1-l\-'<-v 11 u _, q .52 . 4.17:: gig; 13:";- n11.1.r!1/.-».1}1 .1:".:.1'1-'»JJ"..- -~ 1 1 1 t 11.1 r a 1n ' .pfa:1up~1n.11—" $1r¥3,“"r. . ,1, w “-‘" ‘—' Not-q .11."; ,2”; ”a - as,» H...» r’ ‘5. . .. .1.-. , ,1. .ndr a . .: m- '7’»; 1.1"". . Wonk”. 1 r v 2:372" fi‘rrtr-Je: 1 ~ * 1.. "1 11, , rifl" "‘q' I'qm‘ m‘W - 'M -- 11. 1: NW. Hera-In! IE1“ ’ he” - :1" 115:? 4-1.”:2!‘ Harri?“ ’1 gt .531“, I ,.‘1/.v ‘Iit‘ -,.. r 111'!" Lin-11“.” v "'1'11.' 1 1“ h; ' mg: “3?.” ,. "‘u‘fiy" 1 "‘ ~' ‘a . ~z~ .. 1" mini. «1 . ‘r' “‘N’v‘ht ~fih’4un.1:|~n..1,ll LlfiRARY Michigan State University This is to certify that the thesis entitled INTERSPECIFIC HYBRIDIZATION IN ZINNIA: MORPHOLOGY, CYTOLOGY, POLLEN EXAMINATION, AND PONDERY MILDEN RESISTANCE presented by Sheila Dale Linderman has been accepted towards fulfillment of the requirements for Master's degree in Plant Breeding and Genetics/Horticulture <’ v Major professor Date //[/[/’/§7 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution i MSU LIBRARIES _—;_. RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. INTERSPECIFIC HYBRIDIZATION IN ZINNIA: MORPHOLOGY, CYTOLOGY, POLLEN EXAMINATION, AND POWDERY MILDEW RESISTANCE By Sheila Dale Linderman A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Plant Breeding and Genetics/Horticulture 1988 ABSTRACT INTERSPECIFIC HYBRIDIZATION IN ZINNIA: MORPHOLOGY, CYTOLOGY, POLLEN EXAMINATION, AND POWDERY MILDEW RESISTANCE By Sheila Dale Linderman Zinnias are an important garden flower but unprofitable as a bedding plant crop due to disease and cultural prob- lems. Zinnia gngugtifolig is resistant to powdery mildew and has culturally desirable traits. Interspecific hybrid- ization was proposed between L, angugtifolia and L, elegans, L, haageana to attempt transference of desirable traits into the latter two horticulturally important species. A new interspecific hybrid, L, angustifolia x L, haageana, and L, angugtifolig x L, elegans and its recipro- cal, were recovered. Examination of morphology, cytology of pollen mother cells, and pollen staining confirmed hybridity of the plants. Screening for powdery mildew proved L, angustifolia x L, elegans and L, angustifolia x L, haaggana resistant and L, elegans x L, angustifolia moderately sus- ceptible to the disease. This germ plasm could serve as a source of culturally responsive traits and disease resistance to be incorporated into commercial cultivars through further breeding. In remembrance of my father, Harvey Lloyd Rathbun, Jr. iii ACKNOWLEDGEMENTS I would like to thank my major professor, Dr. Lowell Ewart, for his guidance on my research. It has been a pleasure working with and knowing you and your wife. I would also like to express my appreciation to Dr. Chris Stephens and Dr. Ev Everson, my committee members, for their time and advice, and to Dr. Viv Shull for his assistance with the microphotography. The time involved in attaining this degree has been a memorable experience and I have enjoyed sharing it with my fellow graduate students and friends -- you know who you are! I would also like to mention some other people whose support and encouragement have been important to me: my mother, Florence Rathbun, my parents-in-law, Earl and Mildred Linderman, and especially my husband, Glen. Glen, you have contributed as much as I have toward this degree. Thank you for your love, your understanding and your confi- dence in me. iv TABLE OF CONTENTS 2592 LIST or TABLES.... ................................ . ..... vi LIST OF FIGURES.... ........ . ............................ vii INTRODUCTION.... ....................................... . 1 LITERATURE REVIEw.. ..................................... 3 MATERIALS AND METHODS ............................ . ...... 13 RESULTS. ........................ . ................... .... 19 DISCUSSION......... ...... . ...... ........ ..... ........... 42 LIST OF REFERENCES ............................... . ...... 46 liable 1. Pollinations using L, angustifolia as the female parentOIOIIIOOOCCOOOOI.DIOOOIOOOCOOIOIOOIOIIII ..... 2. Pollinations using L, angustifolia as the male parentIOIOOOOOIICIOI00.0.0.0...DID... ...... 0.0.0... 3. Pollen examination using cotton blue stain......... LIST OF TABLES vi LIST OF FIGURES Figure Page 1. Comparison of leaf width of three Zinnia species, from left to right: L, elegans broad leaf type, L, elegans narrow leaf type, L, haageana (2n), L, augustifOIiaIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 21 2. L, an ustifolia x L, elegans; hybrid at center.... 22 3. L, elegans x L, angustifolia; hybrid at center.... 