differences are involved in as female parents, and 20 species the genetic complexes of the pa were used as male parents. A rental species. Many of these seeds total of 8,741 controlled pollina either failed to germinate in cul tions were made. These repre ture or developed abnormally and sented 219 interspecific combina subsequently died. Both normal tions, incl uding reciprocal and abnormal development in cul crosses. A total of 459 hybrid ture are illustrated in figure 1. plants from 5,862 pollinations in 106 combinations were obtained. Morphological Comparisons In table 2, those interspecific com One of the first indications of binations that resulted in hybrid validity of a young hybrid plant is plants are listed, along with the the comparison of its vegetative numbers of pollinations and the morphological characters with numbers and percentages per pol.. those of its parents. Because most lination of seed capsules pro seedlings do not flower before duced, of seeds started in culture, they are :>. to 4 years old, we had and of hybrid plants. The inter to rely heavily on vegetative char specific crosses attempted, but acters rather than floral and seed which proved unsuccessful in the characters. Characters selected establishment of hybrid plants, were those described by Sealy are listed in appendix table 20. (1958) and centered mostly on Chi-square values, based on the twigs and young shoots, leaves, number of pollinations and hy and terminal vegetative buds. The brids obtained from rep.; ;'rocal chief characters used are sum crosses, are listed in table 3. Six of marized for each species in ap·· the 54 reciprocal crosses were sig pendix table 21. nificantly or highly significantly Validation of 211 interspecific different. hybrids, based on the characters Seed capsule development after described in table 21 and used pollination is no assurance of specifically in the discussion and eventually obtaining hybrid analysis in this study, is given in plants. In a substantial number appendix table 22. Those seed of crosses, seed capsules devel lings that could not be proved to oped to apparent maturity, but be valid hybrids on the basis of contained either no seeds or seeds morphological comparisons, on with partly developed embryos. pollen abortion, or on chromosome TABLE 2.-lnterspeci/ic crosses in Camellia 1'esulting in hyb-rid plants

Interspecific crosses Polli Seed capsules Seeds started Hybrid plants nations produced in culture rn 8 c::: Percent per Percent per Percent per t:I ..... Number Number pollinatioll Number pollination Number pollination t::l C. iralema X C. japollica ___" .. 49 1 2.0 1 2.0 1 2.0 rn ReciprocaL" 6 =a 105 5.7 16 15.2 10 9.5 ..... C. ira!ema X C. lulchuclIsis. __ 79 1 1.3 1 1.3 1 1.3 8 ReciprocaL. 50 3 6.0 3 6.0 1 2.0 t:z= C. iralema X C. reliculala_ 218 2 9.1 2 9.1 1 .4 ReciprocaL __ ~ 24 2 8.3 4 16.7 3 12.5 ~ ttl C. hiemlilis XC. kissi__ ._ . 16 2 12.5 3 18.8 1 6.3 ReciprocaL__ .. 35 11 31.4 8 22.9 1 2.9 § C. hiemalis X C. miyagii. _. 30 5 16.7 5 16.7 4 13.3 > Reciprocal __ 32 15 46.9 16 50.0 8 25.0 Z C. Memalis X C. oleiiera 9 3 33.3 2 22.2 1 11.1 t:I ReciprocaL_ 48 8 16.7 9 18.8 4 8.3 ~ t::l t"4 C. hiemalis XC. sasanqua ___ •.•...... __ .... _••• _. 38 8 21.0 10 26.3 8 21.0 > 8 ReciprocaL_ '. ",,,,_, ____ , ___ _ 50 10 20.0 12 24.0 6 12.0 t::l C. hOllgkollgCllsis XC. retlculala ______• _____• ' __ _ 20 9 45.0 9 45.0 3 15.0 t:I C. hOllgkollgensis X C. rusticana ___ , _. _• ______•• __ __ 41 7 17.1 13 31.7 5 12.2 "-l ReciprocaL __ . _. . . ___ " _. ______• ______t::l 60 1 1.7 4 6.6 2 3.3 Z C. ILOllgkollgellsis XC. saliciiolia ______54 2 3.7 2 3.7 1 1.9 t::l ~> C. hOIlgkOllgclIsis X C. sinensis... _•______"______67 2 2.9 5 7.4 2 2.9 C. japollica XC. granthamiana ______75 1 1.3 4 5.3 3 4.0 C.japollica XC. hiemalis ______,, ____ _ 48 1 2.1 1 2.1 1 2.1 C. ILOllgkollgensis ~ japonica XC. ______111 2 1.8 8 7.2 7 6.3 CO ...... TABLE 2.-lnterspecific crosses in Camellia 1'esulting in hybrid plants-Continued ,;:..

8 Interspecific crosses PolJi Seed capsules Seeds started Hybrid plants t.".l nations produced in culture o =Z o..... Percenl per Percent per Percent per > Number Number pollinalion Number pollination Number pollination t"' C.japonica XC. kissi ____ ,_ • ______44 2 4.5 5 11.4 2 4.5 ttl ReciprocaL_____ . __ ,_. ______•______59 7 11.9 10 16.9 5 8.5 q C. japonica lutchuensis______~ XC. 551 28 5.1 86 15.6 62 11.2 t.".l ReciprocaL__.______" _____ . ______8 __ ."' 245 4 1.6 4 1.6 2 .8 ..... C.japonica X C.miyayii_. ______20 1 5.0 3 15.0 2 10.0 Z C. japonica XC. pitardii var. pitardii ______22 1 4.5 2 9.1 2 9.1 .... ReciprocaL_. _ _ ,,______"- ___ .. ______"'"~ 24 1 4.2 1 4.2 1 4.2 .-'1 C . .iaponica X C. reliculala____ ... ______150 4 2.7 6 4.0 5 3.3 q ReciprocaL__ • _.• _ ... ______69 6 8.7 7 10.1 7 10.1 rn C.japonica XC. rosae/lora___ ..____ ._. __ . ___ ._ 38 1 2.6 1 2.6 1 2.6 tj t.:zj C. japonica X C. rmUcana____ _.. ___ " .. __ _"- 20 3 15.0 4 20.0 4 20.0 "t! ReciprocaL_____ ••. __ . ______.. ___ . ______57 8 14.0 17 20.0 13 22.8 !"'l o C. japonica X C. saluenenb'is.. _ ___ . _ _.,_ 41 12 29.3 15 36.6 12 29.3 I:I;j ReciprocaL___ ... _... ______.. - ______13 1 7.7 2 15.4 1 7.7 > C.japonica XC. sinensis______. ______. ___ _ 44 4 9.1 10 22.7 3 6.8 C) ::a ReciprocaL__ . _ . _____ . _. ______.. 4 9.8 1 2.4 ..... 41 5 12.2 o C. japonica X C. laliensis___ . _ .. __ _ . _ • _ . .._ 30 1 3.3 1 3.3 1 3.3 q C.japonica X C./saii___ __ •• _____ .... '.,. _ 86 6 7.0 11 12.8 6 7.0 t"' 8 q C. kissi X C. miyayii _ 52 15 28.8 13 25.0 8 15.4 ::a ReciprocaL_ 37 3 8.1 2 5.4 2 5.4 t.".l

C. kis.~iX C. oleiJera____ . 53 13 24.5 15 28.3 9 16.9 ReciprocaL._ 42 6 14.3 6 14.3 3 7.1 C. kissi X C. reliculala __ 56 5 8.9 3 5.4 1 1.8 C. kissi X C. rosaej/ora.. _. 36 5 13.9 6 16.7 4 11.1

C. kissi XC. rusticana ... . _. ___ • ______•___ ... ___ 35 2 5.6 2 5.6 1 2.8 ReciprocaL __ . - __ • ____ •• ____ . ______707 1 3.7 2 7.4 1 3.7 CIl >-3 C. kil;si XC. salicifolia_. ______. __ •• ___ ._ 63 21 33.3 10 15.9 2 3.2 c: C. kis.~iX C. sasallqua. ______• __ 46 21 45.7 26 56.5 14 30.4 t:::! ReciprocaL_ • 33 .... ______12 36.4 12 36.4 7 21.2 t".l CIl

~ C. kissi X C. sinensis. 44 13 29.5 9 20.5 4 9.1 .... ReciprocaL______•___ . ______45 8 17.8 6 13.3 4 8.9 >-3 C. lutchllcnsis X C. pilardii var pitardiL______50 1 2.0 2 4.0 1 2.0 == ReciprocaL_ .•.. ____ • ___ .______27 2 7.4 3 11.1 2 7.4 ~ C.lulchucn5is X C. ruslicana______67 1 1.5 1 1.5 1 1.5 ReciprocaL__ _ ... __ ._. ______280 12 4.3 41 14.6 24 8.6 ~ t'< C. lulchuensis XC. salicifolia______42 10 23.8 6 14.3 4 9.5 :;: C. miyagii XC. fralerna ____ . ______17 3 17.6 1 5.9 1 5.9 > C. miyagii XC. hongkongensis______. ______44 5 11.4 2 4.5 1 2.3 Z C.11Iiyagii XC. lutchllcnsis______112 6 5.4 8 7.2 4 3.6 t:::! C. miyagii X C. oleifera ______33 16 48.5 21 63.6 13 39.4 ~ ReciprocaL ______.______19 t".l 2 10.6 4 t"' 21.1 3 15.8 > C. miyagii xC. reticulala______37 >-3 4 10.8 3 8.1 1 2.7 l".1 t:I C. miyagii X C. salicifolia______69 8 11.6 5 7.2 2 2.9 C. miyagii XC. sasanqua______40 18 45.0 15 37.5 9 22.5 o ReciprocaL______33 l".1 2 6.1 6 18.2 4 12.1 Z C.oleifera XC. salicifolia______43 4 9.3 2 4.7 1 2.3 l".1 ~ C. oleifera X C. sasanqua______29 > 14 48.3 27 93.1 18 62.1 ReciprocaL______31 7 22.6 26 83.9 16 51.6 C.oleifera XC. sinensis______63 6 9.5 4 6.3 2 3.1 ReciprocaL _____ • ______.______64 ..... 4" 6.3 8 12.5 5 7.8 en TABLE 2.-Interspecific crosses in Camellia resulting in hybrid plants-Continued J-l ~

...; Interspecific crosses PolIi- Seed capsules Seeds started Hybrid plants t;rj nations produced in culture (") ~ Z (j.... Percent per Percent per Percent per > Number Number pollination Number pollination Number pollination t:-< C. pitardii var. pitardii XC. Iraterna ______21 2 9.5 7 33.3 2 9.5 t:C C. pilardii var. pitardii X C. granthamiana ______25 1 4.0 2 8.0 2 8.0 c:: t:-< t:-< C. pitardii var. pilardii X C. reticulata ______25 1 4.0 2 8.0 1 4.0 t;rj Iteciprocal ______1-3 42 4 9.5 5 11.9 4 9.5 .... Z C. pitardii var. pilardii X C. saluenensis______20 1 5.0 2 10.0 1 5.0 IteciprocaL___ . ______. ____ • ______.... 19 2 10.5 3 15.8 1 5.2 ~ 1>:1 C. pilardii var. pilardii X C. tsaiL ______76 8 10.5 11 14.5 4 5.3 ,:'I C. reticulala XC. granlhamiana______39 1 2.5 2 5.1 1 2.5 c:: rn c. reticulata XC. rosaejlora______35 1 2.9 1 2.9 1 2.9 t:l C. reliculala XC. rusticana______12 1 8.3 1 8.3 1 8.3 t;rj C. reliculala XC. saluenensis______. ____ 24 2 8.3 5 20.8 3 12.5 "C ReciprocaL_. ______28 1 3.6 2 7.1 2 7.1 !-3 C. rosaejlora XC. Iralema______0 36 1 2.8 2 5.6 2 5.6 I:I;j C. TUl!ticana XC. lralerna ______162 2 1.2 6 3.7 3 1.9 /;')> C. ruslicana X C. oleilera______t:tI 63 8 12.7 7 11.1 3 4.8 o-t C. TUslicana X C. rosaejlora__• ______46 5 10.9 8 17.4 3 6.5 0 C. ruslicana XC. salicilolia______c:: 53 4 7.5 8 15.1 5 9.4 t:-< C. ruslicana XC. saluenensis______13 1 7.7 1 7.'7 1 7.7 1-3 C. saluenensis XC. Iralema ______c:: 24 1 4.2 1 4.2 1 4.2 ~ l.:rJ C. salttenensis X C. granlhamiana______21 1 4.8 1 4.8 1 4.8 C. saluenensis X C. hongkongensis___ . ______12 1 8.3 " 16.6 1 8.3 C. saluenensis X C. lutchuensis__ .. ______.. ____ .. ___ 19 1 5.2 1 5.2 1 5.2 C. salucnensis X C. rosaeJlora_. ____•_ .. _.. ______24 2 8.3 3 12.5 1 4.2 C. salueuensi.~XC. taliensis ______.. 23 1 4.3 2 8.7 1 4.3

C'l C. sasanqua XC. granthamiana______60 1 1.7 2 3.4 2 3.4 ...; C. sasanqua X C.japonica ___ • ______46 1 2.2 3 6.5 3 6.5 C t:1 C. sasanqua X C. reticulata ..... _____.. _.. __ • ______•_ 88 2 2.3 5 5.7 3 3.4 .... C. sasallqua XC. tenuifiora__ .. ___._. ______.... 16 1 6.2 2 12.5 2 12.5 trJ C/.l C. sinensis X C.fraierna ____ • _. ______• __ .____ 32 7 21.9 6 18.8 1 3.1 :a.... C. sinensis X C. hiemalis______•. _____ ••• __. ____ •_ 66 14 21.2 8 12.1 5 7.6 1-3 II: C. sinensis XC. miyagii. ____ ..... _____ • ___• _____ ._ 58 5 8.6 18 31.0 10 17.2 C. sinensis XC. rosaeiiora ______51 ... 5 9.8 5 9.8 3 5.9 ~ C. sinensis XC. salicifolia______49 3 6.1 1 2.0 1 2.0 l:: C. sinensis XC. sasanqua_ ___ •_____ • ______.______52 19.2 10 19.2 7 13.5 I;Ij 10 !::"I t" ~- > Z t:1 ~ trJ t"' > 1-3 trJ t:1

c<.l trJ Z trJ >~

...... ~ 18 T,ECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

TABLE 3.-Chi-squ.o/re values for 1'ecipr'ocal inter'specific C1'osses where one or both r'ecipr'ocals prod1wed hybrid Camellias 1

Polli- Hybrid Chi- Interspecific crosses nations plants square value

Number Number C. fraterna XC. japonicu,______49 1 2.02 ReciprocaL ______.. ______.. ______105 10 C. frlltema XC. lutchuensis______79 1 .11 ReciprocaL ______50 1 C.fratema XC. reti"ulata ______218 1 19.29" ReciprocaL______24 3

C. hiemalis XC. kissi______16 ReciprocaL ______35 .34 C. hiemalis XC. miyagiL______30 ReciprocaL ______.______32 1.35 C. hiemalis X C. oleifera______9 ReciprocaL ______48 11 .07 C. hiemalis X C. sasanqua______38 Reciprocal______50 1.32 C. hongkongensis XC. reticulata______20 Reciprocal______19 3.09 C. hongkongensis XC. rusticant'l______41 Reciprocal______60 il 2.97 C. japonica XC. granthamiana______75 Reciprocal______11 .46 C. japonica X hiemalis______40 Reciprocal______13 .33 C. japonica XC. hongkongensis______111 Reciprocal______18 II 1.20 OJ C. japonica X C. kissi.______44 Reciprocal______59 .61 C. japo1lica XC. It!lchuensis______551 ReciprocaL ______245 24.98-- C. japonica XC. miyagiL______20 1 ReciprocaL______21 6~ 2.21

C. japonica XC. pitardii var. pitardii.______22 2 .46 Reciprocal. ______24 1 C. japonica XC. reticulata______150 5 4.23- Reciprocal______69 7 C. japonica X C. TUsticama______20 4 .07 Reciprocal______57 13

See footnotes at end of table. STUDIES WITH CAMELLIA AND RELATED GENERA 19

TABLE 3.-Chi-squa?·e values fO?' ?'ecip1'ocal inte'rspecific crosses whe?'e one or both ?'ecip1'ocals produced hybrid Camellias I-Continued

Polli Hybrid Chi Interspecific crosses nations plants square value

C.japonica XC. saluenensis______41 Reciprocal______13 2.51 C. japonica xC. sinensis______44 Reciprocal ____ .. ______41 .91 C. kissi X C. miyagii.______52 Rer:iprocaL ______.. ______.______37 2.16

C. kissi X C. oleifera______53 Reciprocal______42 2.06 C. kissi XC. reticulala______56 Reciprocal_ .. ______._.______18 .33 C. kissi XC. rosaefiora ______• ______36 Reciprocal______34 4.01·

C. kissi XC. rusticana______35 ReciprocaL _'" ______.______27 .04 C. kissi X C. sasanqua______46 Reciprocal.. _.. ___ .. ______33 .84 C. kissi X C. sinensis ___ .. ______44 Reciprocal______45 .01

C. 11llchuensis X C. pilardii var. pitardii. ______50 Reciprocal ______.______27 1.02 C. lulchllellsis XC. TIlsticana______67 ReciprocaL. ______.______280 C. 'miyagii X C. Jralerna______17 Reciprocal. __ '" __ . ______49 2.93 C. miyagii XC. hongkongensis______44 q Reciprocal______21 .48 C. miyagii X C. IUlchuensis______112 ReciprocaL ..... _. __ ._. ___ .______49 1.85 C. miyagii XC. oleifera______33 Reciprocal______19 1~ 1 3,15 C. miyagii XC. sasanqua______40 9 1.33 ReciprocaL ___ . _____ . ______33 4 C.oleifera X C. sa.~anqua______29 18 .67 ReciprocaL ______31 16 C.oleifera XC. sinensis______63 2 1.31 Reciprocal______64 5

See footnotes at end of table. 20 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

TABLE 3,-Chi-squa1'e val~tes !o1'1'ecip1'ocal interspecific C1'osses whet'e one 01' both 1'ecip1'ocals p1'oduced hyb1'id Camellias l-Continued

Polli Hybrid Chi Interspecific crosses nations plants square value

Number Number C. pitardii var. pitardii XC. fraterna______21 2 2.49 Reciprocal______25 0 C, pitardii var. pitardii XC. reticulata______25 1 .69 Reciprocal______42 4 C. pitardii var. pitardU XC. saluencnsis______20 1 .01 Reciprocal. ______19 1

C. relicltlata XC. rosacfiora______35 1 .99 Reciprocal______34 0 C. TeUculata XC. rusticana______12 1 4 07· Reciprocal. ___ .. __ • ______. ______..._.. ______48 0 C. reticu/ata XC. salllcnensis______24 3 .43 Reciprocal______28 2

C. rosaefiora XC. fraterna______36 2 1.49 Reciprocal______26 0 C. Tusticana XC. fraterna______162 3 .94 Reciprocal______50 0 C. saZucncnsis XC. fraterna______24 1 Reciprocal ______25 1.06 0

C. salucnensis XC. hongkongcnsis______12 Reciprocal______36 3.06 C. sa/ucnensis X C. 11ltchuensiB- ______.. ______19 Reciprocal. __ . ____ .. ______14 .76 C. sasanqua XC. granthamiana______60 ReciprocaL______6 ~I .21 C. sasonqua X C. japonica______46 Reciprocal______29 1.97 C. sasanqua XC. relicltlata______88 Reciprocal. ______31 1.08 C. sinensis X C. fratema______32 Reciprocal______27 11 .86 C. sinensis XC. hiemalis______66 5 .89 Reciprocal______11 0 C. sinensis xC. miyagiL______58 10 3.38 Reciprocal______17 0 C. sinensis XC. sasUllqua______52 7 .02 Reciprocal______12 0

1 Statistical calculations based on Snedecor and Cochran (1967), pp. 215-226. "IndicateI' significant difference (0.05). ··Indicates highly significant difference (0.01). STUDIES WITH CAMELLIA AND RELATED GENERA 21