22 4. L, anggstifolia x L, haageana (2n); hybrid at centerIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 000000 23 5. Variation in capitula diameter and ray floret color among hybrids of L, angustifolia x L, elegansIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII ..... 23 6. Pollen of L, elegans: a smaller, clear pollen grain surrounded by large, blue-stained grains (500x)IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 26 7. Pollen of L, an ustifolia: three clear and four dark blue pollen grains (SOOX).................... 26 8. Pollen of L, haageana (2n): one small, clear grain and six large, blue-stained grains, one show- ing its outer surface (500K)...................... 27 9. Pollen of L, haageana (4n): two clear pollen grains surrounded by large, dark blue grains (500x)II...IIIIIIIIIIIIIIIIIIIIIII ..... IIIIII ..... 27 10. Pollen of the hybrid, L, elegans x L, angustifolia: two grains in cross section and four showing their outer surfaces; all grains are clear colored (500K).............................. 28 ll. Pollen of the hybrid, L, angustifolia x L, elegans: one large, dark blue grain among small, Clear grains (500x) I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 28 12. Pollen of the hybrid, L, aggustifolia x L, haageana (2n): one large, dark blue grain and two small, clear colored pollen grains (500K)..... 29 vii Figure Page 13. Germinating pollen of L, angustifolia (4OOX) ...... 29 14. Dividing cell of L, elegans (SOOX). ............... 31 15. Chromosomes of L, ele ans ..... . ...... .. ........... 31 16. Dividing cell of L, haageana (2n) (1250X) ......... 32 17. Chromosomes of L, haageana (2n)........ ........... 32 18. Dividing cell of L, haageana (4n) (lZSOX) ..... .... 33 19. Chromosomes of L, haa eana (4n). .................. 33 20. Dividing cell of L, angustifolia (1250X) .......... 34 21. Chromosomes of L, angustifolia............ ........ 34 22. Dividing cell of the hybrid, L, angustifolia x L, elegans (500x)IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 35 23. Chromosomes of the hybrid, L, angustifolia x L, elegansIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIII I 35 24. Dividing cell of the hybrid, L, elegans x L, angustifolia (1250X)........................ ...... 36 25. Chromosomes of the hybrid, L, elegans x L, anguStifOIiaIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIII I 36 26. Dividing cell of the hybrid, L, angustifolia x L, haagel‘na (2n) (1250x)IIIIIIIIIIIIIIIIIIIIIIIIIIIII 37 27. Chromosomes of the hybrid, L, angustifolig x L, haagean‘ (2n)IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 37 28. Powdery mildew on leaf tissue of L, elegans, L, haageana (2n) and L, haageana (4n)................ 39 29. Powdery mildew on petal tissue of L, eleggn ...... 39 30. Powdery mildew on petal tissue of L, haageana (‘n)IIIIIIIIIIIIIIIIIIIIIIIII IIIIII IIIIIIIIIIIIIII 40 31. Powdery mildew on petal tissue of the hybrid, L, elegans x L, angustifolia................... ..... . 40 viii INTRODUCTION There have been 36 zinnia cultivars chosen as All- America Selections award winners from 1933 to 1987; only two other floriculture crops, marigolds and petunias, have surpassed zinnias in number of award winners (3). A 1976 National Garden Bureau press release listed zinnia as the number one consumer preference among seed packet items in 1975. Even though zinnias are a major seed packet item, wholesale bedding plant growers have been unable to profit from their popularity with home gardeners due to cultural and disease problems. Growers rated zinnias as second poorest in a comparison of various floricultural crops, and they comprised only 1 to 2% of the production sales mix in 1986 (31). Cultural problems encountered by growers are etiolation, yellowing of leaves, weak growth and inability to flower them in flats (13). A narrow foliage trait in Zinnia elegans allows these plants to respond to growth regulators and photoperiod manipulation (14). Three major diseases occur in the greenhouse and the consumer’s garden, powdery mildew caused by Erysiphe cichoracearum DC ex Merat, alternaria blight caused by Alternaria zinniae Pape, and bacterial leaf spot caused by l Xanthomonas campegtris pv zinniae Hopkins and Dawson (2,6,16,27,28,29). L, angustifolia is a species that exhibits disease resistance, narrow foliage and compact free-flowering growth habit (2,6,16,27,28,29). Hybridization of L, angustifolia with L, elegans and L, haageana should allow incorporation of desirable traits into the latter two horticulturally important species, and would broaden the narrow germ plasm base of commercial cultivars. Improved zinnia cultivars would increase consumer satisfaction with their garden performance. They could become a major part of the crop mix of bedding plant growers, an important part of the floriculture industry, whose wholesale value increased substantially from $148,996,000 in 1979 to $546,693,000 in 1986 (1,11). Disease resistant, culturally responsive zinnia cultivars have potential for development as a cut flower crop with their wide range of flower forms and colors. LITERATURE REVIEW Zinnias, as described by Torres, belong to the family Compositae or Asteraceae, tribe Helianthaea, subtribe Zinninae, genus Zinnia and subgenus Zinnia (29). The sub- genus is divided into two sections, Mendezia and Zinnia. Four horticulturally important species are found in these two sections. Section Mendezia is comprised of species whose basic chromosome numbers are x=ll, x=12, and the horticulturally important species is L, angustifolia HBK (syn. L, linearis Benth). It is an erect herb growing to 38 cm with linear to lance-linear leaves. Flower heads are about 9.8 mm in diam- eter with 7 to 9 oblong, bright orange ray florets. It is native to Mexico, found in Sonora and Chihuahua to Michoacan at elevations of 60 to 2100 m and blooms from July to January. The literature regarding chromosome numbers is conflicting; some studies report n812 (5,29,30), some report n=ll (18,21,22,26), and some report both n=1l and n812 (19). Anderson found this species resistant to powdery mildew in that minimal growth of hyphae occurred on the leaves after 24 hours following germination. This was presumed due to the conidia being unable to establish a stable parasitic relationship with the host (2). L, angustifolia has also been found resistant to bacterial leaf spot and ranged from 4 hypersensitive to resistant to alternaria blight (16,27). Section Zinnia contains three other species of horti- cultural significance, L, peruviana L., L, haageana Regel, and L, elegans Jacq., all with a chromosome number of n=12. L, peruviana is an erect herb growing to 90 cm. The leaves are variable, lance-linear to broadly ovate or ellip- tic. Flower heads are 18 mm with 6 to 15 rays, with ray florets variable in shape from linear to spatulate or sub- orbicular and scarlet red or yellow in color. This species can be found growing from southeast Arizona, Central America, Mexico and the West Indies in North America and Colombia and Ecuador to Peru and Argentina in South America growing at elevations up to 3,000 m. Flowering varies from April to October in North America to December to May in South America. Metcalf and Sharma report this species is susceptible to powdery mildew (17). L, haageana is an erect herb growing to 60 cm with lanceolate leaves. Flower heads measure to 1.1 cm in diam- eter with 8 to 9 oblong rays that are bright orange and velvety above, drying to yellow, and dull orange to yellow and hirsute below. Under cultivation, it may produce more than one row of ray florets, and the florets may be varie- gated red and yellow or cream. It is native to Mexico, found from Jalisco to Guerrero at elevations of 900 to 2,010 m and blooms from July to November. It is susceptible to 5 powdery mildew and alternaria blight and is slightly suscep- tible to bacterial leaf spot (16). L, ele ans is the species of greatest horticultural importance. It is an erect herb growing to 90 to 100 cm with leaves lanceolate, ovate or oblong. Flower heads meas- ure to 2.2 cm in diameter with 8 to 20 spatulate to obovate ray florets in a range of colors: white, yellow, orange, pink, scarlet, lilac, purple and red, the latter color usu- ally occurring in nature. The range in Mexico is from Sinaloa and Durango to Guerrero at elevations to 1,560 m and blooms from March to November. It is susceptible to powdery mildew, alternaria blight and bacterial leaf spot (2,6,16,27,28,29). Metcalf and Sharma summarized the commercial breeding history of zinniae (17). Of the four species, L, peruviana was first introduced into cultivation in England in 1753. It has decreased in popularity and is of little significance today. L, elegans was introduced into Europe about 1796, followed by L, haageana in 1862. L, angustifolia was intro- duced by the late 18th or early 19th centuries. "Classic" is the only major cultivar of L, angustifolia available. L, haageana was used in the development of the cultivars ”Persian Carpet" and "Old Mexico", the latter a tetraploid derivative of the former. L, hgageana and L, eleggns were used in interspecific