:rt·1NWWfH \,' 'in. 'iI> t".t

'I ~"':~ . 1:\

PN-2182 FIGURE I.-Normal and abnormal development in Camellia embryo culture. A through E show normal development and F through ! show abnormal: A, Two days after placement in culture jar. B, Five days in culture; early stage of development of radicle. C, Ten days in culture; development of radicle. D, Four weeks in cullure; radicle is developed but partly hidden; hypocotyl arched ove1" cotyledons. E, Six weeks in culture; normally developed seedling with primary leaves unfolding; at proper stage for transfer from c!llture. F, Six weeks in culture; stunted "adicle and hypocotyl; however, seedling de veloped sufficiently after G months for transfer from culture. G, Six weeks in culture; no radicle development; stunted hypocotyl; radicle never developed and sc('dling- finally died. H. Ten weeks in culture; very small embryo; normal radit'le but no hypocot)'I; no further development. I. Ten weeks in culture; embryo very much shriveled at time of culturing, but expanded in culture before abortion. 22 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE counts were omitted from this glabrous stems, all of the study. In table 22, each hybrid progeny also had gla is listed separately and identified brous stems. by its code number. Future refer (2) Hybrids from crosses ences to code numbers of hybrids where C. gmnthamiana are based on this table. was the male parent con Species possessing strongly sistently showed highly dominant characters can be ex raised midrib and veina tremely useful to the plant tion on the lower surface breeder in ascertaining the valid of the leaves, which ap ity of the hybrids when those peared as deep indenta species are used as male parents. tions from the upper sur Specific characters that appeared face. Illustrations of the to be dominant among the F 1 leaves of both parents progeny are as follows: and progeny are shown (1) The pubescent stem char in figure 2 where pollen acter was dominant over of C. g1'anthamiana was the glabrous. Hybrids re used on C. japonica, C. pi sulting from female par tm'dii val'. pita1'dii, C. re ents with glabrous stems ticulata, and C. saluenen crossed with males hav sis. ing pubescent stems were (3) Hybrids from crosses consistently pubescent where C. hongkongensis stemmed, indicating ho was the male parent mozygosity of the pubes showed a distinct re cent stem character in semblance to this species the Glenn Dale parental by having narrow, elon species used. Thus, 62 hy gate leaves. Illustrations brids of C japonica X C. of the leaves of both par lutchuensis, 24 of C. 1'US ents and progeny are ticana X C. l'utchuensis, shown in figure 3, where 10 of C. japonica X C. pollen of C. hongkongen [1'ate1'na, three of C. rus sis was used on C. japon tican(L XC. !1'ate1'na, three ica, C. saluenensis, and C. of C. 1'eticulata X C. !m 1'usticana. te?'na, three of C. japon (4) Hybrids from crosses ica X C. granthamiana, where C. !mtenw was the two of C. japonica X C. male parent showed simi kissi, and one of C. 1'eti larity to this species in a culata X C. gmnthami rather substantial num ana had pubescent stems ber of characters. These like their male parents. include willowy, spread When both parents had ing plant habit with thin, STUDIES WITH CAMELLIA AND RELATED GENERA 23

highly pubescent young silky terminal buds typi shoots, dull grey-green cal of C. l~ttchuensis over foliage, and shape of the long, sharply pointed, leaves characteristic of C. glabrous terminal buds of /,ratenw. Illustrations of C. japonica or C. rusti the last character are cana,' and dominance of shown in figure 4. Figure leaf veination patterns of 4, A, B, and C show the C. japonica or C. 'rusti dominance of C. fmtm'na cana over those of C. lut as the male parent with chuensis. respect to the shape of the Not all species transmitted leaves. As shown in figure such distinct dominance, and the 4, D, similar dominance hybrids were intermediate in was expressed when C. character. Hybrids of this type f1'atenw was used as the are illustrated in figure 5. Here, female parent. In the additional proof of hybridity was latter case, subsequent obtained by chromosome counts chromosome counts or determinations of percentages proved the validity of the of aborted pollen. hybrid. C. fmtm'na is ex Floral characters proved to be tremely floriflerous, and it a valuable supplement to vegeta is not unusual for a ma tive characters and were espe ture plant to have six to cially helpful in some otherwise eight flower buds per questionable cases of hybridity. node along the entire Flower size and form were most length of each branchlet. frequently intermediate between This trait, illustrated in that of the parents. However, figure 4, E, appeared in where small-flowered, wild spe all 18 hybrids involving cies such as C. j1'aterna, C. kissi, this species which have C. lutchuensis, or C. miyagii were flowered to date. crossed with large flowered cul tivated varieties of C. japonica, C. (5) Among the hybrids of C. 1'eticulata, C. 1'ustica~'/,a, or C. japonica X C. lutclmensis sasanqua, the small-flowered spe and C. 1'usticana X C. l1tt des appeared to exert the greater chuensis, the male parent influence on the progeny. C. lutc/mensis, showed Certain species transmitted dominance of pink to red their floral characters more coloration of young strongly than others. This was shoots and leaves over the true of C. gmnthamiana, which light-green coloration of transmitted large, frilled petals C. japonica or C. 1'llstito the majority of its progeny. cana; dominance of short, Figure 6 shows the nature of the stout, bl untly pointed, petals in this species and in typi ~4'- TECHNICAL BULLETIN 1'127, U.S. DEPT. OF AGRICULTURE

PN-2183

PN-2184

FIGURE 2.-Leaf veination of parents and hybrids from crosses where C. gmntha miana was the male parent. Hybrids show indented veination of leaves similar to C. g1'ClllthamiculU: A, Left, C. japonica, female parent; center, hybrid B-3; right, C. [J1'CInthamiunu, male parent. B, Left, C. pitardii val'. pitardU, female parent; center, hybrid B-6; right, C. [J1'CI1Ithu1niu'//u, male parent. C, Left, C. rcticu{uta, female parent; center, hybrid B-5; right, C. [/runthamian{l, male parent. D, Left, C. sa/ucHC'lISis, female parent; center, hybrid B-8; right. C. gl'(lIl/lw1I!ia/ICI, mull' parent. STUDIES WITH CAiHELLIA AND RELATED GENERA 25

PN-2185

PN-2186 26 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

PN-2187

PN-2189 FIGURE 3.-Leaf shape of parents and hybrids from crosses where C. hong kongensis was the male parent. Hybrids show elongate leaves similar to C. hongkongensis: A, Left, C. japonica, female parent; center, hybrid 5-47; right, C. hongkollgensis, male parent. B, Left, C. saiuenensis, female parent; center, hybrid P-48; right, C. hOllgkongensis, male parent. C, Left, C. rusti cana, female parent; center, hybrid 7-1; right, C. hongkongensis, male parent. STUDIES WITH CAMELLIA AND RELATED GENERA 27 cal hybrids. C. hongkongensis has interest. One hybrid selection, a trumpet-shaped flowers and cross between C. -ntsticana 'Yosh rough, grey perules, which are ida' (P.I. 228187) 4 and C. lut also common characters in many chuensis (P.I. 226756) was of its hybrids. The floral form named 'Fragrant Pink' by the characteristic of C. hongkongen Plant Science Research Division sis and its hybrids is shown in in 1966 and has since been distrib figure 7. The floral petals of C. uted to the nursery trade. This is a salllenensis have an illuminative loose, peony-type flower, 2v... quality that is also present in inches bY' 11/t. inches, medium pink, most of its progeny. with 10 'petals and 12 petaloids. Floral scent is rather rare in The scent of the flowers is similar the genus and is not generally to that of C.l'lltch71ensis and it rep present in the cultivated species. resents an improvement in flower In addition to C. sasanqua, which size and form from that of C. l71t has only a few faintly scented chuensis, which is Iv... inches cultivars, six other species, C. across, white, and single. Al /raterna, C. kissi, C. l71tchuensis, though no claim is made that C. miyagii, C. oleifem, and C. 'Fragrant Pink' is commercially tsaii, are scented. Only C. lut outstanding, it appears to be a ch71ensis and C. tsaii have a scent definite step in the right direc that is pleasant to most people. tion. Figure 8 illustrates four of The scent of the other five species the more scented ornamental hy is described as being musky and brids. unpleasant. Floral scent appears to be transmitted to a large per Determination of Pollen centage of the hybrid progeny. Abortion Because this character is unique Early in the interspecific hybri in these species, its presence in dizati~)n program, the question of hybrid progenies provided a valu the selection of suitable parents able marker. Among 59 C. lut arose, especially in regard to chuensis hybrids, 55 were species represented solely by com scented. Only 16 C. fraterna hy mercial varieties. The flowers of brids have flowered to daw, but many cultivated varieties are fre all except one were scented. All quently almost sterile hecause of eight C. miyagii hyhrids observed distorted nonfunctional styles and have the scent characteristic of aborted pollen. Thus, it was nec this species. Nine out of 21 C. essary to evaluate all prospective sasanqua hybrids were scented. parents for quality of the pollen None of the hybrids involving C. kissi, C. oleifera, or C. tsaii have • P.I. refers to the accession number flowered as yet. assigned to foreign introductions by the Plant Science Research Division, Agri Several of the scented hybrids cultural Research Service, U.S. Depart are of considerable ornamental ment of Agriculture. 2~ TEt'Il:\ICAL BL'LLETI:\ 1-l!!7, t:.S, DEPT, OF AGRICl'LTL'RE

PN-2190

I'N 21(11

Fu;l'lu: -I. [l"minllnt f"liar and floral bud l'iwrll,'tl'r:< tran~mittl'cl by C. ()"afrl'llll tll It" hybrid,;, ,-\ through Ii "how dominant 1paf ~hapt' of C, (I'II/rn.a and g show~ tIll' pn>litk tlont! huds ('harHl'tl-ri:;tk uf C 11'(1/('1"1/(1 and it,; hyhrid~: ..1_ lpn, (', ."I/It,liil'((, f.-mal!, parcnt; ,'(-rHpr, hybrid .1-,; right, C, imfaH(J, mal., l'nrl'nt. U, Lpn, c', /" !"'llluICl, t',-nw1" pan-nl; l'('nt.'r. hyhrid ~-,; rig-ht, e', i"l1 fC1'//J(, mall, (lan-nl. e', L(-ft, C, 1"11':1 i(,(IIIC1, [pmal(' [l

PN-2192

PN-2193

PN-2194 :Hl TE('I[:\WAL IH'LLETI:\ 11:!7, l'.:-:, DEPT, OF AGRl

PN-2195

Fr,;l fIr .'. L.'a: ,hap·' ..I' pan'nb and hybrid~ "howing" int('rnwdiat(' nature of 'L·· l:yhnd~ .\. L .. :'t, (', "C, 'I' "'1, f.·nlah· 1':11'1'1:1: ('PBU'!', hybrid <'-1; rig-ht. t f' j, f ri tl .... ~ ru,,: .. p;l!'f'n • /1. Lpt"t, (', > U!HltI:('O, [('lluilp pal'l-nt; t>(Jntpl", hybrid :-. •'. r;~h!. ( ·";,r,'tf"),(', S'-". llla!fI pal'Pllt. (', Lpft ('~ JIl}lfllI ir'(J , fpnudp parpnt;

, .. f ~t'~'. hrhrid -:.:2:;0; rlf.!'ht. (' !''''~'f ",,!.''. 1l1ah· parptH. fJ, Lpft, C. SflSll'''/lUI, f.'lIla;.. par"Jlt; , .. nt.. !', hyhrid :i-;i:!; rig-ht, C, m;i!,!!/ii, male parent, E, Left, C, ."1,.;'. 'I "I, ""Ill:l;. !""'l'll'; ."'Ilt!')', hrhrid B-1; l'l)!ht. (', "/(';:'1'/'0, 11111),' parent. S'ITllll':S WITll (,A.HELL/A AXil HELATIW (;EXl':IL\ :31

PN 2Ur; 32 Tl~Cll~ICAL BULLETIN 1.127, U.S. DEPT. OF AGRICULTURE

.i'N 2200 I'N 2201

PN-2203

P.N :!:!O·I PN ~205

Fwnn: fl., Flow\'I"':: of C. 111"11111 fill III ill 1111 and hybrid:; ,:howing- ll'ansmission of frill.,d IH'tnl,:: [1"<>111 C. 1/I'UII/fluJlliHIIIJ wlll'1I it wali lIsed ali tlll' male pan-nL: ,'I. C, ynw/huIlI/IIlI/a; II, hyhdd g·;l of C, jll])f)lIil'(I " C, !11"Il11t/Wl/lialla; e, hybrid fl·,; (If (', l'i/u/'(liI val'. }lifunlii ", ('. !!l'llll/hamirl/HI; n, hyhrid B-5 of C. l't'fil'lIla/1I ' (', !/I'IIH/haw/HllrI" I~', hyhrid B-~ of C, .~(/hll'IIt'II,~is X C. !l1'tlllf/ltll/liIOil/,' F, hybdd H-,I ur (', SUSIlIlf/II(l " C. !JI'U II IIw III icuw. STUDIES WITH CAMELLIA AND RELATED GENERA 33

PN-2206

PN-2207

PN-2208 FIGUR!;l 7.-Flowers of C. hongkongensis. and hybrids showing transmission of trumpet-shaped floral form when C. hongkongclIsis was used as the male parent: A, C. hongkongensis,' B, hybrid 5-12 of C. japol1ica X C. hong kongensis,. C, hybrid 7-1 of C. 1'usticana X C. hong/wngensis. 34 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

PN-2210

PN-2211 PN-2212 FIGl'RE S.-l~low('rs of ~Cl'nt('d hybrids of possible ornamental value: A, Hybrid, 'Fnlgr:,nt Pink,' of C. I"llslicalla X C. Illlc/lIu.'llsis; B, highly scented hybrid 3-12 of C. japollic(l " C. llllc/wcl/sis; C, mildly scented hybrid 3-4 of C. I"/CStica)1CL X C. fralcl"lW" D, strongly scented hybrid 5-56 of C. sascmqua X C. miyagii. STUDIES WITH CAMELLIA AND RELATED GENERA 35 before any meaningful interspe cent. Introductions with higher cific hybridization program could percentages of aborted pollen be undertaken. Table 4 lists per were omitted from the breeding centages of aborted pollen for the program. Figure 9, A shows uni parents used in interspecific form, filled, stainable pollen and crosses. The percentages of figure 9, B shows shriveled, aborted pollen ranged among the empty, unstainable aborted pol species from 3 percent to 58 per len.

TABLE 4.-Pe?'centages of abo'rted and laTge, 'P1'esmnably '1.ln1·educed pollen, based on observations of 1,000 grains [1'01n a 1nini1n~L1n of six anthe1's f01' each species 01' clone of Camellia

P.I. or Aborted Large B NO.1 Species pollen pollen

Percent Percent 162476_ __ _ C. fraterna______22 3 251534. ___ C. granthamiana ______• ______55 1 B56995 ____ C. hiematis______11 7 B56996______do______29 6 229973. ___ C. hongkollgensis______• _ 19 0

B56272 ____ C.japonica ______17 3 226109______do______22 5 227063 ______do ___ .. ______19 3 228024______do______12 2 230278______do______21 3

231686______do______.______24 3 231687______do ______-______28 1 231689 __ ._ ____ do______7 2 231690. ______do______8 5 231694 ______do ______13 1

231695.___231858______dodo______--_------43 4 231859______do______48 2 238725______do______25 18 274530______do______51 2

274797______do______7 2 274799.______do______12 2 275054 ______.do______4 3 275512______do______21 22 309001______do______13 13 See footnote at end of table. 36 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

TABLE 4.-Percentages of ab01·ted and large, p1'esumably un1'educed pollen, based on obse1'vations of 1,000 gmins from a ?ninimum of six anthers fol' each species or clone of Camellia-Continued

P.l. Aborted Large BNo.1 Species pollen pollen

Percent Percent 319283 ___ _ C. jClpon'icu_ ~ ______.______C. kissL ______20 5 252062 53 1 _ ~_do ___ •____ • ______252064 9 2 226756 C. lulchuellsis__.~ ______C. 1niyagiL__ _, ______21 1 226704~ 22 1

.~.do ___ . ______17 231057 1 162561 C.oleifera __ .. ----______37 do ___ -- ______._ 11 o 235500 ~~~ 2 B58296 ___ _ C. pilardii var. pilardii ______• ___ 33 9 B51685 ___ _ C. reticula/a______.. 33 1

B51686 ___ _ .do______• ______._ 1~ B51689 ___ _ .do______. ______1 1 B51690 ___ _ - _.do ______._. __ .• ______. ______11 2 B51694. ___ _ ~ - .. do ___ -_...... _. _____ • _____ • ______.. 14 4 C. r(lsaejiora ___ •• _. __ " ______•• ______26 B58619 ___ _ 2 228187 __ C. rllsticalla_ -...... _.. __ '" ______• ______34 1 228188 _ .do______._. ______- ______21 o 228190 __ . _do ___ ._~ ______.•. ______11 2 • _do ___ .... ~ __ ..• ______58 233642. _ 3 C. salicifolia _ ~ ~ . .. ___ • ______14 310320 .. 9 C. nalucllensis______21 243862 2 227624 C. 8aSa"qua_. _____ ~. ______22 do ___ --. ______16 2 228025 2 do___ _._. ______~ _____ 19 235568 __ 2 _do ___ ~ .• ~ __ . ______. ______27 237854 3

277763 do. __ .- - ~-- ______. ______17 do ___ _ 2 319284 24 2 . do __ _ 319285 25 2 235570 C. sillellsis. ___ •. 4 2 do ____ ~ 235572 6 2 See footnote at end of table. STUDIES WITH OAMELLIA AND RELATED GENERA 37

TABLE 4.-Pe1·centages of ab01·ted and large, p1'esumably unTeduced pollen, based on obse'tvation..<: of 1,000 g1'ains j?'om a minimum of six anthe'ts f01' each species 01' clone of Camellia-Continued

P.I. or Aborted Large BNo.· Species pollen pollen

Percent Percent 235573____ C. sinensis______3 0 304404______do______6 1 304405______do______10 1 316471______do______14 5 316472 ______do______8 3

316473______do______26 4 316476______do______13 3 316477______do______22 1 316478______do______10 1 B57043____ C. taliensis __ ------______15 0

B57048____ C. tenuijiora______28 5

I P.I. refers to the accession number assigned to ioreign introductions by the Plant Science Research Division, Agricultural Research Service, U.S. Department of Agricul ture. B numbers are code numbers assigned to importations under postentry quar antine permit by the U.S. Plant Introduction Station, Glenn Dale, Md. Observation of the pollen also p. 62) appear to be the result of disclosed that the production of the functioning of unreduced pol large grains with presumably un len grains, and the chromosome reduced chromosome numbers is number observed for one hybrid very common in the genus. All of C. pita1'dii var. pitm'dii X C. but five of 66 introduct;\ons ob gmntha1'/'/.iana (see p. 57) appears served produced at least some to be the result of an unreduced large pollen as shown in table 4. egg. Figure 9, C shows the presence of The use of hybrids in the inter large pollen. Although germina specific transfer of desirable tion tests unfortunately were not characters depends upon fertility. made on large pollen grains, in Furthermore, a greatly increased direct evidence from chromosome frequency of aborted pollen pro counts of parents and hybrids in duced by a presumptive hybrid in comparison with the frequencies dicate that at least some unre produced by the parents provides duced pollen grains and egg cells corroborative evidence of hybrid are viable. The chromosome Tlum ity. Pollen abortion among the bel'S observed for two hybrids of hybrids varied from 100 percent C. 1'eticulata X C. ja;ponica (see to percentages similar to those of 38 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE • • • -. •• PN-2213

~ ~ ~ ~ B

PN-i!214

...... c PN-2215

FIGURE 9.-FiIled, empty, and large pollen following staining with acetocarmine (125 X): A, Uniform, filled, stainable pollen. C. japonica, P.I. 231695; B, shriveled, empty, unstainable aborted pollen. Hybrid 4-4 from C. japonica X C. !rateJ'l1a; C, large, presumably unreduced pollen occurring among filled and aborted grains of normal size. Hybrid A-2 from C. 1'etiC'ltlata X C. japonica. STUDIES WITH CAMELLIA AND RELATED GENERA 39 the parental species. Table 5 sum gmnthamiana, T. virgata X C. mi marizes the pollen abortion deter yagii, and C. pita1'dii var. pitardii minations of flowering hybrids. X T. spectabilis. In addition, six As expected, the percentages of as yet unvalidated plants were se aborted pollen produced by the cured from F. alatamaha X C. hybrids were generally greater hongkongensis. than the percentages produced by the parental species. Large, pre Morphological Comparisons sumably unreduGed pollen was Hybrid validity was based ini also observed in 112 of the 134 tially on com parisen of vegetative hybrids examined. characters of parents and hybrids (appendix tables 23 and 24) and Intergeneric subsequently corroborated by Hybridization floral characters, pollen abortion determinations, and chromosome As an extension of the inter counts of the parents and prog specific hybridization program, eny. In figure 10, foliar charac attempts were made to hybridize ters are shown for T. virgata X C. several Camellia species with g1'anthamiana, T. virgata XC. mi species of related genera. This yagii, and C. pitardii var. pitardii was initiated in a limited fashion X T. spectabilis. The hybrids illus with crosses of Cmnellia with trated for both T. virgata crosses Tutche1'ia spectabilis and T. vir are similar to the male parent, gaia. Some early success encour whereas the hybrid of T. specta aged the later expansion of hybri bilis appears intermediate. dizing to include outdoor plants The floral characters, especially of Fmnklinia alatamaha and Ste petal shape, of t.he hybrids of T. wa1'tia ovata. vi1'gata XC. miyagii are shown in The intergeneric crosses at figure 11, along with those of the tempted are listed in table 6, parental species. The petals of along with the number of pollina two hybrids (fig. 11, D and E) tions and the number and percen were distinctly similar. to the male tage per pollination of capsules parent, four (fig. 11, C, F, H, and produced, of seeds started in cul /) were intermediate, and one ture, and of hybrid plants. A total (fig. 11, G) was similar to the of 1,064 controlled pollinations female parent. Figure 12 shows were made. These rellresented 24 the floral characters of the hybrid intergeneric combinations, includ of C. pitardii var. pitardii X T. ing reciprocal crosses. Eleven spectabilis and its parents. None plants judged to be valid hybrids of the hybrids of T. vi1'gata X C. were secured from T. virgata X C. granthamiana have flowered yet. TABLE 5.-Pe1·ceniages of abor'ied and p1'esumably un'reduced pollen produced by interspecific Camellia hy- 1/::0. b'rids. Obser'vations of 1,000 grains from a minimum of six anthers were made for each hybrid having 0 some filled pollen; 500 gmins were examined in completely male-sterile hybrids 8 aI:=J l1: Ahorted pollen Presumably unreduced pollen Z Hybrids a...... Interspecific crosses counted Range Range > Mean Mean t"' Low High Low High to c::: t"' t"' fVuntbar l'ercent Percent Percent Percent Percent Percent I:=J 8 C. Jratema X C. japonica______• ______1 84 84 84 1.0 1.0 1.0 .... ReciprocaL ______" _. ______Z 8 48 98 78 0 3.0 2.2 .... C. Jraterna X C. reticulata ______• ______lI>.- 1 91 91 91 .5 .5 I>:) Reciprocal______.5 3 23 67 39 0 1.6 .9 ~-1 C. japonica XC. honukolluensis______6 80 97 92 0 5.2 2.1 c::: ~ C. japom'ca XC. kissi. _____ - ______.. 1 83 83 83 .4 .4 .4 t:;j C. ja1Jonica xC. 11ltchuensis ______45 82 100 96 0 3.5 1.1 I:=J C. japonica XC. ntiyauii.______"tI 2 93 96 94 0 1.2 .6 ~ C. japonica X C. reticlllata ______2 54 99 76 0 4.0 2.0 ReciprocaL______"'______0 6 58 93 83 .3 7.2 4.7 '>j > C. japonica XC. rusticana______3 7 46 22 1.1 6.0 3.0 0 C. japonica X C. saluenensis______~ 4 76 82 80 0 1.8 1.3 ..... Reciprocal______1 77 77 77 0 0 0 CJ c::: C. japonica XC. sinensis. ______3 91 94 92 0 2.1 1.7 t"' C.oleiJera XC. hientalis______95 95 95 .3 .3 .3 8 1 c::: ~ C. oleiJera X C. ntiyauii______2 91 97 94 .2 1.2 .7 I:=J C. pitardii var. pitardii XC. granthantiana______1 89 89 89 1.6 1.6 1.6 C. pilardii var. pitardii XC. japonica______1 86 86 86 .1 .1 .1 C. pilardii var. pitardii XC. lutchuensis______1 92 92 92 1.2 1.2 1.2 C. pilardii var. pitardii XC. saluenensis ______2 88 93 90 .2 1.2 .7