crosses which resulted in the release of a few multicolored cultivars, such as the 6 ”Navajo” series, ”Pinwheel”, ”Peppermint Stick” and ”Whirligig”. Most commercial breeding work has been with L, elegans. The overall breeding strategy has been to obtain new flower forms in color mixes followed by separation and extension of the color range, with the simultaneous production of larger flowers on smaller plants. The first major flower form change was attained when double flowers on rather tall plants were developed. Typified by the "Pumila" cultivars, these types were available by the late 18th century. The first large, double-flowered zinnia, ”Giant Mammoth", was released in 1886. Different flower types were also avail- able late in the 1800's. The cactus-flowered type, "Curled and Crested”, was released prior to 1900, and the improved "Fantasy" cactus-flowered cultivars were released in the early 1930's. The ”Dahlia-Flowered” cultivar was introduced in 1919 and ”Giants of California" in 1926 were the first dahlia-flowered zinniae. The first colchicine-induced auto- tetraploids, ”Super Giants” and “Luther Burbank”, were re- leased in 1942 and 1948, respectively. ”State Fair“ is a modern, dahlia-flowered, colchicine-induced autotetraploid. The first true dwarf cultivar, "Thumbelina", was released in 1963. The latest advance in commercial breeding has been production of F hybrids. The first true F hybrid, pro- 1 l duced by crossing two inbred parent lines, was "Trail 7 Blazer”, introduced in 1960. ”Zenith" followed in 1964, the first F hybrid produced using male sterility. Zinnia cultivars are generally classed by height and flower type (17). By height, extra dwarf is less than 20 cm, dwarf is between 20 and 37.5 cm, half-tall is between 45 and 60 cm, tall is between 60 and 75 cm and giant is greater than 75 cm. By flower type, single flowers have one to two rows of ray florets, crested flowers have disk florets that are long and truncated, pompon types are small to medium size double flowers, dahlia-flowered types are large pompons with flat ray florets and cactus-flowered types have ray florets that are rolled or twisted. Scientists and breeders have attempted to improve the quality of zinnias. Gupta and Koak induced autotetraploidy in L, elegans through use of colchicine and evaluated the effect of this mutagen on the tetraploids (15). They found decreased pollen fertility, enlargement of some morphologi- cal organs and some meiotic irregularities. Bose and Panigrahi found similar results when they induced tetraploidy in L, angustifolia (5). Raman, Rangasamy and Ramalingam also induced polyploidy in L, gngustifolia; they crossed the tetraploid with the diploid to obtain a triploid and evaluated all three ploidy types (22). They found some meiotic irregularities in the tetra- ploid with chromosome association ranging from 5 quadri- valents to 22 bivalents. Meiosis in the triploid averaged 5 trivalents; the remaining chromosomes usually formed 8 bivalents and univalents. From this, Raman, et a1. sug- gested that the haploid genome of the diploid was consti- tuted of two basic sets of 5 and 6 chromosomes. The 5 chromosomes have identical homologs in the diploid, while the set of 6 has nearly identical homologs. They proposed that the modern species is the result of hybridization and amphiploidy of two basic species with n=5 and n86. Researchers at Michigan State University have investi- gated the inheritance of male sterility and narrow leaf shape in L, elegans (13). Male sterility is expressed in the presence of three recessive genes as an apetalous, male sterile inflorescence. Male sterility is an important tool in breeding hybrid zinnias —- it eliminates the cost of emasculating‘ the female parents. Additive gene action in- fluenced the leaf width of L, elegans. The narrow leaf type first segregated from the "Wild Cherry” cultivar of the "Fruit Bowl” series. It could be the result of a cross with either L, haageana or L, an ustifolia, or the result of a mutation in L, elegans. Narrow foliage allows light to penetrate the plant to the leaf axils, enabling the plant to respond to growth regulators and photoperiod manipulation (14). By controlling these two factors, growers can consis- tently bloom zinnias with narrow foliage in flats. Recent research has explored the possibility of growing zinnia as a cut flower crop. Boyle and Stimart found L, elegans to be a facultative short day plant in that floral initiation and development is hastened under short days of 9 12 hours or less (7). They found five short