C. reticulata XC. yranlha7niana __ • ______2 63 88 75 3.6 5.8 4.7 UJ __ __ >-3 C. reticulllia X C. saillenensis _____ ••• ._. • ______2 49 75 62 2.0 3.7 2.8 c:: C. ruslicllna XC. frlllenw_ •. ______t1 3 12 99 51 0 9.3 4.7 .... C. rusticllHa XC. llllchuellsis______20 82 100 95 0 6.0 2.4 trJ CIJ C. saluenensis X C. yrcLrllhll7niana______1 100 100 100 0 0 0 ~.... C. sllluenensis XC. Iwnykonyensis______1 84 84 84 1.3 1.3 1.3 >-3 C. sllillenensis X C. lutchuensis__.. _. _. ______1 88 88 88 .7 .7 .7 C. sa"anqull XC. yTllllthll7nialla___ •• __._ •••• ___• ______• ______= 2 19 25 22 1.6 2.9 2.2 ~ C. ______sllsallqua XC. 7niyllyii. ______• . ._. ._. _ 4 5 88 46 .3 1.5 .8 ~ C. saSllllqua X C. oleifera______ttJ 1 44 44 44 1.6 1.6 1.6 t'< t'< C. sasllnqua XC. reticuillta ______• ____ ._._ .. ______•• ____ • ______3 16 43 27 .7 2.1 1.3 ;i: > Z t::I ~ trJ t"' > t-3 trJ t1 G') trJ Z trJ >~

~ ..... TABLE 6.-lnte1·[Jene1-ic C1'osses: Franklinia alatamaha, Stewartia ovata, Tutcheria spectabilis, and T. virgata ,j::o.t¢ c'rossed with species of Camellia >-3 t.zj Interspecific crosses Polli Seed capsules Seeds started Hybrid plants (') nations produced in culture t:r: Z ..... Percent per Percent per Percent per (') pollination Number Number pollinalion Number pollinalion Number >t" 3 3.2 0 0 C.jupOJlica X T. spectubilis ___. . ____ " 95 2 2.1 Il:I C. pi/ardii var. pilardu X T. spectabilis____ 83 1 1.2 2 2.4 1 1.2 c:: t" C. Japonica X T. virga/a ____ 28 0 0 0 0 0 0 0 t" C. miyagii X T. virgata ___ .• ,------ 59 3 5,1 0 0 0 t.zj 0 >-3 C. reticulata X T. virgala_ _. _" •- - 15 0 0 0 0 0 ..... C. ruslicalla X T. virga/a_" __ ,, ______,,.-. 26 1 3.8 1 3.8 0 0 Z 0 ~ C. sasanqua X T. virga(u __,." .. - - _. - --- ------ 16 0 0 0 0 0 0 "'"~ F. alalamaha XC. granlhamiana ___ ------25 4 16.0 0 0 0 :'l 6 11.3 F. alalamaha X C. hongkongensis. ---- - - - --- - -- 53 3 5.6 23 43.4 c:: F. ala/amaha XC. japol!ica____ .. ______25 5 20.0 0 0 0 0 rn 0 0 F. ala/amaha X C. miyagii. ____ . ___ __ - - - ----- 23 1 4.3 0 0 t:l F. alalamaha XC. oleifera__ ., ______23 2 8.6 0 0 0 0 t.zj -- "tI F. ala/amaha XC. si1lensis__ .. ______51 8 15.7 0 () 0 0 !-3 0 S. ova/a XC. japonica __ .. ______25 0 0 0 0 0 0 S.ovala XC. kissi______... ______28 1 3.6 1 3.6 0 0 '%J 0 0 S. ovala X C. miyagii___ __ "__ ____ --- ----- -- 31 0 0 0 0 > S.ovala X'"'. oleifera ______26 0 0 0 0 0 0 c;') ::d ..... T. virgala xC. granlhamiana ______106 5 4.7 3 2.8 3 2.8 (') 0 0 c:: T. virga/a xC. h01!gkol!gensis______29 0 0 0 0 t" T. virga/a X C.japonica ______48 1 2.1 2 4.2 0 0 >-3 T. virga/a X C.llllchllensis______3.7 1 3.7 0 0 c:: 27 1 ::d T. virgala xC. miyal1ii. ______152 3 2.0 7 4.6 7 4.6 t.zj T. virgala xC. rllslicana______45 4 8.9 2 4.4 0 0 T. virgata xC. si1l61lsis______25 2 8.0 5 20.0 0 0 STUDIES WITH CAMELLIA AND RELATED GENERA 43

PN-2216

PN-2217

FIGURE lO.-Foliar characters of parents and hybrids involving intergeneric crosses: A, Left, Tlltcheria 'virgata, female parent; center, hybrid 7-233, showing indented veination of male parent; right, Camellia granthamiana, male parent. B, Left, T. virgata, female parent; eenter, hybrid 5-62, showing leaf shape similar to male parent; right, C. 1niya,gii, male parent. C, ,Left, C. pitcz1"Clii vnr. piturclii, female parent; center, hybrid A-19, intermediate in leaf shape bewteen parents; right, T. spectabilis, male parent. ~14 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

PN-2219

PN-2220

PN 2221 PN ,2222 STl'DIES WITH CA.llELUA A,\D RELATE:D GENIWA 45

PN 2223 PN 222·1

PN 2225

PN 2227

FHaRE II. Fh)ral l'll:lI'al'l('r~ of IHlI'l'nb and hybrids i/1Yolving inLl'I'g'pnl'I'k ('1','''''(,''; ..t. 1'11 1(,1" rill l'iJ'lIll/rr. fl'mall' pal'Pllt, with broad Iwtld,,; n, CUlIll'iiia lI,i!,'I!/ii, mal.' pan'n!, with nalTtl\\' J'l'f1l'xpd Iwlnls; C, hybrid 5..G2, inll'I' nll·dial,· llPl\\'l','n pan'nls; n, hyhrid 5-,Il, "imilal' 10 malt' IHlI'('nL; E, hybrid ~,-71;, similar Itl malt· IJan'lll; F, hyhrid 5 ", illtl'rllll'dinll' 11l'tll'l'l'1l IHII'l'J1t~; Ii. hyhritl ii-Sil, ~illlilaJ' to fl'malt'; II. hybrid 5-81, intl'J'lllpdintl' 1ll'IIl'l'l'll paJ'l'lIl,;; I, hybrid ii-l-l:.!, inlt'rJ111'diall' 11l'lwppIl pan'nl:;, 46 TECIINICAL BL"LLETIN 1.127, U.S. DEPT. OF AGRICULTURE

PN-2228 PN-2229

PN-2230

1·'IGI·Rl-: 1:!.- Floral l'hamdt'r::; of parents and hybl'id involving an intergeneric l'ross: A, ('a md/ia }lila rdi; var. pita rdii, female parent; D, Tutcheria specta bi/is, mall parpnt; C, hybrid A-HI. STUDIES WITH CAMELLIA AND RELATED GENERA 47

Determinations of Pollen with a mean of 86 percent. Pollen Abortion abortion for the one hybrid of C. The parental plants of the gen pitCt1'd'ii val'. pitm'dii X T. SlJect~ era related to Camellia showed bilis was 77 percent. Although considerable variation in percen these percentages appear to be tages of aborted pollen. Counts sufficiently high to give reasona disclosed that F. alatamaha had 34 ble certainty of hybridity, enough percent aborted pollen, S. ova,ta fertility is indicated to warrant had 69 percent, T. spectabilis had the use of the intergeneric hy 26 percent, and T. v'i1·ga:,'.;, had 9 brids in further crosses. percent. Large pollen was found Large pollen was found in five only in the two Tutche1'ia species. of seven hybrids of T. vi1'gata X T. spectabilis had 1 percent and C. miyugii. The range was from 0 T. vil'gata 8 percent large pollen. to 2.3 percent, with a mean of 0.7 In addition, approximately 4 per percent. The one hybrid of C. pi cent of the pollen of T. vi1'gatct tct1'dii val'. IJita1'dii X T. spect~ had an unusual elongate shape. bilis did not have any large pol This odd form of pollen, shown in len. figure 13, was also observed in four of seven hybrids of this species which flowered. Pollen abortion for the seven hybrids of T. vi1'gata X C. miyagii • ranged from 77 to 99 percent, 'I' \ , I • A B I PN-2231 PN-2232

FIGURE l3.-Pollen of Tutchel'ia, lIil'ucMa. and hybrid, stained with acetocarmine (X 125) : A, Pollen of T. vil'gata showing normal pollen and unusual elongate pollen i 8, pollen of hybrid 5-70, typical of four of seven hybrids of T. virgata X Camellia miyagii, showing normal pollen, aborted pollen, and unusual elongate pollen. 48 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

Species Compatibility of aborted pollen than did their Relationships parents. Pollen sterility of hy brids, a valid criterion of compat Sealy (1958) divided the genus ibility relations between specIes, Camellia into 12 sections of 82 was also used as a criterion in species plus an unplaced group, this study where possible, but Dubiae, of 16 species. Representa many hybrids have not yet flow tive species of only six of the 12 ered. The data are presented pri sections plus the Dubiae group marily as a possible guide to are presently grown in the United plant breeders to illustrate those States and in the Glenn Dale combinations that may be ex collection. These six sections are: pected to produce hybrids suc Camellia, CameIliopsis, Heteroge cessfully. nea, Paracamellia, Thea, and Theopsis. Species relationships in The largest amount of informa this study are based on these sec tion for intrasectional hybridiza tions. tion was available within the sec Hilsman (1966) and Savige tion Camellia. Here, six species, (1967) have proposed that for C. hongkongensis, C. japonica, C. breeding purposes the most suc pitardii val'. pitardii, C. ?'eticu cessful interspecific crosses will lata, C. rusticana, and C. salu be found within a section, partic enensis, were intercrossed in a ularly species with the sarne ploi substantial number of combina dy. Hybridizing results generally tions. Figure 14 summarizes the substantiate this assumption, but successful and unsuccessful com there are notable exceptions. binations within the section. Most Also, certain sections are more of the interspecific combinations, closely related than others, based with the possible exception of on the cross compatibility be those involving C. honglcongensis, tween their representative spe could be made without particular cies. difficulty. The compatibility ratio The number of valid hybrids in of the number of hybrids divided relation to the total number of re by the total number of cross-polli ciprocal cross-pollinations for nations within the section was 9 each parental combination af percent. forded a rough indication of the The following cr()sses failed: C. degree of compatibility. Neither honkongensis with C. pita?'dii the number of representative val'. pitarclii; C. hongkongensis as species in each section nor the the female parent with C. japon number of intrasectiollal or in ica and with C. saluenen.sis; C. tersectional combinations was honglcongensis as the male parent equal, so the data can only sug with C. reticulata; and C. ?·u.sti gest relationships. In general, all cana as the female parent with C. hybrids had greater percentages ?·eticulata. STUDIES WITH CAMELLIA AND RELATED GENERA 49

Q. '-..

~va, ~ FIGURE 14.-Compatibility between species within the section Camellia: Solid lines denote successful combinations and dotted lines, unsuccessful. Arrows point in direction of male parent for each cross.

The section Paracameliia was each was obtained from 79 C. fra represented by three species, C. terna X C.lutchuensis crosses and kissi, C. olei[era, and C. sasan50 reciprocal crosses. These hy qua. All three of these species hy brids were weak in growth and bridized very readily with each have not flowered. Two hybrids other. The compatibility ratio of were obtained from 36 crosses of hybrids in relation to total cross C. rosaeflora X C. fraterna, but no pollinations was 29 percent, the seed capsules were produced by highest within any of the sec 26 crosses of the reciprocal. In tions. The section Theopsis was rep sufficient blossoms prevented test resented by four species, C. fra ing of other combinations within tel'na, C. lu,tch'uensis, C. 'J'osaethis section. The compatibility flora, and C. tsaii. One hybrid ratio was 2 percent. 50 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE The unplaced group, Dubiae, of suIted in at least some hybrids, 16 species was represented by C. but successful crosses were much hiemalis, C. miyagii, and C. tenui more easily made between certain flora. C. hiemalis and C. miyagii sections than between others. The were very readily hybridized in width of the solid lines between either direction. Their compati sections is proportionate to the bility ratio was 19 percent. Insuf compatibility ratio. The greater ficient blossoms prevented testing the width, the greater the com C. ten'lti/lora with either of the patibility between sections. Bro other species. ken lines denote unsuccessful Compatibility between species combinations. of different subgeneric sections Species within section Camellia is summarized in figure 15. As produced hybrids with some rep shown in the figure, most in resentative species of all other tersectional combinations re- sections, including Dubiae. Sec f\ Theapsis

'" Camellia ~ Heterogenea ~ FIGURE 15.-Compatibility brtween species of different subgeneric sections: The width of the solid lines between sections denotes the rela.tive ease of hybridiza tion and dotted lines denote unsuccessful combinations. STUDIES WITH CAMELLIA AND RELATED GENERA 51 tion Camellia hybridized more of Dubh~e than with any other easily with Camelliopsis, Hetero section. genea, and Theopsis than with Theopsis species hybridized any other sec.tion. However, hy more easily with Camellia and brids resulting from crosses with Came1liopsis than with any other Heterogenea and Theopsis were section. Successful crosses with highly sterile, mostly with 80 per Paracamellia and with C. miya,gii cent or more abortive pollen. of Dubiae were difficult to achieve, None of the hybrids involving and crosses with Heterogenea and Camelliopsis have flowered yet. with C. hiernalis of Dubiae faHed. Camelliopsis species produced hybrids with some representative Chromosome Numbers species of all sections except Het Chrom050me Numbers of erogenea. Camelliopsis hybridized Parents more easily with Theopsis and Chromosome numbers for most with Camellia than with species of the species used in this study of any other section. have already been reported by the In addition to producing hy following: Darlington and Janaki brids with some representative Ammal (1945), Darlington and species of section Camellia, C. Wylie (1961), Janki Ammal g'ranthamiana of section Hetero (1952), Longley and Tourje genea produced two hybrids with (1959, 1960), Morinaga and oth C. sasanqua of Paracamellia, but ers (192.9), Morinaga and Fuku none with any other section. shima (1931), Patterson, Long Paracamellia species produced ley, and Robertson (1950), and hybrids with some representative Simura (1935). However, it was species of all other sections. Para necessary to establish the chro camellia hybridized with C. hie mosome numbers of the particu malis and C. miyagii of Dubiae as lar clones used as parents. The easily as when intrasectional import.ance of establishing these crosses were made within Paraca numbers is illustrated by Longley mellia. Also, Paracamellia hybri and Tourje (1959, 1960) who dized more easily with Thea than listed chromosome numbers for with any other section and only 59 C. japonica varieties; 47 were with considerable difficulty with diploids and 12 were triploids. Camelliopsis, Heterogenea, and Thus, one must be certain of the Theopsis. chromosome number of each par Thea species produced hybrids ent before determining numbers with some representative species in the hybrids. of all other sections except Heter Table 7 gives determinations of ogenea. Thea hybridized more root-tip chromosome numbers of easily with Paracamel1ia and Ca?nelliO" species and species of with C. hiemalis and C. miyagii related genera used as parents; 52 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

TABLE 7.-Determinations of root-tip chromosome numbe1'S of Camellia species and species of related genera used as parents

P.I. or Root-tip BNo.l Scientific name chromosomes

Number 162476 ____ C.Jralerna______90 251534 ____ C. Ilranlhamiana______60 B56995____ C. hiemaZis 'Shishi-Gashira'______90 B56996____ C. hiemalis 'Bill Wylam'______90 229973_ __ _ C. honllkonllensis______30

226109 ____ C. japonica No. 6______30 227063___ _ C. japonica 'Bon-Shirotama' ______30 228024 __ _ _ C. japonica 'Tsubaki' ______30 230278 ____ C.japDnica______30 231686_ __ _ C. japonica 'Hasumi-Shiro' ______30

231687 ___ _ C. japonica 'Komyotai' ______30 231689____ C. japonica 'Moshio' ______30 231690 ____ C.japonica______30 231694___ _ C. japonica 'Utamakura' ______30 231695___ _ C. japonica 'Yuki botan' ______30

231858 ____ C. japonica 'Beni botan'______30 231859 ____ C. japonica 'Kanyo tai' ______30 238725____ C. japonica 'Saudade de Martins Blanco' ______30 274530 ____ C. japonica No. 825______30 274797. ___ C. japonica No. 870______30

274799 ____ C.japonica No. 872______30 275054____ C. japonica No. 913______30 275512____ C. japonica D______30 309001.___ C. japonica 'Le Lys' ______30 319283 ____ C.japonica K-20______30

252062_ ___ C. kissi. ______30 252064______do______30 226756_ ___ C. Zutchuensis______30 226704._ __ C. miyagiL______90 231057______do______90

See footnote at end of table. STUDIES WITH CAMELLIA AND RELATED GENERA 53

TABLE 7.-Dete'rminations of root-tip chromosome nu.mbers of Camellia species and species of 'related genem used as pm'ents-Continued

P.I. or Root-tip B NO.1 Scientific namE' chromosomes

Number 162561- __ _ C. oleifera______do ______90 235500 ___ _ 90 B58296___ _ C. pitardii var. pitardii.______30 B58619 ___ _ C. rosae;'lora ______90 228187 ___ _ C. rusticana 'Yoshida' ______30

228188___ _ C. rusticana 'Hatano' ______30 228190 ___ _ C. rusticana 'Koshiji' ______30 233642~ __ _ C. rusticana B White plena ______30 243862 ___ _ C. sC!luenensis No. 6093 ______30 227624 ___ _ C. sasanqua 'Kokinran' ______90

228025 ___ _ C. sasanqua 'Sazanka' ______90 235568 ___ _ C. sasanqua______90 237854 ___ _ C. sasanqua No. 791. ______90 C. sasanqua______319284 ___ _ 90 319285___ _ C. sasanqua 'Onishiki' ______90

235570 ___ _ C. sinensis 'Tama-midori' ______30 235572 ___ _ C. sinensis Y-2 ______30 235573 ___ _ C. sinensis Z-l ______30 304404 ___ _ C. sinensis No. 1______30 304405___ _ C. sinensis No. 2______30

316471. __ _ C. sinensis 'Benji-Fuji'______30 316472___ _ C. sinensis 'Beni-Homare' ______30 316473 ___ _ C. sinensis 'Makinowara Wase' ______30 316476___ _ C. sinensis 'YamatomidorF ______30 316477___ _ C. sinensis______45 316478______do ______60

230368___ _ Tutchena spectabilis ______30 229881. __ _ T. virga/a ______75

TABLE 8.-CMomD.'wme numbers of Camellia species and species of ?'elated genera used as parents

Scientific Name Root-tip Source I !!hromosomes

Number C. japonica 'Ville de Nantes' ______30 Patterson, Longley, and Robertson (1950) C, reticulata 'Crimson Robe' ______90 Longley and Tourje (1959) C, reticulata 'Great Shot Silk' ______90 Longley and Tourje (1959) C. reticulata 'Chang's Temple' ______90 Longley and Tourje (1959)

C. reticulata 'Large CornelIian' ______90 Longley and Tourje (1959) C, I'eticulata 'Lion's Head' ______90 Longley and Tourje (1959) C. saBanqua 'Narumi-gata'______75 Longley and Tourje (1960) C. laliensis ______;30 Janaki Ammal (1952) C, tenuifiora ______60 Longley ~nd Tourje (1960)

Franklinia alatamaha ______36 Santamour (1969) Stewartia ovata ______30 Santamour (196tJ)

I Chromosome numbers are not directly determined in this study; recorded from literature. STUDIES WITH CAMELLIA AND RELATED GENERA 55

PN-22S6 A 1, l'N-2233

PN-2237 FIGURE 17.-Root-tip chromosomes of Tutcheria spectabilis and T. virgata PN-2234 (X 2,300): A, T. spectabilis, 2x =30; B, T. virgat:!, 5x=75.