days to be the minimum required for irreversible floral initiation in all the cultivars studied. Twenty to 25 long days of at least 14 hours after transplanting of seedlings produced cutting- quality plants with rigid stems, large flower diameters and a maximum number of ray florets. Boyle, Stimart and McIntosh investigated the effect of seasonal variation on zinnia production (8). They deter- mined that zinnia cultivars can be selected for earliness to flower based on a minimum number of nodes below the inflor- escence. Seasonal differences in irradiance, photoperiod and temperature affect the vegetative and reproductive de- velopment of L, elegans. From spring through fall, flower- ing is mainly influenced by temperature, while during winter flowering is delayed through an increase in the time between initiation of successive leaf primordia and/or a slower rate of floral development. They proposed photoperiod control during the short days of fall through spring to be essential in the production of quality flowers. Production time var- ied from 6 to 8 1/2 weeks during early summer, fall and spring to 10 1/2 to 14 1/2 weeks during midwinter. Stimart and Brown found that a preservative solution containing 200 mg/l 8-hydroxyquinoline citrate and 1% su- crose extended the decorative postharvest life of L, elegans to 11 to 14 1/2 days; this solution reduced vascular blockage and intensified and maintained flower color (25). 10 The literature contains several reports of interspe— cific hybridization within the genus Zinnia. Olorode per- formed highly successful crosses between four species in the section Mendezia and concluded that their genomes are highly homologous (18). Shahin, Campbell, Pollard and Hamson ob- tained hybrids from the cross L, elegans x L, peruviana through the tissue culture technique of embryo culture (23). Their work could prove a valuable guide for future interspe- cific hybridization where embryo abortion is a problem. Interspecific hybrids have been occasionally obtained from L, angustifolia and L, elegans. This cross could be invaluable for transferring disease resistance from L, angustifolia to L, elegans. Ramalingam, Rangasamy and Raman obtained hybrids using L, angustifolia as the female parent (21). Hybrids had some intermediate morphological charac- teristics and very low pollen fertility. They had a chro- mosome number of 2n-23 and an average meiotic chromosome association of 6 bivalents and 11 univalents. Ramalingam, et al. proposed two basic sets of 5 and 6 chromosomes for the haploid genome of L, an stifolia, as did Raman, et a1. They further postulated that one set of 6 chromosomes in L, anggstifolia is partially homologous to a similar set of 6 chromosomes in the haploid genome of L, ele ans, and that the other set of 6 L, eleggns chromosomes were distinct from the former set. Thus, they theorized that both L, anggstifolia and L, elegans resulted from hybridization and ll amphiploidy of two species, one n=6 and one n=5 in L, angustifolia and two n=6 in L, elegans. Boyle and Stimart also obtained interspecific hybrids from the cross L, angustifolia x L, ele ans, using commer- cially available cultivars for the latter species (6). They recovered one plant from the reciprocal cross by excising the embryo from the seed before it aborted. Hybrids resem- bled L, gngugtifolia and had very low fertility. Colchicine treatment doubled the chromosomes, increased the pollen fertility and increased the'size of some morphological or- gans. Preliminary screening of all plants for powdery mil- dew resistance classified all hybrids as resistant. Cytological studies of the amphiploids obtained in the experiment by Boyle and Stimart were performed by Terry- Lewandowski, Bauchan and Stimart (26). The F hybrids usu- ally had 23 univalents during meiosis, but 1 to 6 bivalents were observed. The amphiploids usually had 23 bivalents during meiosis, although a low percentage of univalents was observed. The absence of multivalents during meiosis of the amphiploids suggested a genetic control of pairing, where the double dose of the gene present in the amphiploid elim— inated the homoeological pairing seen to a small extent in the F hybrid. Terry-Lewandowski and Stimart evaluated the amphi- ploids' resistance to the three major zinnia diseases (27). The amphiploid derived from the cross L, anggstifolia x L, elegans was resistant to powdery mildew, hypersensitive to 12 alternaria blight and intermediately resistant to bacterial leaf spot. The amphiploid from the reciprocal cross, L, elegans x L, angustifolia, was intermediately resistant to powdery mildew in that senescing petal tissue developed nonsporulating colonies of L, cichoracearum. It was hyper- sensitive to alternaria blight and intermediately resistant to bacterial leaf spot, although in the latter disease, it was less resistant than its reciprocal amphiploid. Terry-Lewandowski and Stimart investigated the inheritance of resistance to powdery mildew in the F hybrids and col- chicine-derived amphiploids through reciprocal crosses be- tween two amphiploids originating from the crosses: L, elegans x L, angustifolia and L, anggstifolia x L, ele ans (28). Ray petal resistance of the F progeny of these re- ciprocal crosses was intermediate ind not significantly different from each other, although it did not resemble either amphiploid parent. The F progeny of each cross displayed a 15 resistant:l susceptible ratio suggesting two major dominant genes responsible for disease resistance in the ray florets of the amphiploids. The phenotypic similar- ity of the F and variation in susceptibility observed in the F suggeit nuclear and not cytoplasmic inheritance of these genes. Terry-Lewandowski and Stimart suggest that possibly all of the genes responsible for resistance to powdery mildew are inherited from L, angustifolia. MATERIALS AND METHODS Growing of Plants Seed of the following species was used: various inbred lines of L, elegans developed at Michigan State University, with a red or orange ray floret color and exhibiting the characteristics of broad or narrow foliage and/or apetalous male sterility; L, angustifolia, cultivar “Classic”; L, haageana, cultivars "Persian Carpet”, a diploid type, and ”Old Mexico", a tetraploid. Plants were grown in a green— house with temperatures ranging from 27 C daytime and 16 C nighttime during the warm months of the year (May to September) and 18 C daytime and 16 C nighttime during October to April. Plants were watered with warm water (27 C) from October to April. Normal seed germination practices were followed. Seed- lings were transplanted to 32-cell flats containing commer- cial peat-lite medium when the cotyledons were fully expanded. Plants were grown in these flats until they flow- ered: selected plants were potted in 15 cm pots or 25 cm pans. Pollinations The following interspecific crosses were performed: L, elegans x L, an ustifolia, L, haa eana (2n) x L, angustifolia, L, haageana (4n) x L, an ustifolia, and the 13 14 reciprocal crosses of each. Capitula of the female parent chosen for pollination had disk florets removed with fine- tip forceps. Apetalous male sterile L, elegans plants re- quired no emasculation. Pollination was effected by brushing the stigmas of the female plant with a detached capitulum of either L, gnqugtifolg; or L, haageana or with detached disk florets of L, elegans. Pollinated flower heads were tagged with the date of pollination and identity of the pollen parent. Seed was harvested four weeks after pollination and allowed to dry at least two weeks before sowing. All seed from interspecific crosses was sown and all plants grown to maturity. Plants exhibiting some character- istics intermediate between the parent species were selected and potted for further study. Pollen Examination Pollen of a plant selected from each species and of a number of possible hybrids was examined under a binocular microscope at 200x magnification using cotton blue stain (12). At least five fields of each sample were examined and the pollen grains counted. They fell into two categories: large, plump, dark blue grains and smaller, shriveled grains that stained a light blue to clear color. The percentage of deep blue to total pollen grains was determined. An attempt was made to germinate pollen of the species and hybrids to compare with the results obtained from stain- ing with cotton blue. Since pollen production in zinnias 15 peaks in the middle of summer when temperature and light intensity are at their highest, germination techniques were carried out during the six month period of April through September. Sucrose content of the germination media was varied from 10 to 12.5 to 15% to determine the optimum amount of sucrose, 10%, which was used in subsequent media. Solid media consisted of 200 ppm boric acid, 400 ppm calcium nitrate, 10% sucrose and 4% agar. Squares of solid media, 2 x 2 x 0.4 cm were placed on microscopic slides, pollen was shaken onto the media, and the slide stored in a sealed petrie dish with a distilled water-saturated filter paper for 24 hours at 30 C in the dark. A drop of aceto- carmine