Chromosome Numbers of Interspecific Hybrids Table 9 gives determinations of root-tip chromosome numbars of the 88 interspecific hybrids of Camellia that produced root tips suitable for study. Hybrids resulting from crosses between diploid species were in all but c one case also diploid. Figure PN-2235 18 shows the chromosomal complements of representative FIGURE 16.-Root-tip chromosomes of diploid hybrids, including C. C. iutc/mOLsis, C. miyagii, and C. X C. rosaefiol'a (X 2,300): A, C. lut japonica honkongemis, chue1!sis, 2x=30; B, C. miyagii. C. japonica X C. lutchuensis, 6x=90; C, C. rosaefiora, 6x= 90. C. japonica X C. saluenesis, C. ja 56 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

TABLE 9.-Dete'r'lninations of 'j'oot-tip ch'ro1nosom.e nU1ltbel'S of 88 interspecific hyb1'ids of Camellia

Crosses Hybrids Root-tip chromosomes

Number Number

Diploid X diplid (30 X 30): C. japoniea XC. hongkongensis ______4 30 C. japoniea XC. kissi. ______1 30 C. japolliea X C.lulehuensis______9 30 C. japolliea XC. lulehuensis______1 28 C. japoniea X C. pilardii var. pilardiL ______1 30 ReciprocaL______1 30 C. japolliea X C. ruslieallo, ______• ______2 30 ReciprocaL ______• ______1 30 C. japolliea X C. saluellens1s______1 30 ReciprocaL _____ ...______• ______1 30 C. japoniea xC. sinensis ______2 30 C. japoniea X C.laliensis ______1 30 C. kissi XC. rustieana______1 30 ReciprocaL ____ • ______1 30 C. pil(lrdii var; pilardii X C. soluenellsis ______1 30 ReciprocaL ______1 30 C. rustieana XC. hOllllkollllensis______1 30 C. rustieana X C. lulehuensis ______6 30

Diploid X tetraploid (30 X 60): C. japoniea X C. grallthamiana______2 45 C. pilardii var. pilardii XC. Ilranlhamiana ______1 45 C. pitardii var. pitardii xC. granthamiana ______1 60 C. saluenellsis XC. Ilranthamiana _____ ... ______1 45

Diploid X hexaploid (30 X 90) C. japoniea XC. fralema ______5 60 C. japolliea XC. fralerna ______- ______1 56 C. japoniea X C. fralerna ______1 59 C. japoniea XC. Memalis ______1 60 C. japolliea XC. miyallii.______1 60 C. japoniea XC. reticulala ______:2 60 C. japoniea X C. rosaeflora ______- ___ _ 1 60 C. pilardii var. pitardii xC. fralema ______1 60 C. pitardii var. pilardii XC. reticulala______1 60 C. ruslieana X C.fratemu ______1 60 C. saluenellsis X C. fra/ema ______.___ -- ___ - . - -- 1 60 C. saluellensis X C. reticulala ______2 60 STUDIES WITH CAMELLIA AND RELATED GENERA 57

TABLE 9.-Detenuinations of 1'00t-tip ch?'omosome nU1ubm's of 88 inte'tspecijic hyb'rids of Camellia-Continued

Crosses Hybrids Root-tip chromosomes

Number Number

Hexaploid X diploid (90 X 30): C. fratenlU X C. japonica __ • ______• ______• 1 60 C. reticulala X C. japonica ______• _. 3 60 C. reticuluta X C. japollica _. _____ • ______••• ___ _ 2 '75 C. reticulata X C. pitardii var. pitardii. __ • _. ___ ..___ • ____ _ 1 60 C. reticu/ala X C. rttsticana ______.._..____ _ 1 60 C. reticulata XC. saluellellsis _____ ...... ___ ..... , ______1 60 C. sasal/qua X C.japonicu ______... ______2 60 C. sasallqua XC. (enuijlora______1 60

PentapJoid X tetraploid (75 .x 60): C. sasal/qua 'Narumi-gata' X C. gralllhamiana. ____ • _•• ___ _ 1 60

Pentaploid X hexaploid (75 X 90): C. sasanqua 'Narumi-gata' XC. reticulata______1 90

Hexaploid X hexaploid (90 X 90): C. fraterna XC. reticulala______1 90 ReciprocaL_____ • __ ._. ______..• ______._ 2 90 C.oleifera XC. hiemal·s_. ______1 90 C.oldfera XC. miyagii.______2 90 C. reticula/a XC. sasallqua______1 90 C. sasanqua XC. hiemalis______1 90 C. sasallqua X C. miyagii__...______• ______3 90 C. susallqua XC. miyagiL ______•• ______1 86 C. sasanqua XC. oleifera______1 90 C. sasanqua XC. reticulata______2 90 ponica XC. taliensis, C. kissi xC Hybrids resulting from crosses 1'u,st'icana, C. 1"ltsticana X C. l'utbetween diploid female parents chuensis, C. saluenensis X C. iaand tetraploid male parents re ponica, and C. pitardi?: var. pitarsulted in triploid:; in four out of dii X C. saluenensis. One aneu five hybrids studied. One of the ploid hybrid with 28 chromo two hybrids resulting from somes occurred from a cross of C. crosses between C. pitardii var. iaponica X C. lutchuensis. This pitardii (2x-30) X C. gmntha plant was very weak in growth miana (4x=60) was tetraploid. and has subsequently died. Figure This hybrid was probably the re 19 shows the chromosomal com sult of the functioning of an un plement of this aneuploid hybrid. reduced egg cell from C. pitardii 58 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

A B PN-2238 PN-228\1

• c o PN-224() PN-2241

•• - ;?" ~ ;;.~: - '.. J

E F PN-2242 STUDIES WITH CAMELLIA AND RELATED GENERA 59

• ,.~l .... v~· -.\f ..•••..;

G H

PN-2244 PN-2245

FiGURE lS.-Root-tip chromosomes of interspecific hybrids from crosses of diploid parents (X 2,300); A, Hybrid 5-37 of C. japonica X C. Iwngkongensis, 2x=30; B, hybrid 3-32 of C. japonica X C. lutchuensis, 2x=30; C, hybrid 5-1 of C. ja.ponica X C. saluenensis, 2x=30; D, hybrid P-37 of C. japonica X C. taliensis, 2x=30; E, hybrid 5-69 of C. kissi X C. Tustieana, 2x=30; F, hybrid 1-1 of C. Tusticana X C. /utchwmsis, 2x=30; G, hybrid 5-29 of C. saluen6ltsis X C. japonica, 2x=30; H, hybrid P-42 of C. pita1'dii val'. pitardii X C. saillenensis, 2x=30.

... A - B ,

PN-2246 PN-2247

FIGURE 19.-Root-tip chromosomes of aneuploid hybrid of C. japonica X C. lutchu cllsis with 28 chromosomes (X 2,300) : A and B, Two division figures from the aneuploid hybrid. 60 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

var. pitm·dii. Figure 20, A, B, All but two of 18 hybrids stud and C show the chromosomal ied from crosses between diploid complements of the triploid hy female parents and hexaploid brids resulting from crosses be male parents resulted in tetra tween C. japonica X C. gmnth~ ploid hybrids. Figure 21 shows mia.na, C. saluenensis X C. g'ran the chromosomal complements of tluL1nia.na, and C. pitardii var. pi the tetraploid hybrids resulting tardU X C. gmnthamiana, and from crosses between C. japonica figure 20, D shows the comple X C. fmten/a, C. japonica ment of the tetraploid hybrid re X C. reticulata, C. japonica X C. sulting from C. pitardii var. pi 1'osaejlom, C. pita1'dii var. pitar tU1'dii XC. gmnthamiana. dii X C. fraterna, C. rusticana X

A B

PN-2248 PN-2249

c D

PN-2250 PN-2251 FIGURE 20.-Root-tip chromosomes of interspecific hybrids from crosses of diploid female parents and tetraploid male parents (X 2,300): A, Hybrid 7-96 of C. japonica. X C. gl'antilamiana, 3x=45; B, hybrid B-2 of C. saluenensis X C. granthamiana, 3x=45; C, hybrid P-56 of C. pita1'clii var. pitardii X C. granthanrialla, 3x=45; D, hybrid B-6 of C. pitarclii var. pitarclii X C. granthamialla, 4x=60. STUDIES WITH CAMELLIA AND RELATED GENERA 61

PN-2262 PN-2263

c PN-2264 PN-2266

E F PN-225o PN-2257 FIGURE 21.-Root-tip chromosomes of interspecific hybrids from crosses of diploid fEmale parents and hexaploid male parents (X 2,300): A, Hybrid 4-4 of C. ja,ponica X C.fretterna, 4x:;::60; B, hybrid P-13 of C. japonica X C.reticulata, 4x=-60; C, hybrid P-46 of C. japonica X C. 'rosaejlora, 4x=60; D, hybrid 7-159 of C. pita1·dii 1}(Lr. pitantii X C. fra,ierna, 4x=60; E., hybrid 3-2 of C. Tusticana X C. fratentel., 4x:;::60; P, hybrid P-5 of C. saluenensis X C. reticulata, 4x=60. 62 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

C. fratenta, and C. saluenensis X C. japonica, C. reticulata X C. ja C. 1·cticulata. Two hybrids with ponica, C. ?·eticulata X C. pitardii aneuploid chromosome counts var. pitm·dii, C. ?·eticulata X C. were observed, one with 56 chro ?"itsticana, C. ?·eticulata X C. sal mosomes, the other with 59. Fig 'llenensis, C. sasanqua X C. ja ure 22 shows the chromosomal ponica. and C. sasanqua X C. tcn complements of these two aneu nifiom. The two pentaploids were ploid hybrids. obtained from C. reticulata X C. Ten hybrids from crosses be japonica. If the somatic numbers tween hexaploid female and dip of both parents were exactly loid male parents were tetra halved during meiosis, only tetra ploid, whereas two hybrids were ploid hybrids would result. A pos pentaploid. Tetraploid hybrids sible explanation for the occur were, obtained from C. jrate1-na X rence of the two pentaploids is• the I

PN-2258 PN-2259

PN-2260 PN-2261 FIGURE 22.-Root-tip chromosomes of two aneuploid hybrids of C. japonica X C. fl'Utel"llu, one with 56 chromsomes and the other with 59 (X 2,300): A, Aneuploid hybrid 4-44 with 56 chromosomes; B, same aneuploid hybrid as in A, 56 chromosomes; C, aneuploid hybrid 4-48 with 59 chromosomes; D, same aneuploid hybrid as in C, 59 chromosomes. STUDIES WITH CAMELLIA AND RELATED GENERA 63

PN-2262 PN··2263

•

c PN-2264 PN-2265

FIGURE 23.-Root-tip chromosomes of interspecific hybrids from crosses of hexaploid female parents and diploid male parents (X 2,300) : A, Hybrid 5-58 of C. rcticulata X C. 8aluencnsi8, 4x=60; B, hybrid A-6 of C. 'reticlllata X C. japnnica, 4x=60; C, hybrid A-8 of C. rcticulata X C. japonica, 5x=75; D, hybrid A-14 of C. rctic:nlata X C. japonic:a, 5x=75. fertilization of the hexaploid fe numbers of its hybrids seem un male parent with unreduced pollen predictable. A hybrid, A-24, re from the diploid parent. Figure sulting from C. sasanqua 'Naru 23 shows the chromosomal com mi-gata' (5x=75) X C. gmn plements of tetraploid and penta thamiana (4x=60) was tetraploid. ploid hybrids resulting from hex The morphological characters of aploid X diploid crosses. Thus, this hybrid were intermediate. It crosses between C. reticulata X C. is generally difficult to assess japonica can produce either tetra accurately the contribution of ploid or pentaploid hybrids. each parent to the hybrid in 'Narumi-gata,' a pentaploid va crosses involving polyploid spe riety of C. sasanqua, produced cies without the aid of genetical or hybrids when used as the female cytological markers. However, parent. However, the chromosome 'Narumi-gata' may have produced 64 TECHNICAL BULLE'l'IN 1427, U.S. DEPT. 01<' AGRICULTURE

an egg with 30 chromosomes, Chromosome Numbers of which united with a sperm carry Intergcneric Hybrids ing 30 chromosomes from C. grantham;iana. A hybrid of C. Table 10 gives determinations sasanqua. 'Narumi-gata' X C. reti of root-tip chromosome numbers cula.ta (6x=90) was hexaploid. of intergeneric hybrids. Among In this case, 'Narumi-gata' may the hybrids of Tutcheria virgata have produced an egg cell with 45 (5x=75) X Camellia miyagii chromosomes. Figure 24 shows (Sx=90), two hybrids each were the cht'omosomal complements of found with 75 and 90 chromo both these hybrids. somes, respectively, and one hy Among 15 hybrids studied brid each with 80, 82, and 105 from crosses between two hexa chromosomes. .Figure 26 shows ploid species, all but one were hex the chromosomal complements of aploids. Figure 25, A, B, C, and D these intergeneric hybrids. Little show the chromosomal comple apparent difference was noted in ments of the hybrids resulting general vigor and growth be from crosses between C. sctsanqua tween the euploids and aneuploids X C. 'nLiyagi-i, C. olei/em X C. mi of this cross. yagti, C. 'I'eticulata X C. /ratel'na.. Three hybrids of T. virgata and C. sctsunqua. X C. olei/era. In (5x=75) X C. granthamiana addition, one aneuploid hybrid (4x=60) had 75 chromosomes. Figure 27, A, B, and C show the with 86 chromosomes was found chromosomal complements of among the four hybrids of C. sas these three hybrids. The chromo (l1lqua X C. miYctgii. Figure 25, E some numbers of these hybrids and F show the chromosomal might suggest that they resulted complements of this aneuploid from accidental selfing or apomic hybrid. tic development. However, the hy

PN ·2266 PN-2267 !<'IGURE 24.-Root-tip chromosomes of interspecific hybrids from ~rosses of pent~ ploid female parel.t C. S(lS(I/(qll(l 'Narumi-gata' with tetraplolCl ~nd hex.aplOl~ male parents ('< 2,300): A, Hybrid A-24 of C. SCls(lnqua. Naruml-gata (5x::::75) X C. l/1'(I1It/w/lta.illl/(I (4x::::60), 4x=60; H, hybrId A-32 of C. sasllIlqllll 'Nurumi-gnta' (5x::::75) X C.1'eliC:ldata (6x::::90), 6x::::90. STUDIES WITH CAMELLIA AND RELATED GENERA 65 ., • , .... "

B PN-2268 PN-2269

D PN-2270 PN-2271

F PN-2272 PN-2273

FIGURE 25.-Root-tip chromosomes of interspecific hybrids from crosses of hexaploid parents (X 2,300): A, Hybrid 5-52 of C. sasanqlta X C. miyagii, 6x=90; B, hybrid 5-61 of C. alei/era X C. miyagii, 6x=90; C, hybrid 2-5 of C. I'eticulata X C. /raterna, 6x=:90; D, hybrid B-1 of C. sasanqua X C. alei/era, 6x=90; E, aneuploid hybrid 5-55 of C. sasanqua. X C. miyagii with 86 chromosomes; F, same aneuploid hybrid as in E, 86 chromosomes. 66 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

TABLE lO.-Dete1·minations of 1·00t-tip chromosome numbers of -inte1· gene7ic hybrids of Camellia and Tutcheria

Crosses Hybrids Root-tip chromosomes

Number Number Diploid X diploid (30 X 30): C. pilardii var. pitardii X T. speclabili8______1 30 Pentaploid X tetraploid (75 X 60): T. virgata XC. granlhamiana______3 75 Pentaploid X hexaploid (75 X 90): T. virgala XC. miyagiL ______])0 ______2 75 ])0 ______1 80 ])0 ______1 82 ])0 ______2 90 1 105

A ,

PN-2274 PN-2275

PN-2276 STUDIES WITH CAM.ELLIA AND RELATED GENERA 67 ..

o PN-2277 PN-2278

F PN-2279 PN-2280

1I ..,,."It.,- 'l , I

, "I - I " PN-2281 PN-2282

FIGURE 26.-Root-tip chromosomes of intergeneric hybrids of Tutcheria virgata X Camellia miyagii (X 2,300): A., Hybrid 5-62, 5x=75j B, hybrid 5-77, 5x=75j C, hybrid 5-70, 6x=90j D, hybrid 5-81, 6x=90j E, hybrid 5-80, 7x=105j F, aneuploid hybrid 5-70 with 80 chromosomesj G, same aneuploid hybrid as F, 80 chromosomeSj H, aneuploid hybrid 5-82 with 82 chromosomes; I, same aneuploid hybrid as H, 82 chl'omosomes. 68 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

~ ..~;~ ~ " ~... ,. .... / "~'\ ~:~D~'~

,~ . ,6 .....•• "". ; ,.... I A B • " I PN-2283 PN-2284

•

PN-2286 FIGURE 27.-Root-tip chromosomes of intergeneric hybrids of Tutcheria virgata X Camellia g1'anthamiana and C. pitardii var. pitardii X T. spectabilis (X 2,300): A, Hybrid 7-233 of T. virgata X C. granthamiana, 5x=75; B, hybrid 7-235 of T. virgata X C. grantitamiana, 5x=75; C, hybrid 7-255 of T. vil'gata X C. g1'antha?niana, 5x=75; D, hybrid A-19 of C. pitardii var. pitanlii X T. spectabilis, 2x=30. brids showed transmission from chromosomal complement of this C. g1'unthu?niana of vegetative hybrid. As with the previously characters as listed in appendix mentioned hybrids of T. vi1'gata table 24 and illustrated in figure X C. gmnthamianu, the chromo 10, A. some number cannot prove the The one hybrid obtained from validity of the hybrid. Again, the crosses between C. pitU1'd'ii var. proof of hybridity rests on the piturdi'i and T. spectabilis had 30 vegetative morphology of the hy chromosomes like the two par brid as listed in appendix table 24 ents. Figure 27, D shows the and illustrated in figure 10, C. STUDIES WITH CAMELLIA AND RELATED GENERA 69 Chromosome Morphology same parent were not being as The major characteristics used signed as mates. Therefore, the to identify individual chromo chromosome morphology studies somes were its total length and were restricted to the diploid species. the position of the centromere. In most of the species the centrom Karyotypes of the somatic eres were relatively easy to chromosomes of eight diploid loce teo The centromeres them species are illustrated in figures selves are achromatic and flanked 28 through 35 with the chromo on either side by chromatic re somes arranged in pairs of homo gions. logs from the longest to the short Observations disclosed that est pair. All measurements for whereas the shortest and longest each pair of homolog's within a pairs of homologs among the di species were essentially constant, ploid species could be easily dis though some minor differences tinguished, the large group of were encountered. 1'he data pre medium-length chromosomes. sented in tables 11 through 18 re comprising most of the comple flect the averages of any differ ment, were very difficult to clas ences that may have occurred be sify into individual pairs. Accu tween homologs. rate identification of sets of hom (1) C. l'utclmensis. Table 11. ologS was not possible for poly The chromosomes range in size ploid species, so the study was from 16.6 microns to 7.3 microns. confined to diploid species and hy The longest pair exceeds the next brids. pair by 2.1 microns; the second Initially, we thought that, after longest pair exceeds the third the karyotypes of each of the di longest pair by 2.3 microns .. The ploid species had been estab other pairs decrease in length lished, the study could be ex down to the shortest pair in tended to include the hybrids be rather even gradations. Five tween different diploid species. pairs were classified as median The assumption was that the ge with indices from 1.00 to 1.19, nome contribution of each parent four pairs were submedian with could be identified in the hybrid. indices from 1.56 to 2.30, and six This line of reasoning was found pairs were subterminal with ind ices from 2.44 to 7.00. Figure 28 to be an oversimplification and proved impractical for a number show the karyotype of somatic of reasons. For example, it was chromosomes of C.lutchuensis. (2) C. rusticana. Table 12. The difficult to decide which of two or chromosomes range in size from more possible mates should be as 15.6 microns to 7.3 microns. The signed to a given parental species. longest pair exceeds the next pair Also, there was no assurance by 2.5 microns. The other pairs that two chromosomes from the decrease in length down to the 70 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

TABLE n.-Basic ch?'omosomes of C. lutchuensis

Chromosome length Long arm Chromosome Arm to short No. Total Long Short length arm arm arm rlltio index

Microns Microns Microns 1______16.6 2,2 100:15 6.55 2______14.4 100:23 4.37 3______-- 14.5 l1.8 2.7 4______12.2 8.3 3.9 100:47 2.13 5______l1.8 6.1 5.7 100:93 1.07 6______l1.2 9.8 1.4 100:14 7.00 7______11.1 5.9 5.2 100:88 1.13 8______11.0 7.8 3.2 100:41 2.44 9______10.5 5.7 4.8 100:84 l.19 10______10.4 7.5 2.9 100:39 2.58 100:100 1.00 11 ______10.0 5.0 5.0 9.7 6.9 2.8 100:41 2.46 12 ______13______9.3 4.8 4.5 100:94 l.06 8.9 6.2 2.7 100:44 2.30 14 ______8.2 5.0 3.2 100:64 l.56 15______7.3 5.0 2.3 100:46 2.17

TABLE 12.-Basic chromosomes of C. rusticana

Chromosome length Long arm Chromosome Arm to short No. Total Long Short length arm arm arm ratio index

Microns Microns Microns 1_____ 15.6 13.3 2.3 100:17 5.78 ------~--- 2_ .. ___ 13.1 l1.1 2.0 100:18 5.55 3_____ . ______~---~~~------4______. _____ 11.8 9.5 2.3 100:24 4.13 5 ______._ 11.4 6.1 5.3 100:87 l.15 6______- ______11.1 8.8 2.3 100:26 3.83 7______10.8 5.5 5.3 100:96 1.04 100:20 4.94 8 ______10.7 8.9 1.8 9______10.6 6.1 4.5 100:74 1.36 10 ______10.5 8.4 2.1 100:25 4.00 11 ______10.4 6.5 3.9 100:60 1.67 9.6 7,1 100:35 2.84 12 ______2.5 13 ______9.2 5.4 3.8 100:70 1.42 14 ______8.9 6.1 2.8 100:46 2.18 7.8 4,6 100:70 15 ______3.2 1.44 7.3 4,1 3.2 100:78 l.28 STUDIES WITH CAjUELLIA AND RELATED GENERA 71 shortest pair in rather even gra with indices from 2.84 to 5.78. dations. Two pairs were classified Figure 29 shows the karyotype of as median with indices of 1.04 somatic chromosomes of C. rusti and 1.15, six pairs were sub cana. median with indices from 1.28 to (3) C. japonica. Table 13. The 2.18, and seven were subterminal chromosomes range in size from 16.4 microns to 7.3 microns. The II); Ilil (

," ., I• ,• ,( i j )J ) I I , J ; , J

•• •• • 0 'I '" C, I I (I il (f ; 5 PN-2287 PN-2288 FIGURE 28.-Karotype of somatic chro FIGURE 29.-Karotype of somatic chro mosomes of C. lutchueusis. mosomes of C. rusticana.