stain was placed on the media and a coverslip ap- plied before examining under a light microscope for germinating pollen. Liquid media was prepared of 10% sucrose, 100 ppm boric acid, 300 ppm calcium nitrate, 200 ppm magnesium sulfate and 100 ppm potassium nitrate (9). Pollen was shaken over a microscopic slide and media was pipetted onto the slide in standing drops of about 0.02 ml, allowing a small population of about 20-30 pollen grains per drop. The slide was sealed in a petrie dish with a water-saturated filter paper and stored in the dark at 30 C for 24 hours. Slides were exam- ined under a light microscope for germinating pollen grains. Five fields were counted under the light microscope to determine germination percent. 16 Cytological Techniques Pollen mother cells were examined by the smear tech- nique to verify chromosome numbers of the three species and to verify hybridity of selected plants resulting from inter- specific pollinations. Capitula were dissected with the aid of binocular mag- nifying glasses and fine-tip forceps to obtain immature disk florets approximately one to two mm in length. The florets were placed in modified Carnoy's solution for 24 hours to kill and fix the cells (4). After fixing, florets were rinsed in distilled water and four to five florets were placed on a microscopic slide in a drop of aceto-carmine stain (24). The florets were macerated with a spear-pointed needle and excess tissue removed with fine-tip forceps with the aid of an Olympus binocular dissecting microscope. A cover slip was applied and the slide placed between folds of a paper towel before pressure was exerted upon it. The slide was gently heated, then examined under a binocular microscope at 400x magnifi- cation for dividing cells. Slides which contained dividing cells were preserved by sealing the cover slips with molten gum mastic and storing them in a glass bottle in the hydrator of a refrigerator. Slides prepared in this manner remained in good condition for several weeks. Several examples of dividing cells of each plant type were examined. l7 Photomicrographs of pollen and dividing cells were made at magnifications of 250x, 500x and 1250K. Selected prints were enlarged and images of the chromosomes were traced onto wet-media acetate with a permanent marker. If necessary, the positions of the chromosomes were verified by re-examining the dividing cell under the microscope and focusing through the depth of the cell. Prints were made of the chromosome images on the acetate for clarification (20). Powdery Mildew Inoculation Ten plants each of the following were placed in the greenhouse for inoculation: L, ele ans, L, angustifolia, L, haageana 2n and 4n, and the hybrids, L, elegans x L, angustifolia, L, Lngugtifolig x L, ele ans, and L, angustifolia x L, haageana (2n). As a check, two plants of each were planted in the field for summer growth followed by natural inoculation in the fall. By September 10th, powdery mildew was beginning to develop naturally in the greenhouse, as small colonies of mycelium about 0.5 cm in diameter were present on L, haageana 2n and 4n. Eight heavily infested plants of L, elegans from the field served as a source of inoculum. They were placed on the greenhouse benches and were randomly surrounded by eight to nine test plants. The infested plants were shaken every two to three days for 14 days to allow conidia to fall onto the test plants. 18 Four weeks later, the test plants were examined and evaluated as follows: resistant = no colonies of mycelium visible, moderately susceptible = few colonies present with some mycelium covering 50% of plant surface, susceptible = heavily infected with mycelium covering 100% of plant sur— face. Leaf and senescing petal tissue were evaluated sep- arately. Plants grown in the field were evaluated on October lst using the same criteria. RESULTS Crossing Results Hybrids were obtained from the following crosses: L, angustifolia x L, elegans and its reciprocal, L, elegans x L, gngugtifolia, and L, angustifolia x L, haageana (2n). Of these three crosses, the first resulted in 172 plants classified as hybrids. The other two crosses were much less successful: the reciprocal cross produced one hybrid and the last cross mentioned above produced two hybrids. Tables 1 and 2 summarize the relative success of each cross. Hybrid Morphology Morphological characteristics of the hybrids were vari— able: some were similar to one of the parents, some were intermediate to the parents. Leaf width and length was intermediate to the parents, ranging from 2 cm in