TABLE 13.-Basic ch'romosomes of C. japoni,'-!a

Chromosome length Long arm Chromosome Arm to short No. Total Long Short length arm arm arm ratio index

Microns Microns Microns L 16A 13.9 2.5 100:18 5.56 2. 13.7 11A 2.3 100:20 4.96 3. 13.0 10.5 2.5 100:24 4.20 4" 12 7 7.1 5.6 100:79 1.27 5_ 12.0 9.5 2.5 100:26 3.80 6. 11.4 6.4 5.0 100:78 1.28 7 11.0 8.8 2.2 100:25 4.00 8. 10.7 6.5 4.2 100:65 1.55 9. 10.2 8.1 2.1 100:26 3.86 10._ 9.7 5.4 4.3 100:80 1.26 I!. 9.6 7.2 2.4 100:33 3.06 12_ 8.7 4.7 4.0 100:85 1.18 13. 8.6 4.9 3.7 100:76 1.32 14. 8.2 4.7 3.5 100:74 1.34 Hi. 7.3 5.2 2.1 100:40 2A8 72 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE longest pair exceeds the next pair by 2.7 microns. The other pairs decrease in length down to the shortest pair in rather even gra dations. One pair was classified as median with an index of 1.18, six pairs were submedian with ind ices from 1.26 to 1.55, and eight () ii " t t " pairs were subterminal with ind ices from 2.48 to 5.56. Figure 30 shows the karyotype of somatic chromosomes of C. japonica. (4) C. saluenensis. Table 14. The chromosomes range in size PN-2289 from 12.8 microns to 4.8 microns. FIGURE 30.--Karotype of somatic chro The longest pair exceeds the next mosomes of C. japonica. pair by 2.2 microns. The other 1.00 and 1.16, five pairs were sub pairs decrease in length down to median with indices from 1.25 to the shortest pair in rather even 2.24, and eight pairs were subter gradations. Two pairs were clas minal with indices from 2.60 to sified as median with indices of 5.24. Figure 31 shows the kary-

TABLE 14.-Basic ch1'ornosomes of C. saluenensis

Chromosome length Long arm Chromosome Arm to short No. Total Long Short length arm arm arm ratio index

Microns Microns Microns 1_____ . _____ . __ . _____ .___ 12.8 10.6 2.2 100:21 4.82 2_.. .. ______• ______. 10.6 8.9 1.7 100:19 5.24 3... ___ .. _____ •. __ . _. __ . 10.4 7.9 2.5 100:32 3.16 4 ______.______10.1 7.3 2.8 100:38 2.60 5___ . ______. __ ._____ 9.4 7.7 1.7 100:22 4.53 6____ ._. _. ______. _...___ 9.0 5.0 4.0 100:80 1.25 7_- __ ._._.______8.2 6.8 1.4 100:21 4.86 8______.______8.0 4.3 3.7 100 :86 1.16 9_____ • ______•• ______7.6 6.3 1.3 100:21 4.85 10._._ .... ______7.0 4.0 3,0 100:75 1.33 11 ______. ______._.____ 6.8 4.7 2.1 100:45 2.24 12_. ___ - ______•. __ ._____ 6.4 3.2 3.2 100:100 1.00 13.• __ •. ______. ___ ._ 6.1 4.8 1.3 100:27 3.69 14 •..• _._. . __ .______5.5 3.1 2.4 100:77 1.29 15._. __ .". _ . ___ .______4.8 3.1 1.7 100:55 1.82 STUDIES WITH CAMELLIA AND RELATED GENERA 73

longest pair exceeds the next pair i \. I i i i ; I ).J by 2.0 microns. The other pairs decrease in length down to the shortest pair in rather even gra dations. Two pairs were classified as median with indices of 1.00 ; , i J I I ,; and 1.12, five pairs were sub I J median with indices from 1.27 to 1.71, and eight pairs were subter minal with indices from 2.75 to 4.60. "Figure 32 shows the kary otype of somatic chromosomes of ; i i i I I c" sinensis~ (6) C. pita?'dii var. pitardii. PN-2290 FIGURE 31.-Karotype of somatic chro Table 16. The chromosomes range mos.mes of C. sa/u('n(,lI8"is. in size from 16.5 microns to 6.7 microns. The longest pair exceeds otype of somatic chromosomes of the next pair by 1.5 microns; the C. saluenensis. second longest pair exceeds the (5) C. sinensis. Table 15. The third longest pair by 1.8 microns. chromosomes range in size from The other pairs decrease in 17.0 microns to 7.5 microns. The length down to the shortest pair

TABLE 15.-Basic Ch1"OmOSOmes of C. sinensis

Chromosome length Long arm Chromosome Arm to short No. Total Long Short length arm arm arm ratio index

Alicrolls l'vlicrolls llficr01l8 l. 17.0 13.0 4.0 100:31 3.25 'J ~- 15.0 11.5 3.5 100:30 3.29 3. 14.0 10.5 3.5 100:33 3.00 4 12.7 8.0 4.7 100:59 1. 70 O. 12.0 9.0 3.0 100:33 3.00 6. 11.7 7.0 4.7 100:67 1.49 7'. 11.2 9.2 2.0 100:22 4.60 8 .. 11.0 5.8 5.2 100:90 1.12 9. 10.8 6.6 4.2 100:64 1.57 10. 10.2 5.7 4.5 100:79 1.27 II. 10.0 8.0 2.0 100:25 4.00 12. 9 5 6.0 3.5 100:58 1.71 13_ 9.0 4.5 4.5 100:100 1.00 14. 8.5 6.5 2.0 100:31 3.25 15. 7.5 5.5 2.0 100:36 2.75 74 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE in rather even gradations. Three nine pairs were subterminal with pairs were classified as median indices from 2.72 to 7.25. Figure with indices from 1.16 to 1.20, 33 shows the karyotype of so three pairs were submedian with matic chromosomes of C. pitm'dii indices from 1.25 to 2.31, and var. pita1·dii. it II II)J 'I j j J j " (; J'

)1 II I \ .\ 1 i j ,), ~)

I , ,\ I I I ; , J I) It J I PN-2291 PN-2292 FIGURE 32.-Karotype of somatic chro FIGURE 33.-Karotype of somatic chro mosomes of C. sinensis. mosomes of C. pitardii var. pitardii.

TABLE 16.-Basic clw'omosomes of C.pitardii var. pitardii

Chromosome length Long arm Chromosome Arm to short No. Total Long Short length arm arm arm ratio index

l\ficrons Microns Microns 1__ '" _ 16.7 14,,5 2.0 100:14 7.25 -----~-----~------2. __ ------15.0 12.3 2.7 100:22 4.56 3•. ______•• 13.2 10.7 2.5 100:23 4.28 4. _____ . __ . ______. ____ 13.0 7.0 6.0 100:86 1.17 5_ . __.. __ 12.9 9.0 3.9 100:43 2.31 ~- .. - ... - ... ----- ... -- 6.- ___ ._. ______. _____ 12.5 6.7 5.8 100:87 1.16 12.1 9.8 2.3 100:23 4.26 L. ~~-----~-~------8_ .•. __ • __ ••.• ______11.8 6.6 5.2 100:79 1.27 9_____ . ______. ______11.5 10.0 1.5 100:15 6.67 10 ______. ___ • ____ . ______11.0 6.0 5.0 100:83 1.20 11. ______._- __ . __ 10.8 9.3 1.5 100:16 6.20 12 ____ ~Q~~ ______M~ ______~ 10.5 8.5 2.0 100:24 4.25

13._ .. _ M .. _~ ...... , .. __ .. _~_ 9.0 5.0 4.0 100:80 1.25 - 100:26 3.87 ~ .- - 7.3 5.8 1.5 14 .. , - -- - ...... _... 4.9 1.8 100:37 2.72 15. o _ " . .. ,.... - ... ~ ... ""-- 6.7 STUDIES WITH CAilJELLIA AND RELATED GENERA 75

TABLE 17.-Basic chromosomes of C. kissi

Chromosome length Long arm Chromosome Arm to short No. Total Long Short length arm arm arm ratio index

lIiierons Prh"crons Mierolls 1. 17.5 15.0 2,5 100:17 6.00 2 15.0 12.0 3.0 100:25 4.00 3 14.0 11.5 2.5 100:22 4.60 4 13.5 7.5 6.0 100:80 1.25 5 13.0 10.1 2.9 100:29 3.48 6 12.8 6.8 6.0 100:88 1.13 7. 12.0 9.3 2.7 100:29 3.44 8 117 8.0 3.7 100:46 2.16 9 11.0 8.5 2.5 100:29 3.40 10 10.1 5.1 5.0 100:98 1.02 11 10.0 7.8 2.2 100:28 3.55 12 9.9 6,4 3.5 100:55 1.83 13. 8.5 6.1 2.4 100:39 2.54 14 7 7 4.2 3.5 100:83 1.20 15. 75 5.4 2,1 100:39 2.57

(7) C. kissi. Table 17. The ., , • • chromosomes range in size from ,, I , 17.5 microns to 7.5 microns. The ; \ J; I i longest pair '2xceeds the next pair by 2.5 microns. The other pairs decrease in length down to the shortest pair in rather even gra , , • • I \ , I ,• , • j ... ,'" dations. Three pairs were classi i fied as median with indices from 1.02 to 1.20, three pairs were sub median with indices from 1.25 to 2.16, and nine pairs were subter • • • u • • minal with indices from 2.54 to ,, J I c. I i i 6.00. Figure 34 shows the kary otype of somatic chromosomes of PN-2293 FIGURE 34.-Karotype of somatic chro C. k ·ssl. mosomes of C. kissi. (8) C. salicifolia. Table 18. The chromosomes range in size from pair by 1.7 microns. The other 17.5 microns to 5.8 microns. The pairs decrease in length down to longest pair of homoJogs possess the shortest pair in rather even a microsatellite. The longest pair gradations. Three pairs were of chromosomes exceeds the next classified as median with indices 76 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE TABLE 18.-Basic ch1'C1nosomes of C. salicifolia

Chromosome length Long arm Chromosome Arm to short No. Total Long Shorf. length arm arm arm ratio index

Microns Microns llficTOIlS 1..••___ ._. __ • ___ ..____ . __ 17.5 15.0 2.5 100:17 6.00 2._ .••_._ ..•______._ 3 ______15.8 12.3 3.5 100:28 3.51 15.0 12.5 2.5 100:20 5.00 4. ______~~w_~_~_~ ______14.0 8.5 5.5 100:65 1.55 5. ____ ... _. ___ ._ 6____ . ___ • ______- ... -, ~ -...... 13.~ 11.0 2.6 100:24 4.23 12.7 6.5 6.2 100:95 1.05

L ,- ~ ~ ~ ~ ~ ~ "" - ~ 12.5 10.5 2.0 100:19 5.25 8 ______..... - ...... - - ...... 9______...... _"'- - - ,.. ... - ~ .. - "" 11.2 8.7 2.5 100:29 3.48 .. ~ .. -- ...... -- 10.5 7.5 3.0 100:40 2.50 10. __ • ___ ...... ~ ...... 10.0 7.0 3.0 100:43 2.33 1L.___ . '" " - ...... " _ ....., .. - 9.5 5.0 4.5 100:90 1.11 12 __ . _.. " .. ---~-,...... "- 9.0 5.0 4.0 100:80 1.25 13 •. _. __ - ...... ~" .. ~~ ... ~-~" ..... 7.0 5.0 2.0 100:40 2.50

14._ ..... - ". ~ .. ..,...... , - - - + -0...... 6.4 3.2 3.2 100:100 1.00 15•...•.. ,. ....",. - . .. - ~ .. 5.8 4.3 1.5 100:35 2 87 from 1.00 to 1.11, three pairs .. 't were submedian with indices ; i fJ.. l,' j,' from 1.25 to 2.33, and nine pairs JJ were subterminal with indices from 2.50 to 6.00 Figure 35 shows the karyotype of somatic chromosomes of C. salicifol1'a. If jf, If An analysis of the karyotypes of t I the basic chromosomes of eight diploid species of Camellia is given in table 19. The species are listed in the table in ascending , , , , order of the number of subter I • , , i 1 : J ; i minal chromosomes in each. PN-2294 Chromosomes 1, 2, and 7 are FIGURE 35.-Karotype of somatic chro· subterminal in all eight species; mosomes of C. salicifolia. whereas chromosome 3 is subter minal in all species except C. lut is subterminal in all except C. si clwensis; chromosome 5 is sub nensis. Chromosomes 6 and 10 terminal in all except C. pitardii are either median or submedian val'. pitco'dii, and chromosome 9 in all eight species; whereas chro en >-,3 d t:I .... t:j TABLE 19.-Analysis of k(t1oyotypes of basic chromosomes of 8 diploid species of Camellia 1 en

~.... Chromosome No. Type chromosome >-,3 Species ::r: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 m 8m st ~ i::: Number Number Number C. lulchuellsis______st ~ st sm m st m st m st m st m sm sm sm 5 4 6 ;b: C. rustica/w___• __ • ______• _. ______st st st m st m st sm st sm st sm sm sm sm 2 6 7 C. japollica______st st st sm st sm st sm st sm st m sIn sm st 1 6 8 > C. saluellensis______st Z st st st st sm st m st sm sm m st sm sm 2 5 8 t:I C. sinensis ______st st st sm st sm st m sm sm st sm m st st 2 5 8 ~ C. pilardii var. pilardii______• ____ -_ st st st m sm m st sm at m st st sm st st 3 3 9 t:j t" C. kissi______st st st sm st m st sm st m st sm st m st 3 3 9 C. salicifolia______> st sm st st sm st 9 ~ st st m st st sm m m st 3 3 t:j t:I 1 m = median; sm = submedian; st = subterminal. til t".i Z t:j ~ >

...::J ...::J 78 TECHNICAL BULLETIN 142'1, U.S. DEPT. OF AGRICULTURE mosome 4 is median or submedian or submedian in all except two in all except C. saluenensis; chro species, chromosome 13 is median mosome 8 is median or submedian or submedian in all except three in all except C. salici/olia; and species, and chromosome 15 is chromosome 12 is median or sub subterminal in all except three median in all except C. pitm'dii species. Thus, more variability var. pitarcUi. Chromosome 11 is appears to exist among the short subterminal in all except two chromosomes than among the species, chromosome 14 is median long ones.

DISCUSSION AND CONCLUSIONS

Rather wide differences were male than was hexaploid C.1'eticu observed in the percentages of lata. Also, hexaploid C. 1'osaejlom hybrids developed per pollination was less fertile as the female than between the reciprocals of inter diploid C. kissi. Three of the specific crosses, as shown in tables species that performed poorly as 2 and 3. In most cases, the num females, C. /,rateJ'na, C. lutchuen bers of pollinations and hybrids sis, and C. 1'osaejl07'a, are of sec were not sufficiently large to es tion Theopsis. C. /mte1'na and C. tablish significance by chi-square, lutchuensis at least may require but differences were significant higher temperatures than other for six of 54 reciprocal combina species of Camellia for success tions shown in table 3. The six when used as the female parent. reciprocal crosses showing signif C. lutchuensis was highly fertile icant or highly significant differ when used as the female parent ences were: diploid with diploid in Gonzales, La.,G and Long crosses C. japonica with C. lut Beach, Calif.,G but was poorly clwensis and C. rllsticana with C. fertile at Berkeley, Calif.' Simi lutcJwensis; diploid with hexa larly, C. /mtenla was highly fer ploid crosses, C. japonica with C. tile at Los Angeles, Calif.' I'eticlllata, C. J'llsticana with C. When used as the female par 'rcl'iculata, and C. kissi with C. ent in reciprocal crosses with C. rosae/lora; and hexaploid with japonica and C. 1'ust'icana, the hexaploid crosses, C. /rate?'na hexaploid C. J'eticulata was supe with C. reticlllata. The diploid C. l1ttclmensis as • Palmer, F. A., Jr. 1969. [Personal the female parent produced fewer correspondence.] progency in crosseS with C. ja "Neptune, H. 1969. [Personal cor ponica. ancl C. l'ustic(tJw than did respondence.] ., Cutter, R. K. 1966-69. [Personal either C. japonica or C. I"llst'icana. correspondence.] Similarly, the hexaploid C. /ra , Parks, C. R. 1967-68. [Personal cor terna was less fertile as the fe respondence.] STUDIES WITH CAMELLIA AND RELATED GENERA 79 rior to both of these diploid spe grantha?niana produced some cies. partly fertile hybrids. The percentages of aborted pol The same factors that cause re len produced by the hybrids gen duced pollen fertility may also be erally exceeded those of the par expected to cause reduced mega ents. Individual hybrids with to spore or egg fertility. Therefore, tally aborted pollen were obtained attempts to successfully cross onl:.) from C. japonica X C. l~ltchuensis, hybrid of low pollen fertility with C. 'l"ltsticana XC. lutchuen.sis, and another hybrid of similar low fer·, C. saluenensis X C. gl'anthami tility are usually very difficult. In ana. Eleven of the interspecific instances where the desiI't'd g'e w crosses produced hybrids having netic characteristic has bE.len Sf)-, averages of 90 percent or more cured in the hybrid, a morH pr'!}" aborted pollen. In contrast, some ductive procedure may be to "se of the clones of the diploids C. hybrids of low fertility a'fi raale japonica, C. kissi, and C. 1"ltsU parents in crosses with female calla had more than 50 percent clones of high fertility and pos aborted pollen as did the tetra sessing other desirable character ploid C. gnmthamialla. Although istics lacking in the hybrid. reduced fertility in polyploids Compatibility relationships of often can be attributed to irregu species within subgeneric sections lar meiotic distribution of chro and between sections were inves mosomes, varying frequencies of tigated. Within the section aborted pollen encountered Camellia there is a close relation among different clones of the pa ship between C. japonica, C. pi rental diploids also suggest the tardii val'. 1Jitardii, C. 1'eticulata, presence of chromosomal struc C. 1'usticwna, and C. saluenensis. tural changes or segregation of All of these species hybridize genes affecting fertility in Camel quite readily with each other. C. lia. The variation in frequencies hongkongensis, on the other of aborted pollen among the par hand, failed to hybridize in one ents undoubtedly was a factor in half ·of the reciprocal combina differences in fertility of hybrids tions attempted, as shown in obtained from the same interspe figure 14. Therefore C. hong kong cific cross. Success in the inter cnsis apparently is the least specific transfer of desirable compatible species within the sec genes in future hybridization pro tion. This observation agrees grams requires some degree of with Parks and Griffiths (1966) fertility in the F 1 hybrids. Al who stated that from both chemo though the hybrids exhibited re taxonomic and cross-compatibility duced fertilit;{ in comparison criteria, C. hongkongensis is with the parental species, all suc apart from other species in this cessful crosses except the triploid section. Thus, five of the species hybrid of C. salllenensis X C. seem to be closely related ecospe 80 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE cies and C. hongkongensis a pos with species within any other sec sible cenospecies in relation to the tion. Actually, the compatibility others. ratios of these two species with The three species, C. kissi, C. those of Paracamellia is 18 per oleife1'a and C. sasanqua, within cent for C. miyagii and 13 per section Paracamellia all hybridize cent for C. hiemalis. This com freely with each other, indicating pares with 9 percent for intrasec a close relationship typical of tional crosses of species within ecospecies. secti0n Camellia. Compatibility C. /,raterna, C. lutch'llensis, and evidence thus suggests that these C. l'osaefiom of section Theopsis two species might in fact belong were the most difficult of the in within section Paracamellia. trasectional crosses to hybridize. Section Thea appears more Their behavior, typical of ceno closely related to section Paraca species, suggests the existence of meIIia and to C. hiemalis and C. genetic isolation barriers more m'iyagii of Dubiae than to species typical of intersectional species of other sections. crosses than that expected of Section Camellia shows a some crosses within a single section. A what closer compatibility rela factor that may playa part here tionship with Heterogenea, is that all three species are poor Theopsis, and CameIIiopsis than female parents. Seed set has al with Paracamellia or Thea, al ways been low, regardless of type though at least some hybrids of pollination. A rather large were possible from all combina number of reciprocal crosses in tions. High pollen abortion volving one of these species com among the hybrids indicates that bined with another species from a the species of each of these sec different section showed that in tions act as cenospecies in rela variably the number of hybrids tion to species of other sections. was always much higher where C. granthamiana of section the Theopsis species was the male Heterogenea is most closely asso parent. A possible explanation ciated with species of section for this deficiency was previously Camellia and wiII hybridize with discussed on page 78. difficulty with C. sasanqua of sec C. lriemalis and C. m'iyagii of tion Paracamellia, but not at all section Dubiae hybridize readily with species of other sections. In between themselves and also with general, C. gmnthamLiana showed species within section Paracamel a higher overall incompatibility lia. A lack of any apparent ge in intersectional crosses than did netic isolation barriers suggests any of the other species tested. that C. Mentalis and C. miyagii In summary, the subgeneric are ecospecies and that they are sections appeared to fall into two more closely related to species more-or-Iess distinct groups. In within section Paracamellia than the first, species of Paracamellia, STUDIES WITH CAMELLIA AND RELATED GENERA 81 Thea, and Dubiae appear to be 56 chromosomes and hybrid 4-48 more generally compatible with with 59 chromosomes, resulted each other than with species of from C. japonica .(2x=30) X C. the other sections. In the second, f'rate1'1w (6x=90). These two hy species of Theopsis and Camel brids developed from separate liopsis appear to be closely re seed capsules grown on the same lated to each other and in turn C. japonica parent and pollinated more closely compatible with the same day in 1964. Both hy Camellia than with other sec brids survive and resemble the tions. Heterogenea would appear hexaploid male parent in most to fall within this second group vegetative characters (appendix because of its closer association table 22), the exceptions being with Camellia than with any leaf veination and margins for other section. 4-44 and leaf veination for 4-48. Determinations of chromosome General vigor and growth of both numbers were made for the 88 in of these hybrids are comparable terspecific hybrids of Camellia with those of the other eight hy that produced root tips suitable brids of this cross. Both hybrids for study. Among these, four hy prodnced 98 percent aborted pol brids had aneuploid chromosome len, suggesting considerable ir complements; ull others had mul regularities in meiosis of the dn tiples of the basic chromosome balanced chromosomal comple number of 15. Only one aneuploid ments. was found reported in the litera Among four hybrids of the ture; Longley and Tourje (1959) hexaploid cross C. sasanqua X C. list the hybrid variety 'Saluta miyagii, hybrid 5-55 had a com tion' from C. saluenensis X C.1·e plement of 86 chromosomes. This ticulata as having 53±2 chromo aneuploid survives and shows somes. However, this variety had greater similarity to the female been reported earlier by Janaki parent, C. sasanqua" than do the Ammal (1952) as being a tetra other three hybrids of this cross. ploid with 60 chromosomes. Be Of eight vegetative characteris cause of the loss of vital genes tics listed in table 22, hybrid 5-55 and genetic imbalance, the ab resembles the female parent in sence of a single chromosome is four of these and the male parent usually lethal in diploid plants in the other four. As with the (Muntzing, 1961). An aneuploid, other polyploid hybrids with deficient for two chromosomes, aneuploid chromosome numbers, occurred among the hybrids of and in contrast to the 2x-2 aneu the diploids C. japonica X C. l'ltt ploid, there is little apparent dif chuensis. The aneuploid had small ference in vigor and growth be distorted leaves, was very weak tween hybrid 5-55 and the three in growth, and subsequently died. euploid hybrids of this cross. Pol Two aneupioids, hybrid 4-44 with len abortion was 88 percent, 82 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE somewhat lower than that for the rads of microspores, diploid va other aneuploids. rieties occasionally produced Observations of the pollen of dyads and sporads with one or the Camellia species used as par two extra small spores. Patter ents disclosed that the production son, Longley, and Robertson sug of large, presumably unreduced, gested that the triploids origi grains is very common in the nated after the functioning of un genus. Only five of 66 introduc reduced diploid pollen developed tions listed in table 4 did not pro from dyads of spores. Other trip duce at least some unreduced pol loids in Ca?nelUa have also been len. Spontaneous polyploidy detected. Darlington and Wylie through the production of func (1961) reported the triploid num tional unreduced gametes is not ber for C. iaponica val'. gmndi unique in Camellia but has been fiom. Longley and Tourje (1959, reported in a number of horticul 1960) listed chromosome counts tural plants including Dianthus for 59 cultivated varieties of C. by Brooks and Mehlquist (1961) ; japonica, and 12 were triploids. Mal'u.,~ by Einset (1952); Na'rcis C. vernalis 'Hirya' was reported sus und b'is by Emsweller and to be triploid by the $ame au Uhring (1960); and Vitis by thors. Further evidence for the Olmo (.1952). The occurrence of functioning of unreduced gametes spontaneous triploids in diploid in Camellia is provided by the species is attributed to the func pentaploid hybrid 'Inamorata' de tioning of unreduced gametes rived from C. saluenensis (2x= (Elliott, Hi5S) . 30) X C. retiC'lllata (6x-90). Janaki Ammal (1952) attributed Triploidy in Camellia was first the origin of the pentaploid to the detected in C. sinensis. Karasawa functioning of an unreduced egg (1.9.32, 1935) found a large from the diploid parent. lea'ved plant of tea, which had 45 chromosomes. Patterson, Long The functioning of an unre ley, and Robertson (1950) first duced egg from a diploid parent is reported the existence of triploids also indicated in this study. among cultivated varieties of C. Among five hybrids of the cross japonica. The three triploids C. pita1'dii val'. pitardii (2x=30) found by Patterson, Longley, and X C. gmnthamiana (4x-60), a Robertson had irregular meiotic tetraploid hybrid was obtained. As association. Univalents were de shown in table 4, C. pitardii val'. tected, and more than the normal pita1'dii was observed to produce number of four spores from a 9 percent large, unreduced pollen. single microsporocyte often Presumably, meiosis would be sim occurred, resulting in the produc ilar in microsporocytes and mega tion of aneuploid, aborted pollen. sporocytes, and unreduced mega In addition to the expected tet spores occurred, resulting in a dip STUDIES WITH CAMEL.LIA AND RELATED GENERA 83 loid egg produced by C. pitardii x C. granthamiana, a tetraploid var. pitm·dU. species. Two pentaploid hybrids re Similar to the unpredictable be sulted from crosses of C. reticu havior of C. sasanqua 'Narumi lata (6x=90) X C. iaponica gata' was that of another penta (2x=30). Ali of the 21 clones of ploid, Tutaheria virgata, when C. iaponica produced large, un used as the female parent in in reduced pollen in percentages tergeneric crosses with C. gran ranging from 1 to 22. With the thamiana (4x=60) and C. miya exception of one hybrid with the gii (6x=90). Three pentaploid unpredictable pentaploid C. sas hybrids resulted from crosses of anqua 'Narumi-gata,' all of the T. vi'rgata (5x=75) X C. gran other hybrids involving C. reticu thamiana, (4x=60). Pentaploids lata as one parent, and listed in are more erratic in meiotic be table 9, had the expected chromo havior than tetraploids. Assum some numbers. Accordingly, the ing that tetraploid C. grantham origin of the two pentaploids is iana produced reduced diploid attributed to the functioning of sperms, the production of eggs unreduced pollen of C. iaponica. with three sets of chromosomes With the exception of the pen by T. vi?'gata could account for taploid C. sansanqua 'Narumi the origin of the pentaploid inter gata,' all species of Camellia used generic hybrids. as parents had even multiples of The seven intergeneric hybrids the basic chromosome number. resulting from crosses of T. vir Because 'Narumi-gata' is penta gata (5x=75) X C. miyagii ploid, varying chromosomal con (6x=90) gave greater variation tributions to its hybrids may be in their chromosome complements expected. C. 1'eticulata is hexa than any of the interspecific com ploid and, as shown in table 9, in binations. As shown in table 10, crosses with other species with two hybrids had 75 chromosomes, even multiples of the basic chro two had 90, and one had a comple mosome number, produced 16 hy ment of 105. In addition, two hy brids having expected chromo brids had aneuploid numbers of some numbers. The hybrid C. ,c;as 80 and 82 chromosomes, respec anqua 'Narurr:.i-gata' X C.1'eticu tively. The source of variation in latn had 90 chromosomes, Accord chromosome numbers of &11 these ingly, it seems likely that 'Naru hybrids is difficult to assess. mi-gata' produced a functional Atypical chromosome numbers egg cell with three sets of chro would be expected in the eggs of mosomes. The functioning of an the pentaploid female parent. Al egg cell with two sets of chromo though the hexaploid male parent somes may account for the origin produced six hybrids with the ex of the tetraploid hybrid obtained pected chromosome numbers in from C. sasanqua 'Narumi-gata' crosses with other Camellia spe 84 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE cies, an aneuploid hybrid had 86 more, the hybrid might be ex chromosomes rather than the ex pected to show considerably phe pected number of 90 when C. notypic similarities to the male myiagii was used as the male par parent. Of the seven hybrids ent in a cross with C. sasanqua. from the cross, hybrid 5-80 with In consideration of the instability 105 chromosomes was the only of both parents and the absence one to resemble the female parent of adequate cytological markers, in floral form (fig. 11), and simi the chromosomal contributions of larly resembled the female parent the parents is quite speculative, more than the male in vegetative particularly in regard to the two characters (appendix table 24). aneuploids. If C. miyagii contrib Accordingly, the combination of uted the reduced number of 45, as 15 chromosomes from the female it apparently did in six of seven parent and 90 from the male interspecific hybrids, the two pen seems most unlikely. With the ex taploid intergeneric hybrids ception of one aneuploid, all 14 would have been derived from other hybrids of hexaploid X eggs carrying 30 chromosomes hexaploid crosses resulted in the produced by the pentaploid fe expected chromosome numbers. male parent. Similarly, eggs car Excluding the contribution of 45 rying three sets of chromosomes chromosomes by C. miyagii, the would have functioned in the de other combinations would require velopment of the hexaploid hy a degree of gametic variability in brids. chromosome number unsupported It seems quite unlikely that from the results of hexaploid X the hybrid with 105 chromo hexaploid crosses. Therefore, the somes, exactly seven times the most likely explanation of the ori basic number, could have arisen gin of the hybrid with 105 chro from two gametes carrying un mosomes is the union of an egg balanced chromosome numbers. carrying 60 chromosomes from T. Assuming the functioning of ga virgata with a normal sperm metes carrying multiples of the from C. miyagii. basic number, the following pos Studies of chromosome mor sible chromosomal contributions phology were intended primarily exist, with the chromosomes from to ascertain whether karyotype the female parent listed first: 75 analysis in Camellia would be X 30, 60 X 45, 45 X 60, 30 X fruitful in distinguishing species 75, and 15 X 90. The last combi and in elucidating phylogenetic nation would require the union of relationships. Thus, representa a monoploi'd egg from the pentn tive diploid species were investi ploid T. vi'rgata and an unreduced gated without attempting to es sperm from C. miyagii. Only 1 tablish fine differences through a percent unreduced polien was de statistical approach. tected in C. 111;iyagii. Further Some differences were noted in STUDIES WITH CAMELLIA AND RELATED GENERA 85 chromosome size. Particularly ob largest differences in centromere vious wa!'l the shorter length of position appear to exist between the chromosomes of C. saluenen chromosomes 13 and 15, where sis. This condition may have been the contrast in arm length is more apparent than real and the greater in C. rusticana for chro result of greater colchicine pene mosome 13 and greater in C. ja tration of these root-tip prepara ponica for chromosome 15. tions. The main differences en Species known to be closely re countered were in relation to cen lated are C. japonica, C. pitardii tromere position and in one case var. 1Jita?'dii, C. nlsticana, and C. to the presence of a microsatel saluenensis. Sealy (1958) places lite. these species among others in the Although there were striking subgeneric section Camellia. similarities between the genomes Therefore, the karyotypes of of all of the species in respect to these species would be expected to centromere position, there were be more similar to each other also marked differences, indicat than to other species. Ignoring ing that the species can be distin possible chromosome size differ guished from each other through ences, this is certainly true for C. careful karyotype analysis. saluenensis, whose long arm to There are a number of serious short arm indices (table 14) com limitations in comparing similari pare quite favorably with those ties and differences of karyotypes of both C. japonica and C. rusti of various species with known ex cana. Major differences in cen ternal morphological relation tromere position are apparent only ships. However, such comparisons in chromosomes 4 and 13. The do offer one more tool in the clas karyotype of C. pitardii var. pi sification of species and may help twrdii, however, appears dis clarify relationships that cannot tinctly different from those of be detected by the taxonomist. other species within section The karyotypes of C. japonica Camellia. At least six chromo and C. 1'usticana, appeared the somes (1, 5, 9, 11, 12, and 14) most similar among the eight appear quite distinct from their species studied. Here, the differ counterparts in the other species. ences are perhaps more surpris Among the species outside the ing than the similarities. Some section Camellia, C. kissi and C. taxonomists, including Sealy lutchuensis appear more similar, (1958), consider C. 'fusticana to at least superficially, to those be a subspecies of C. japonica. within this section than do C. sal The ve;.;,etative morphology of the icifolia or C. sinensis. C. lut two species is rather similar. chuensis hybridizes fairly readily Comparison of the indices of with species of the section Camel tables 12 and 13 shows a number lia, but many of the hybrids are of minor dissimilarities, but the largely sterile. The karyotype of 86 'rECHNWAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

G. salicifolia is perhaps the most otypes of the various sp~cies and dissimilar in relation to all other their ability to hybridize and pro species. Also, here the first evi duce fertile hybrid plants. The dence was found of a chromosome lack of more positive correlation pair with a satellite. would appear to indicate that structural and genetic differences In general, there was only a affecting hybridization exist, rough correlation between the which are not visible in the kary visual appearance of the kary otypes.

SUMMARY

Twenty species were used as parents. The identification of 211 parents for interspecific crosses in interspecific and 11 intergeneric Camtellia, resulting in 459 hybrids hybrids was based on vegetative from 106 combinations. Analysis characters. Certain characters by chi-square disclosed that one were found to be strongly domi parent was superior in six of 54 nant. Among these, the most use reciprocal crosses in the numbers ful were pubescent stems of cer of hybrids obtained per pollina tain species, the indented midrib tion, Intergeneric crosses were and veination of the leaves of C. also made involving Cmneliia and g?'anthamiana, and the narrow, F1'anlclinia alatamaha, Stewm·tia elongated leaf character of C. ovata, Tutche'l'ia, spectabili8, and hongkongensis. T. virgata. Eleven hybrids were Floral characters proved to be obtained from crosses of Camellia a valuable supplement to vegeta species with T'utchel'ia spectabilis tive characters. Certain species and T. vi1'gata, and six as yet un transmitted their floral charac validated plants were obtained ters more strongly th::-.n others. C. from F tanklinia :tlatamaha X gmnthamicma transmitted frilled Camellia honglcongensis. petals to the majority of its pro In a substantial number of geny, C. hongkongensis transmit crosses, seed capsules developed ted trumpet-shaped flowers and to apparent maturity, but Con rough, grey perules to many of tained partly developed or its progeny. Floral scent is re aborted embryos. Many of the stricted to seven species, C. f'ra former were grown to normal te?'na, C. kissi, C. lutchuensis, C. plants through the use of embryo miycLgii, C. oleifera, C. sasanqua, culture techniques. and C. tsaii. As an inherited The first evidence of validity of character, floral scent was trans a young hybrid plant is the com mitted to most of the progeny of parison of its vegetative morphol these species, ogical characters with those of its The use of hybrids in the inter STUDIES WITH CM,IELLIA AND RELATED GENERA 87 specific or intergeneric transfer each parental combination. Pollen of desirable characters depends sterility among the hybrids was upon fertility. Also, a greatly, in also considered, but these data creased frequency of aborted pol were far from complete because len produced by a presumptive many hybrids were too young to hybrid in comparison with fre flower. quencies produced by the parents Intrasectional hybridization provides corroborative evidence within section Camellia showed a of hybridity. Thus, percentages of dose compatibility relationship aborted pollen, after staining between C. japonicct, C. pitardii with acetocarmine, were deter var. pitarciii, C. 'reticulata, C. mined for all parental species and ntst'icana, and C. saluenensis. C. those hybrids that flowered. Al hong/congensis, on the other though the percentages of hand, was more difficult to hybri aborted pollen in the hybrids gen dize with other members of this erally exceeded those in the pa section, indicating a more distant rental species, enough fertility relationship. Species within sec was indicated to suggest the use of tion Paracamellia hybridized the hybrids in the transfer of de more readily with each other than sirable characters. Observation of those of any other section. In con the pollen also disclosed that the trast, species of section Theopsis production of large, presumably were the most difficult to hybri unreduced grains is very common dize with each other. Two species in Ca?nellia. Only five of 66 paren of Dubiae, C. hiemalis and C. m'i tal species or clones observed did yagii, very readily hybridized in not produce at least some unre either reciprocal direction. duced pollen. Unreduced pollen Analyses of the intersectional was found in 112 of 134 interspe compatibility relationships sug cific hybrids examined. Tutche1'ia gest that the seven sections fall spectabilis and T. vi1'gata also into two more-or-Iess distinct produced unreduced pollen. Un groups. In the first group, species reduced pollen was found in five of Paracamellia, Thea, and Du of the eight intergeneric hybrids biae appear to be more generally of Tutchel'ia and Camellia that related to each other than with flowered. species of the other sections. In Species compatibility relation the second, species of Theopsis ships were based on Sealy's and Cammeliopsis appear to be (1958) division of the genus into sections. Representatives of six closely related to each other and, subgeneric sections and the un in turn, more closely compatible placed group Dubiae were studied with Camellia than with other for the number of valid hybrids sections. Heterogenea would ap in relation to the total number of pear to fall within this second reciprocal cross-pollinations for group because of its closer com 88 TECHNICAL BULLETIN 1427. U.S. DEPT. OF AGRICULTURE patibility with Camellia than However, the chromosome num wit.h any other section. bers of its hybrids were unpre Root-tip chromosomes were ex dictable. A hybrid of C. sasanqua amined in parental species and 'Narumi-gata' (5x=75) X C. hybrids. Chromosome numbers not granthamiana (4x=60) was tet previously reported were deter raploid. A hybrid of C. sasanqua mined for the diploid C. lutchuen 'Narumi-gata' X C. 7'eticulata sis and the hexaploids C. miyag'ili (6x=90) was hexaploid. Among and C. 7'osaejlora. T. vi?'gata, pre 15 hybrids studied from crosses viously reported as diploid., was between hexaploid species, all but found to be pentaploid. one were hexaploids. One hybrid Determinations of chromosome with an aneuploid chromosome numbers were made for 88 inter count of 86 was observed among specific and 11 intergeneric hy the four hybrids of C. sasanqua brids that produced root tips suit X C.1niyagii. abll1 for study. Chromosome Three hybrids of Tutche1"ia Vi7' counts of 43 hybrids resulting gata (5x=75) XC. gmntha from crosses between diploid miana (4x=60) had 75 chromo species disclosed one variant from somes. Among the hybrids of T. the diploid number. An aneuploid vi7'gata X C. miyagii (6x=90), hybrid with 28 chromosomes two hybrids had 75 chromosomes occurred from a cross of C. ja and two had 90. Individual hy ponica X C. lutc/mensis. brids with 80, 82, and 105 chro Hybrids resulting from crosses mosomes also occurred. The one between diploid femaie parents hybrid obtained from the diploids and tetraploid male parents re C. pitardii val'. pitardii X T. sulted in triploids in four out of S1Jectabilis was also diploid. five hybrids studied. One hybrid The probable origins of the of C. pitm'dii val'. pita1'dii X C. complements of euploid variants gmnthan/'iana was tetraploid. All among the interspecific and inter but two of 18 hybrids from generic hybrids are discussed. crosses between diploid female 'l'he functioning of unreduced ga parents and hexaploid male par metes is indicated in three inter ents were tetraploids. Two hy specific hybrids. brids with aneuploid chromosome Studies of chromosome mor counts were observed; one had 56 phology were intended primarily chromosomes, and the other 59. to ascertain whether karyotype Ten hybrids from crosses be analysis in Camellia would be tween hexaploid female and di fruitful in distinguishing species ploid male parents were tetraploid and in elucidating phytogenetic and two hybrids were pentaploid. relationships. The major charac C. sasanqua 'Narumi-gata,' a pen teristics used to identify individ taploid variety, produced hybrids ual chromosomes were its total when used as the female parent. length and the position of the STUDIES WITH CAMELLIA AND RELATED GENERA 89 centromere. Accurate identifica chromosomes 4, 11, and 13. How tion of sets of homologs was not ever, C. pitwrdii val'. pita1'dii dif possible for polyploid species, so fered from C. japonica and C. 7''US the study was confined to eight tic ana in the centromere position diploid species. of six chromosomes. Although there were striking Among the species outside the similarities between the genomes section Camellia, C. kissi and of all of the species in respect to C. lutchuensis appear more simi centromere position, there were lar, at least superficially, to those also marked differences, indicat within this section than do C. sal ing that the species can be distin 'icifolia and C. sinensis. The kary guished from each other through otype of C. salicifolia showed the careful karyotype analysis. least similarity to those of the Species known to be closely re other species. lated are C. japonica, C. lJ'ita1'dii In general, there was only a var. pitm'dii, C. 1"ust'icana, and C. rough correlation between kary saluenensis. The major differ otypes of the various species and ences in the karyotypes of C. ja their ability to hybridize. The ponica and C. ntsticana appeared lack of more positive correlation in the position of the centromere in chromosomes 13 and 15. The indicates that structural and ge karyotype of C. saluenensis dif netic differences affecting hybri ferE~d from that of C. japonica dization exist, which are not visi and C. 7'usticana primarily in ble in the karyotypes.

LITERATURE CITED

ACKERMAN, W. L. 1963. INTERSPECIFIC HYBRIDIZING OF CAMELLIAS. Amer. Camellia Yearb. 1962-63: 113-119.

1970a. INTERSPECIFIC HYBRIDIZATION OF CAMELLIA. Amer. Camellia Yearb. 1970: 65-79.

1970b. INTERGENERIC ~jYBRIDS OF CAMELLIA WITH RELATED GENERA. Amer. Camellia Yearb. 1970: 100-105. ANDERSON, E. B. 1961. CAMELLIAS. 115 pp. Blandford Press, London. BAMMI, H. K. 1965. CYTOGENETICS OF RUBUS. IV. PACHYTENE MORPHOLOGY OF RUBUS PARVIFOLlUS L. CHROMOSOME COMPLEMENT. Canad. Jour. Genet. and Cytology 7 :254-258. BROOKS, H. J., and "MEHLQUIST, G. A. L. 1961. INHERITANCE IN THE CARNATION (DIANTHUS CARYOPHYLLUS). VI. TRIPLOID AND ANEUPLOID PRODUCTION. Amer. Soc. Hort. Sci. Proc. 77: 544-551. 90 TECHNICAL BULLETIN 1427, U.S. DEP'l'. OF AGRICULTURE

DARLINGTON, C. D., and JANAKI AMMAL, E. K. 1945. CHROMOSOME ATLAS OF CULTIVATED PLANTS. 397 pp. University Press, Aberdeen, England. --- and LACOUR, L. F. 1960. THE HANDLING OF CHROMOSOMES. Ed. 3, 284 pp. Allen and Unwin, Ltd., London. --- and WYLIE, A. P. 1961. CHROMOSOME ATLAS OF FLOWERING PLANTS. 519 pp. Allen and Unwin, Ltd., London. EINSET, J. 1952. SPONTANEOUS POLYPLOIDY IN CULTIVATED APPLES. Amer. SOC. Hort. Sci. Proc. 59: 291-302. ELLIOTT, F. C. 1958. PLANT BREEDING AND CYTOGENETICS. 395 pp. McGraw-Hill Book Co., New York, N. Y. EMSWELLER, S. L., and UHRING, J. 1960. BREEDING LILlUM LONGIFLORUM AT THE TETRAPLOID LEVEL. Amer. Soc. Hort. Sci. Proc. 75: 711-719. HILSMAN, P. L. 1966. INTERSPECIFIC CAMELLIA HYBRIDS. ArneI'. Camellia Yearb. 1966: 113-142. HOOKER, J. D., and JACKSON, B. D. 1895. INDEX KEWENSIS, WITH SUPPLEMENTS 2 THROUGH 13 (1896-1960). Clarendon Press, Oxford, England. JAGATHESAN, D., and SREENIVASAN, T. V. 1967. CYTOGENETICAL STUDIES IN NARENGA PORPHYROCOMA. I. STUDY OF KARYOTYPE AND ABNORMALITIES IN MEIOSIS. Cytologia 32 (1) : 11-18. JANAKI AM MAL, E. K. 1952. CHROMOSOME RELATIONSHIP IN CULTIVATED SPECIES OF CAMELLIAS. ArneI'. Camellia Yearb. 1952: 106-114. KANEKO, K. 1966. CYTOLOGICAL STUDIES ON SOME SPECIES OF HOSTA. I. KAROTYPES OF H. MONTANA, H. LANCIFOLIA, H. CHIBAI, AND H. CAPITATA [In Japanese with F.nglish summary.] Bot. Mag. [Tokyo] 79: 131-137 KARASA W A, K. 1932. ON TRIPLOID THEA. Bot. Mag. [Tokyo] 46: 458-460.

1935. ON THE SOMATIC NUMBER OF TRIPLOID THEA. Jap. Jour. of Genetics XI (6): 320. KAWANO, S., IHARA, M., SUZUKI, M., and ILTIS, H. H. 1967. BIOSYSTEMATIC STUDIES ON MAIANTHEMUM (LILIACEAE-POLYGO- NA~AE). I. SOMATIC CHROMOSOME NUMBER AND MORPHOLOGY. Bot. Mag. [Tokyo] 80: 345-352. LA;\;Ii\IERTS, W. K 1950. THE USE OF EMBRYO CULTURE IN GERMINATION OF CAMELLIA SEEDS, In Camellia Research, Camellia Res. Comm., ed. pp. 7-8. South Calif. Camellia Soc., Pasadena, Calif. LONGLEY, A. E., and TOURJE, E. C. 1959. CHROMOSOi\IE NUMBERS OF CERTAIN CAMELLIA SPECIES AND ALLIED GENERA. Amer. Camellia Yearb. 1959: 33-39. STUDIES WITH GMrIELLIA AND RELATED GENERA 91

1960. CHROMOSOME NUMBERS OF CERTAIN CAMELLIA SPECIES AND ALLIED GENERA. ArneI'. Camellia Yearb. 1960: 70-72. MALLICK, R., and SHARMA, A. K. 1966. CHROMOSOME STUDIES IN INDIAN PANDANALES. Cytologia 31 (4) : 402-410. MEYER, 1<'. G. 1959. PLANT EXPLORATIONS-ORNAMENTALS IN I'rALY, SOUTHERN FRANCE, SPAIN, PORTUGAL, ENGLAND, AND SCOTLAND. U.S. Agr. Res. Servo ARS 34-9, 180 pp. MORINAGA, T., FUKUSHIMA, E., KANG, T., and others. 1929. CHROMOSOME NUMBERS OF CULTIVATED PLANTS. II. Bot. Mag. [Tokyo] 43: 589-594. ---and FUKUSHIMA, E. 1931. CHROMOSOME NUMBERS IN CULTIVATED PLANTS. 1Il. Bot. Mag. [Tokyo] 45: 140-145. MUNTZING, A. 1961. GENETIC RESEARCH. Lts. Forlag. Stockholm. 345 pp. MURIN, A. 1962. CBROl\10S0ME STUDY IN ALLIUM STRICTUM SCHRAD. CaryoloO'h 15 (1): 139-142. NAKAJIMA, G. 1963. KARYOTYPE OF GENUS IPOMOEA. Cytologia 28 (4): 351-359. OLMO, H. P. 1952. BREEDING TETRAPLOID GRAPES. ArneI'. Soc. Hort. Sci. Proc. 59: 285-290. OURECKY, D. K. 1966. CHROMOSOME MORPHOLOGY IN SAMBUCUS CANADENSIS VAR. MAXIMA SCB., THE AMERICAN ELDERBERRY. Canad. Jour. Genet. and Cytology 8(2): 188-191. PARKS, C. R., and GRIFFITHS, A., JR. 1966. EXPERIMENTAL TAXONOMIC STUDIES IN SECTION CAMELLIA OF THE GENUS CAMELLIA. Amer. Jour. Bot. 53: 636. PATTERSON, E. B., LONGLEY, MARY 0., and ROBERTSON, D. S. 1950. CHROMOSOME NUMBERS IN CULTIVATED CAMELLIAS. ArneI'. Camellia Yearb. 1950: 107-113. SANTAMOUR, F. S., JR. 1963. CYTOLOGICAL STUDIES IN THE THEACEAE. Morris Arboretum Bu\. 14: 51-53. SAVIGE, T. 1967. CAMELLIA SPECIES AND HYBRIDS. ArneI'. Camellia Yearb. 1967: 77-100. SEALY, R. J. 1958. A REVISlON OF THE GENUS CAMELLIA. 239 pp. Spottiswoode, Bal lantyne & Co., Ltd., London and Colchester. SErTLE, W. J. 1967. THE CHROMOSOME MORPHOLOGY IN THE GENUS SILPHIUM (COMPOSI TAE). Ohio Jour. Sci. 67(1): 10-19. SHINDO, K., and KAMEMOTO, H. 1963. KARYOTYPE ANALYSIS OF SOME SPECIES OF PHALAENOPSIS. Cytologia 28 (4): 390-398. 92 TECHNICAL BULLETIN 1427, U.S. DEPT. OF AGRICULTURE

SHINDO, K., and KAMEMOTO, H. 1963. KAROTYPE ~NALYSIS OF SOME SARCANTHINE ORCHIDS. Amer. Jour. Bot. 50 (1): 73-79. SIMURA, T. 1935. CYTOLOGICAL INVESTIGATIONS IN TEA PLANT. Crop Sci. Soc. Japan Proc.: 7: 121-133. [In Japanese with summary in Esperanto.] SNEDECOR, G. W., and COCHRAN, W. G. 1967. STATISTICAL METHODS. Ed. 6,534 pp. Iowa State Univ. Press, Ames. TUKEY, H. B. 1934. ARTIFICIAL CULTURE METHODS FOR ISOLATED EMBRYOS OF DECIDUOUS ~'RUITS. ArneI'. Soc. Hort. Sci. Proc. 32: 303-322. ZEILINGA, A. E., and KROON, G. i1. 1965. A METHOD ~'OR MAKING ROOT-TIP SQUASHES PERMANENT, WITHOUT REMOVAL OF THE COVER SLIP. Euphytica 14: 36-38. APPENDIX

T,\I3LE 20.-Intcrspeci[zc crosses in Camellia 1'esuZting in no hybrid plants rn 1-3 c::! t:j Interspecific crosses Polli Seed capsules Seeds started ..... t.:r.l nations produced in culture en

~ ..... PzrCIl11tper Percent per 1-3 Number Number pol/ination Number pollinati911 .....'"';

0 0 &: C. ira/ema X C. gran/hamiana --- 63 2 3.2 0 0 ~..... C. ira/ema X C. hiemalis__ _ 20 1 5.0 0 0 C. ira/ema X C. h01!gkollgensis_ - 18 0 0 ~ 0 0 ReciprocaL __ 15 0 0 t: 0 0 ;t. C. ira/ema X C. kissi__ 25 0 0 25 4 16.0 0 0 > ReciprocaL __ _ Z t:j C.iraiema X C.miyagii ______.______------49 3 6.1 2 4.1 42.1 1 5.3 ~ C. iraiema X C. oleiiem__ ------ 19 8 t.:r.l 0 0 t"' C.iratema xC.pitardiivar.piiardiL ___ ------25 0 0 > 26 0 0 0 0 1-3 C. iraiema X C. rosaejiora__ - - . ------.. ------ 50 4 8,0 2 4.0 t.:r.l C.iratema XC. rusiieana ______- .------.. t:j 4.0 2 8.0 C.iraiema XC. salieifolia____ --.------.---- .. --_.- 25 1 o t.:r.l 18 1 5.6 3 16.7 Z C. iraiema X C. saluenensis_ ------t.:r.l C. ira/ema X C. sasal/qua______-- -- - _. ------. --- 34 1 2..9 0 0 0 0 ~ ReciprocaL______------ 31 0 0 > 0 0 0 C. iraiema X C. sinensis_ -- -- -. ------ 27 0 C. ira/ema X C. taliensis______---- 16 0 0 0 0 11 0 0 0 0 C. granthamiana X C.japoniea______~ ~ TABLE 20.-Inte1·specific crosses in Camellia resulting in no hybrid plants-Continued c:o ~

Interspecific crosses Seed capsules "'3 Polli Seeds started l:J nations produced in culture o =Z o.... Percent per Percent per > Number Number pollination Number pollination t"' C. granthamiana X C. miyagiL ______• ______18 0 ~ Fteciprocal______. ______0 0 0 C 54 4 7.4 3 5.6 t"' C. granthamiana XC. oieifera______• ______16 0 0 0 0 t"' Fteciprocal__• ______. ______t ,-'I C. hiemalis XC. japonica______• ______26 2 7.7 0 0 C. hiemalis XC. lutchuensis______17 0 0 0 0 c FteciprocaL______en 29 0 0 0 0 C. hiemalis XC. pitardii var. pitardii ______11 1 9.1 0 0 t1 C. hiemalis X C. reticulata______23 0 l:J lteciprocal______0 0 0 "tj 4 0 0 0 0 !-3 o C. hiemalis X C. sinensis______12 0 0 0 0 ~ C. hongkongensis XC. frGtema ______15 1 6.7 0 0 > C. hongkongensis X C. granthamiana______20 2 10.0 2 10.0 Q ::0 C. hongkongensis XC. hiemalis______..... 16 1 6.3 0 0 o C. hongkongensis X C. japonica______18 0 0 0 0 C C. hongkongensis XC. kissi. ______.. ______~ Fteciprocal______• ______• ______11 0 0 0 0 c:: 31 2 6.4 1 3.2 ::0 l:J C. hongkongensis X C. lutchuensis ______.. ______• ______.. ___ _ 12 0 0 0 0 Fteci pro cal _ . ______.. 12 0 0 0 0 C. hongkongensis X C. miyagii______.. _____ .•• ______• ______21 0 0 0 0

C. hongkongensis XC. oleifera. ___ _• ______. 24 o o o o ReciprocaL ___. ______. ___ . 42 2 4.8 o o C. hongkongensis X C. pitardii var; pitardii. ___.. _____ . _. ______14 o o o o ReciprocaL__ .. ______25 o o o o C. hongkongensis saluenensis______36 XC. 2 5.6 '0 o CIJ 1-,3 q C. japonica XC. oleiJera______18 2 11.0 1 5.5 tj ReciprocaL_ _. ______28 ..... 1 3.6 o o t:tj C. japonica X C. rosaeflora. __ •. ______•______• __ 24 o o o o CIJ ReciprocaL___ .. ___ . ______24 o o o o ~ C. japonica X C. saliciJolia_. ______16 ..... ••• o o o o 1-,3 p:: C. japonica XC. sasanqua______29 2 6.. 9 2 6.9 Q ;l>. C. kissi X C.fraterna_ .. ___ • __•• ______25 1 4.0 o o i::: C. kissi X C. granthamiana______66 5 7.6 7 10.6 G. kissi X C. lutchuensis. ______108 31 28.7 o o ~ Reciprocal______21 o o t-o o o ~ C. kissi XC. pitardii var. pitardii.______17 1 5.9 o -0 C.lutchuensis X C. granthamiana______14 o o o o > tjZ ______C. lutchuensis X C. hiemalis______29 2 6.9 1 3.4 ~ C.lutchuensis X C. hongkongensis______12 t:tj o o o o t"' C. lutchuensis X C. miyagii______49 o o o o > C. lutchuensis X C. oleifera______26 1-,3 o o o o t:tj Fteciprocal______65 6 9.2 3 4.6 tj Q C. lutchuensis X C. reticulata______38 3 7.9 2 5.3 tz:j ReciprocaL______.______55 Z 3 5.5 o o tz:j C.lutchuensis X C. saluenensis______14 o o o o ~ C. lutchuensis X C. sasanqua______32 o o o o > Reciprocal ______. 12 o o o II

C. lutchuensis X C. sinensis_____ 26 o o o o ~ ReciprocaL ______23 2 8.7 o o 01 TABLE 20.-Inte'rspecific crosses in Camellia resulting in no hybrid plants-Continued ~ ~

Interspecific crosses Polli Seed capsules 1-3 Seeds started l:j nations produced in culture (") II: Z .... Percent per Percent per (") Number Number pollination Number pollination ~ t"' C. miyagii X C. japonica_", _. __ .. _. ___ • ______21 0 0 0 0 t:I:j C. miyagii XC. rosaefiora ______.• _____ • ______30 2 6.7 0 0 C Fteciprocal_____ ••• ______t"' 23 0 0 0 0 t"' l:j C. miyagii XC. rusticana ______.•. ______1-3 .. 42 5 11.9 0 0 .... FteciprocaL___ . ______. ______. . • .. ______26 1 3.8 0 0 Z C. miyagii X C. sinensis ______._ ...... ______I-' 17 6 35.3 0 0 C.olei/era X C. grantltamiana ______"'"~ 12 1 8.3 ~-'I C. olei/era X C. japonica ______. ______0 0 28 1 3.6 2 7.1 c C.olei/era XC. pitardii var. pitardii.______27 2 7.4 0 0 ~ C.pitardii var. pitardii X C. rosaefiora ______I:j 20 0 0 0 0 l:j C. pitardii var. pitardii X C. salici/olia______22 0 0 0 0 "tI C. pitardii var. pitardii XC. taliensis______25 0 0 0 0 ~ C. reticulata XC. hiemalis____ . ______o 14 0 0 0 0 "t:r.j C. reticulata X C. hongkongensis______• ______,-:;. 19 2 10.5 0 0 C. reticlllata X C. kissi. ____ • ______~ 28 3 10.7 2 7.1 C) ~ C. reticulata XC. miyagii______..... 11 0 0 0 0 (") C. retic1l1ata X C. salici/olia______. ______22 0 0 0 C C. reticulat,), xC. sasanqua______.______0 t"' 31 0 0 0 0 1-3 C. rosaefiora XC. granthamiana______. ______26 C 0 0 0 C C. rosaefiora XC. hongkongensis______21 0 0 0 0 ~ C. rosaefiora X C. kissi______. ____ _ 83 2 2.4 0 0 C. rosaeflora X C. lulchuensis•.•. 24 0 0 0 0 C. rosaeflora X C. reticulala._ . .. 34 0 0 0 0 C. ruslicana X C. hiemalis.__ . 19 1 5.3 0 0 C. ruslicana X C. reticulala .. 48 0 0 0 0 29 0 0 0 0 Ul C. rusticana X C. sasanqua. 8 ReciprocaL_. _ 22 0 0 0 0 q t::l ...... C. rusticana X C. sinensis.... -_ .• ___ ... __ •. ____ .. _.• _...... 23 0 0 0 0 l".l ReciprocaL._.. . __ . ______.•. _ .... -.,. 18 1 5.6 0 0 r:tJ. 0 0 0 0 .~ C. salicifolia X C. pilardii var. pitardii____ -. _. ------ -. -- -- 16 ...... C. saluenensis xC. salicifolia. ______._. ______.. -- --.----. 20 0 0 0 0 8 ::t: C. sasanqua X C.. hongkongensis______22 0 0 0 0 C'J :t:. 0 0 C. sasanqua XC. pitardii var. pitardiL _ _ _ _. - . - - - - -.. ------... -- . - . - 20 0 0 ~ ______12 0 0 0 0 ttl C. sasanqua XC. rosaeflora_. ._. ______------.------ -. --- t"I (I 0 C. sasanqua XC. saluenensis____ • _• _______.. -------• --------. - 33 2 6.1 t:-< C. sasanqua X C. sinensis___ . ______·_ 12 1 8.3 0 0 ~ C. sasanqua X C. laliensis______• ____ - -. ___ ____ - ----- ---.--- - -- - -- 20 0 0 0 0 > Z t:i C. sinensis X C. l;fmgkongensis_____ - - _- - - - - -- -- - -- 30 2 5.7 0 0 C. sinensis XC. iulchuensis______.-_. __ _ 3~ 5 15.2 3 9.1 l:tl l".l C. sinensis XC. pilardii var. pitardii______16 0 0 0 0 t" C. sinensis XC. reliculala ____ . ______20 1 5.0 0 0 > 8 C. sinensis X C. TlIslicana ______- ------ -- 18 0 0 0 0 l".l t::l Q l".l Z l".l l:tl >

~ ..;:j TABLE 21.-Vegetative cha?'acters of parental species of Camellia useful f01' identification of hybrids to 00

Twigs and young shoots Leaves ~ t.:r:j Terminal Species 0 Habit Color Color Lower surface vegetative Pubescence of of new =Z of upper MHgins buds ...... growth growth surface Midrib Veins a II> t" C. Ira/ema. __ ". _____ .' ___ Pubescent Thin, Pink to Dull, Raised, Obscure Basal U Medium t:I:1 willowy, red grey fairly q smooth, long, t"' spreading green prominent bluntly pointed, t"' t.:r:j ::Ierrulate C. granlhamiana______pubescent. ~ ••• Pubescent Medium Light Glossy, ...... Very Very Bluntly Medium Z thin, green bright highly highly serrulate short, inter .... green raised, raised, bluntly ~ mediate prominent prominent pointed, :'I pubescent. q C. hiemalis______Pubescent Medium Light Glossy, Slightly Apparent Mostly Long, rn stout, green bright raised, as dark crenate, sharply b inter green apparent lines some pointed, t.:r:j mediate only serrate "0 C. hongkongensis______• __ pubescent. ~ Glabrous Stout, Dark Dull, Highly Slightly Very Short, upright 0 purple grey raised, raised, bluntly stout, I>j green prominent dark shallow, bluntly II> lined crenate pointed, Q ~ C. japonica______glabrous...... Glabrous Stout, 0 Bright Glossy, Raised, Not Bluntly Long, q upright green bright prominent raised, denticu sharply t"' green but late to pointed, q>-'3 apparent crenate glabrous. ~ denticulate t.:r:j C. kissi ..•• l'ubescent Medium Pink to Semi Raised, Apparent Basal U Short, thin, red glossy, prominent as dark to }~ stout:at inter medium lines only smooth, basE', en serrula~esharply mediate green >-3 pointed, d t1 pubescent...... t:.l C. lulchuensis Pubescent Thin, Pink to Dull, Not Obscure Bluntly Short, en willowy, red olive raised, serrulate bluntly ~ spreading green but pointed, ...... apparent pubescent. >-3 ::t: Pink Semi Not Obscure Crenate Short, C. miyagii_ . Pubescent Thin, C':l or glabrous spreading glossy, raiged, stout, :z,. medium but bluntly ~ green apparent pointed, ~ glabrous. t-<.... C. olei/era ___ Mostly Thin, Light Semi Slightly Obscure Serrate Shori, ~ pubescent spreading green glossy, raised, stout, ~ bluntly Z bright apparent t1 green poiri'ted, ~ pubescent. l".! Glossy, Raised, Raised, Serrulate Short, t" C. pilardii var. p1·tardii._ .. Glabrous Very Light > stout, green bright prominent dark to serrate stOll.t, >-3 upright green (lned, bluntly t:.l t1 prominent pointed, 0 pubescent. t:.l Z C. ret1·culata______. Pink Semi Raised, Raised, Basal U Short, t:.l Glabrous Stout, ~ upright, glossy, prominent dark smooth, stout, > rigid, dark lined, serrulate bluntly open green prominer,:; pointed, glabrous. ~ ~ TABLE 21.-Vegetative characte1'S of parental species of Camellia useful for identification of hybrids~Con,~ 0 0

Twigs and young shoots Leaves 1-3 Terminal t=j Species a Habit Color Color Lower surface vegetative lJ:: Pubescence of of new of upper Margins buds Z..... growth growth surface Midrib Veins a >t"' C, rosaeflora ______•• _. Pubescent Medium Pink Glossy, Slightly Obscure Widely Medium Cd thin, lax, dark green raised, q crenulate long, t"' straggly apparent serrulate, pointed, t"' l:tj teeth 2-3 pubescent. 1-3 mm. apart ..... C. rusticana______. ___ • __ Z Glabrous Stout, Bright Glossy, Raised, Not Coarsely Medium .... upright green dark green -.::.. prominent raised, but crenate long, r-:> apparent to serrate sharply ~""'I

pointed, ~ glabrous. ?2 C. saUcijolia __•..•• ___ .• Pubescent Thin, Pink to Semi Raised, Raised, Widely and Short, tl willowy, red glossy, prominent prominent shaJlowlr slender, l:tj spreading, olive serrulate, sharply '"d :-'3 pendulous green 'teeth pointed, incurved, highly "j0 black pubescent. tipped ::> C. saluenensis. __ •.• __ •.• Pubescent Stout, Pink Semi Slightly Slightly Serrulate Medium ~.-. upright, glossy, a raised, raised, long, q densely olive apparent dark pointed, t"' leaved 1-3 green lined, pubescent. q apparent ::0 t::j c. sasanqua._ Mostly Stout, Light Glossy, Raised, Not Variable, Medium pubescent upright green bright fairly rabed, crl!nately short, green prominent darl:. serrate bluntly lined, pointed, apparent pubescent. Cf.I C. sinensis__ Mostly Medium Light Glossy, Highly Raised, Bluntly Long, qf-3 t;::l glabrous to stout, green dark raised, prominent serrulate pointed, ..... upright green prominent to sinuate- pubescent. I?j serrulate, Cf.I ~ teeth .... incurved, black ~ tipped ~ C. taliensis______._ Glabrous Stout Light Glossy, Raised, Raised, Bluntly Long, E::: green bright prominent prominent serrulate pointed, ~ green pu'b·escent. t< C. tenuijiora____ ..• Pubescent, Thin, Light Semi- Slightly Obscure Crenulate, Short, ~ then spreading green glossy, raised, denticulate bluntly > glabrous medium apparent pointed, Z green glabrous. t:I ~ t?-j Source: Information partly from narrative description in Sealy (1950) and partly from aui;hor's observations at Plant Intloduction t"' Station, Glenn Dale, Md. > f-3 l':j t:I !;1 l':j Z l':j >~

I-' o I-' TABLE 22.-Vegetative characters of Camellia used for initial identification of interspecific hybrids 1 ...... 0 ~ ~ Twigs and young shoots Leaves "'3 trJ Terminal .0 Code of hybrids Pubes- Habit Color Color Lower surface vegetative ;:r: Z cence of of new of upper Margins buds H growth growtb surface Midrib Veins .0 >t" C. iraterna X C. japonica: tl:l B-7 ______s m m f c:: C. japonica X C. iraterna: ~ 4-2______t".J m m m m s m m m >-3 4-4______H m m m m s f f m 4-5______Z m m m m s i m m .... 4-7______.::;. m m m m s m m m t.;) 4-8______m i f m s m m m :-'I 4-9______m m ------m s i f m 4-43______S m m r:. f m m 4-44 ______------Vi m m m m s f m 4-47______I:j m m m m s i m m t".J 4-48______s f m m "'d m m m m !"' C. iraterna X C. lutchuensis: 1-3______m 0 s s s m s ~ > C. lutchuens's XC. iraterna Q 2-1______f m ~ s s s m s H .0 c:: C. a-25fracterno.______X C. reticulata: t" s m s m s ""c:: ~ t".J C. 2-4reticuZala______X C. !ralerna: m s m s s m 2-5______m s m ;; m s m 2-7______•. m s m s m s m UJ 1-3 q t1 C. japonica X C. ! ranlhamiana: ..... 7-67 ______.-_- ____ . -- -- m s s m m m l".l 7-96 ______UJ m s s m m m B-3______. ______s s m m m m =S m ..... 1-3 tIl C. 5-84japonica______X C. hiemalis: m s s m m m C'::l ~ s:: C. japonica X C. hongkongen,,"is: t:t.l 5-10______t'< s s m m f m t'< 5-12______------ s s m m m m f m 5-37 ______~ s s m m f m 5-41______------ ~ s s m m f f f Z 5-47______- -- -- s s m m m f m t:l 5-48______s s m m m f m ::0 5-49______m l".l s s m m m f t"' ~ 1-3 C. 5-26japonica ______X C. kissi: l".l m f f f s m f m t1 5-27 ______m f m f s f m m c;'l I::l Z C. 3-6japonica______X C. lulchuensis: l".l m f m m m f m m ::0 2-7 ______~ m m m m f m m 3-8______m m m f m m 3-12______m m m m f m 1Ii 3-13______m ~ m m m f f 0 See footnote at end of table. ~ TABLE 22.-Vegetative characters of Camellia used tor initial identification of interspecific hybrids I-Con. I'-" o ~ Twigs and young shoots Leaves ~ l"j ------Tenninal Code of hybrids Pubes Habit Color Color Lower surface vegetative @ cence of of new of upper Margins buds ~

growth growth surface Midrib Veins .~ t"' C. japonica X C. lutchue1Ul1s--Con. : to 3-14______c:: 3-15______m m m m f m m E:: 3-16______m f f m f m m l"j m m m f m m ~ 3-17______...... 3-18______m m m m f m m Z m m m f m m ..... ~ 3-19______l\:I m m m f m m .-'1 3-20______.. ______3-21______m m m m f m m c:: 3-23______m f m m f f m ~ m m i f m m t:j 3-24 __.. ______m m m m f m m l"j 3-26______"t! m 3-28______m m m m m m ~ 3-29______m ------ m f f m m o 3-30______m m f f f m m f%j 3-32______m m m m f m m > m m m m f f m Q ::0 4-16______...... 4-19______m f i m f m m m f m m f m m ~ 4-20______~ m m m m f m 4-21 ______f c::: m m m m f f m ~ 4-23______I:t.1 4-31______m m m m f m m 4-32______m m m f f m m m m f m m 4-33______m m m f m m 4-35. ______.. ____ 4-36______m m m f m m m m m f m m 4-40______•______m m m m f m m rn 4-41 __• ______8 m m m f m m 4-42______.• ______~ m m m f m m I:::) 4-45______. ______. ____• ______H 4-46______m m m f f m t;l m m f m f m m rn 4~53 ______• ______m m m f m m ~ 4-54. ______..... m m m f f m 4-55 ______m "'3 m m f f m ::t:: 4-56______• ______m m m f m m 4-57 ______~ m m m m f a-: 4-58______- ______m f f m ~ 4-59 ______m m t-< m m f m m .... 4-60______m f ;l:. i f m m 4-61______m f m f m m > 4-62______m m m Z m f m I:::) 4-63______m m m m m m f m m ~ 4-64______t;l m m m f m m 4-65 ______m t"' m m f m m 4-66______m ------ > m m f m m t;5 4-67 ______m m m m m m f m m t1 q 4-68______m m m m t;l 4-69______m m m Z m f m .m 4-70 ______m m t;l m m f m m ~ 4-71______m m m f m m > 4-72 ______m f m f m m 4-73______m -- -- -- m m m m m m f m ~ 0 See footnote at end of table. 01 TABLE 22.-Vegetative characters of Camellia used for initial identification of interspecific hybrids l-Con. f-.6 o C!) Twigs and young shoots Leaves 8 l?:1 ------Tenninal (') Code of hybrids Pubes Habit Color Color Lower surface vegetative cence of of new of upper Margins buds =~ o growth growth surface Midrib Veins :> t"' c. japonica X C. lutchuensis--Con. : b:l 4-74______c::: m m f t"' 4-75______m f m m t"" 4-76______m f m m f m m l':1 4-77______m m m i f m m ~ m i m f m m Z 5-3______~ m f m i f f m ,;.. 5-4______" __ r-:> m m m m f m m :'I c. lulchuensis X C. japonica: c::: 5-20______i:n f m In m 5-23 ______m f f f m m m m f f tl t:<.I "'d C. japonica X C. miyagii: !"3 5-7______m f f m f f o 5-9 ______I'%j m f m m f f m > C. japonica X C. pitardii var. pitardii: ~ P-2______..... P-3______5 f 5 5 5 m f m (') c::: 5 f ~ ·s 5 m m m t"' >-3 C. pitardii var. pilardii X C. japonica: c::: A-16 ______::0 5 5 5 5 5 m m f I:.:::j C. 3-22japonica ______X C. reliculala: 3-23______s m m m 5 m m m s m m s m m f 3-31______------.... P-I3 ______s r f m 5 r m f s f m s m f m P·-16. __ .____. ______Ul s m ------f 5 m m m 1-3 tjc::: C. reliculala X C. japonica: .... A-2 ______• ______.l'9 s m f f 5 f f m A-6 ______00 s m f m 5 f f m A-8______:g s m f 5 f m m .... A-9______1-3 s f In 5 f In f A-II ______I:I: 5 f f s m A-1a______In Q s m f s f f In ;t.. A-l4.______s m s m fi!:: f f m l>:j t-< t-< C. japonica X C. rosaeflora: P-46______m m f m f m ~ > C. japonica X C. rusticana: Z 5-13______tj s s s m s s m 5 5-14______l:tI 5 s 5 m 5 5 m s l'9 5-24______t"' 5-25______s s s m s s m s > s s !' In 5 5 In S 1-3 tjl".l C. rusticana X C. japonica: 5-28__~ ______!;') s s s f s s m 5 l::r:J Z l".l C. 5-1iaponica______X C. saluenensis: l:tI 5-5______m f f f m m m > 5-6______m i f I m m m m m f m f m m m m 5-42______m f f f f m m m I--' 0 See footnote at end of table. ~ TABLE 22.-Vegetative chamcters of Camellia used for initial identification of interspecific hybrids I-Con...... 0 (X) Twigs and young shoots Leaves 1-3 t.:;j Terminal 0 Code of hybrids Pubes Habit Color Color Lower surh~e vegetative ttl Z cence of of new of upper Margins buds ..... growth growth surface Midrib Veins Q >t"' C. japonica X C. saluenensis-Con.: b:j 5-44______q m f 5-45 ______f f m m m t< m f m m m m m m t"' 5-46______t.:;j m f m f m m m m 1-3 ..... Z C. saluenensis X C. japonica: 5-29______I-' f m ------m m f m f "'"~ .-'1 C. japonica X C. sinensis: q 5-17. ______f s s f m m m m ~ 5-22______m s s f f f f m 7-250______t! m s s m m f trJ m m "0 !""3 C. japonica X C. laliensis: P-37______0 s s s s s m m m I:;l > C. kissi X C. rusticana: 0 5-69 ______~ f m f m s m m f H Q q C. rusticana X C. kissi: t"' 5-51. ______I-j m m ------f s m m m q ~ C. lulchuensis X C. ruslicana: t.:;j 1-4______f m ------f m m m f C. Tusticana X C. lulchuensis: 1-1______1-3______m m f m m 1-5 ______m m m m f m m m f m m 2-2 ______.______f m m m 2-3 ______m m m m f m m 00 8 2-6______m m m f f f m q m m f 2-10______f m m t::::I m ..... 2-11 ______m m m i lil m to;! m m m f m m 00 ~ 3-9 ______..... 3-11 ______m m m m f f m 8 3-27 ______m m m f m m m tIl 4-1______m m m m f f m Q m m f f f f m :t> 4-3______•______~ 4-6______m m m f f m m ~ 4-10 ______m m m m f m m 4-13______m m i f f f m m m m f m m E ~ 4-14______Z 4-15______m m m f f f m t::::I 4-17 ______m m m m f m m ~ m m i f f m to;! 4-18______I:"" m m m 4-37 ______f f f m ~ m m m f f m m 8 4-38______to;! 4-39______m m f f m m t::::I 5-39______m f m m f m m I:;') m m m f m m to;! Z to;! C. miyagii X C. lutchu~nsis: 7-168______~ m s s s s m ~ 7-180______f 7-182______m s s m s s f m m s s m s m 7-248______s f I-l s s s m s s f m 0 See footnote at end of table. ~

--~ TABLE <~ 22.-Vegetative chamcters of Camellia used for initial identification of interspecific hybrids I-Con. ~ o Leaves 1-'3 Twigs and young shoots t.1j ------Terminal o Code of hybrids Pubes Habit Color Color Lower surface vegetative ::= Margins buds z cence of of new of upper ....o growth growth surface Midrib Veins ~ t"' to C. oleifera X C. hiemalis: d C-1 ______------s s s s m m m t"' t"' t.1j oleifera X C. miyagii: 1-'3 r. 5-60______.... f s n~ s m s f f 5-61______Z f s m s m s f f ... ~ ,:'I C. pitardii var. pitardii X C. fralerna: 7-159______m m f s f f i c:::: P-19______m m m s f f m rn t::J t.1j C. pilardii var. pilardii X C. qranlhamiana: 0,; B-6 ______m m s s m m s s P-56______;3 m f s s m m s s o ~ C. pilardii var. pitardii X C. IUlchuensis: P-8 ______~ m m f m i f s s Q P-9. ______m m f s s :E m m o c:::: C. pitardii Vl1r. pilardii X C. reticulata: t"' ______1-'3 P-24 _ m s s m s f c:::: ::t! t.1j C. reticulalo X C. pilardii var. pilard;:: F-55 ______s m m f s s m C. pilardH var. pilardii X C. salllenens'is: ·,12 p .• .... """,,,,,~, -_... ,.,.,,..- ...... "' ..._--- ...- ..- .. _- m m m s m

C. saluencnsis X C. pilardii var. pilariIi'i:

P~45 ~ '" .. -- ~ -,.. - .. >~ ,...... - -'< ... ~ -- ..., ~. ""' ...... - _ ...... 0- m m m s en >-:2 C. reticlIlq'(( X C. granthamiana: c:: ....t:::! B,·5 .. _.~._, ..' _. __ .••. , .... _".~_.___ ._. m m m ni m m t;tj U1 C. reticulata X C. rosaeflora: :!l P-57. __ ._ .,.. ____. ____ •______..... m s m m f m >-:2 ~ C. retjclllata X C. rusticana: Q 5-50______;t.. s m s m m f 1:: tt.l C. 5-58reticulata______X C. saluenensis: ______1:-1 ._...... t 5-65 ______m 5 m f s m m ;t.. m f s f m s m m 5-83______. ______:> m f s f f s m m Z t:::! C. rosaeflora X C. fralema: 7-266______•____ • ______?:i s m s m m s f s 7-·267. ______. ______~ s m s m m s f s :> 8 t;tj C. rllsticana X C. fralerna: 3-1 .. ______t:::! m 3-2_____._. __. ______m s f m c;') m m m s m m t'1j 4-49 ______Z m m m s m m t'1j l=d C. ruslicana X C. hongkungensis: :> 7-1______7-5______s s ------m m m f i s s m m m m f ------..... See footnote at end of table...... TABLE ...... 22.--Vegdat.ive chamcte1's of Camellia used for initial identification of interspecific hybrids I-Con. -...... '-"0 t-3 Twigs and young shoots Leaves t::l Terminal 0 vegetative ~ Code of hyhrids Pubes- Habit Color Color Lower surface Z of upper Margins buds .... cence of of new 0 growth growth surface Midrib Veins > t"' C:l C. rusticana X C. saluenensis: 4-77 ______q m ------m f m f m t"' t"' t::l t-3 C. 7-302sa'llenensis ______X C. Jralerna: .... s m s m f m s Z I-' C. salllenensis C. granthamiana: "'" X ~ B-2____ - ______s m m m s f ~ VJ C. P-48salucncnsis______X C. hongkongcnsis: m m m s { t:::;j t::l "'Ij C. salllcnensis X C. lulchuensis: ~ A-26______m s m s m m f s 0 "%j

C. P-4salucnensis ______X C. reticulala: > f s m m m f f 0 ~ P-5 ______m m m f f .... f s 0 c:: t"' C. 5-85salucnensis______X C. roseafiora: t-3 s m s s m m s C! ~ t::l C. 7-249saluenensis______X C. laliensis: - ______m --_.-<------m m m s C. sasanqua X C. granthamiana: A-24 ______"____• ______" m m s B-4 "__ . ______. ______s f s s m s s s s

C. 6-41sasanqua ______X C. hiemalis: _ rn s m s s m s m 8 c:: t1 C. sasanqua X C. japonica: ...... ___ t.".l P-33 . ______rn P-34.______f s s s s f m f f s s s s f m f =1l P-36 ______..... m s s s s f m f ~ C. sasanqua X C. miyagii: 5-52______--- ______f m m m m ~ 5-53______m f a: 5-55______f m m m m m s f f m m m m f 5-56______f f ~ f m m m m f s f ~ C. sasanqua X C. oleifera: > B-1 ______Z m s ------m m s m t1 ~ C. (Usanqua X C. reticula/a: t.".l A-2L ______t"' A-22 ______f m ------m s m m f > m m m m s m m f 8 A-23 ______t.".l f f m m s i m f t1 Q C. sasanqua X C. tenuijloTa P-29______t.".l s s m m m Z P-32______f f t.".l s s m m f ~ f f >

I m=resemblance to male parent; f=resemblance to female parent; i=intermediate characters; s=characters similar to both parents. I-l COI-l ..... TABLE 23.-Vegetative chamcte?·s of parental species of Tutcheria useful for identification of hybrids J-l ~ ~ Twigs and young shoots Leaves tzj Terminal o Lower surface vegetative III Species Habit Color Color Z buds ..... Pubescence of of new of upper Margins o growth growth surface Midrib Veins ~ t" to T. spec/abilis • ______• ___ Pubescent Stout, Red to Dull, Very Slightly Widely Long, very c::: t" upright, purple olive highly raised, shallow, sharply t" rigid, open green raised, dark lined, crenate pointed, tzj prominent prominent pubescent. ~ Z

T. triTuata______Glabrous Medium Red to Dull, Raised, Obscure Widely Long, very ~ shallow, sharply t¢ to stout, purple olive prominent ."" upright, green crenate pointed, pubescent. c::: rigid, open ~ t) tzj "C ~ o "l:j

~ Q ~ .....o c::: t" >-3 c::: ~ t=.1 TABLE 24.-Vegetative characters 1.t8ed for initial identificatiOn of intergeneric hyb1ids I

Twigs and young shoots Leaves Tenninal rn Code of hybrids Pubes- Habit Color Color Lower surface vegetative t-3 cence of of new of upper Margins buds C t:::I growth growth surfaC'e Midrib Veins ..... t1j rn C. pitardii var. pitardii X T. spectabilis: ~ A-19______..... m - ...------"""'- m s f m .~ T. virgata X C. granlhamiana: 7-233______(') m m m f f m j ib 7-235______~ m ------.... m f f m m .. 7-255______I:tJ ~ m f m f f m m t"< )/l t"< ..... Q T. virgata X C. miyagii :t. ~ 5-62______~ m m m s f :> z 5-70______Z " 5 76______m m m m s f t'1 § m m m m s r 1 -; 5-77______------ ~ ~ m m r m m s f m l?i1 5-80______t.-' f f m m s ~ 5-81 ______f :> m i f m m s f i t-3 ~ 5-82______l?i1 :; m m f m m s f m tj

~ !;) l'l l?i1 1 m resemblance to male parent; f resemblance to female parent; i intermediate characters; s characters similar to both Z ~ parents. l?i1 o ~ :> ... c'" ;;; ...... en