length by 0.7 cm in width up to 5 cm by 1.7 cm. Figure 1 shows leaves of the three Zinnia species. Considerable variation in flowering was observed. Some plants formed buds but anthe- sis never occurred. Other plants produced distorted flowers consisting of a few ray florets and/or disk florets. Some showed seasonal variation in flowering, blooming from ap- proximately May through September, during times of high light intensity. Fifty-four of the hybrid plants obtained from the cross L, angustifolia x L, elegans produced 19 20 Table l. Pollinations using L, angustifolia as the female parent. Male Parents Capitula Pollinated Hybrids % Success L, elegans 409 172 42.05 L, haageana (2n) 229 2 0.87 L, haageana (4n) 199 0 0 Table 2. Pollinations using L, angustifolia as the male parent. Female Parents Capitula Pollinated Hybrids % Success L, elegans 426 l 0.23 L, haageana (Zn) 175 0 0 L, haageana (4n) 189 0 0 21 edesens I: Inafifl‘dna .Z. aq,fl*345ha . J " f‘ m1... 1“ I r . ‘ -‘ h 1 _ Figure 1. Comparison of leaf width of three Zinnia spe- cies, from left to right: L, elegans broad leaf type, L, elegans narrow leaf type, L, haageana (2n), L, angustifolia. 22 Figure 2. L, angustifolia (left) x L, ele ns (right): hybrid at center. Z. elegans X Z. angustifolia Figure 3. L, elegagg (left) x L, anggstifolia (right): hybrid at center. 23 I t _A_Wz.,qggg§gigplia x z. haageana i Figure 4. L, angustifolia (left) x L, haageana (2n) (right): hybrid at center. Z. angustifolia X Z. elegans Figure 5. Variation in capitula diameter and ray floret color among hybrids of L, angustifolia x L, elegans. Table 3. elegans angustifolia haageana (2n) haageana (4n) elegans x anggstifolia anggstifolia x elegans anggstifolia x haageana (2n) 24 285 236 189 348 610 2174 740 276 211 183 134 ll Pollen examination using cotton blue stain. Total Pollen Blue Pollen % Blue Pollen 96 89 96. 38 .84 .41 83 .51 .49 .51 .68 25 undistorted flowers -- the hybrids had a short, mounded growth habit and freely flowered yearround (Figure 2). The one plant obtained from the reciprocal cross produced flowers (Figure 3). Of the two plants obtained from the cross L, gngustifolig x L, haageana (2n), one of these was a flowering plant (Figure 4). Flower diameter was intermediate to the parents but variable among the hybrids, ranging from 3 to 6 cm. Color of the ray florets was variable; some were bright orange, similar to L, gngugtifolia, while others were a red-orange or red-orange mottled with lighter orange (Figure 5). This range of colors was observed among the population of hy- brids, as well as within individual plants. Plants tended to produce larger, darker flowers during the summer months. All flowers resembled L, angustifolia in that they had a single row of ray florets surrounding the disk florets. Pollen Examination Table 3 summarizes the results obtained from examining pollen stained with cotton blue. The species plants showed a high proportion of stained pollen, ranging from 89.41% to 96.84%, while the tetraploid version of L, haageana showed a considerable decrease in stained pollen, 38.51% as compared to the diploid’s 96.83%. The interspecific hybrids all show a very low but similar percentage of stained pollen grains, ranging from 0.49% to 0.68%. Figures 6 through 12 are pic- tures of pollen grains from the species and hybrid plants. Figure 6. Pollen of L, ele ans: a smaller, clear pollen grain surrounded by large, blue—stained grains (500K). '0 Figure 7. Pollen of L, angustifolia: three clear and four dark blue pollen grains (500K). 27 K) “‘ >| I 5 'i \ e \ . I Figure 8. Pollen of L, haageana (2n): one small, clear grain and six large, blue-stained grains, one showing its outer surface (500K). 9e... 4 Figure 9. Pollen of L, haa eana (4n): two clear pollen grains surrounded by large, dark blue grains (500X). 28 Figure 10. Pollen of the hybrid, L, ele ans x L, angustifolia: two grains in cross section and four showing their outer surfaces; all grains are clear colored (500K). .1 )3 Figure 11. Pollen of the hybrid, L, angustifolia x L, elegans: one large, dark blue grain among small, clear grains (500K). Figure 12. Pollen of the hybrid, L, an stifolia x L, hgagegna (2n): one large, dark blue grain and two small, clear colored grains (500K). Figure 13. Germinating pollen of L, angustifolia (400K). 30 Germination of pollen was better in liquid media, al- though overall germination was low regardless of time of year. Only two species, L, anggstifolia and L, elegans, showed pollen tube growth (Figure 13). Germination in liquid media ranged from 23.07% to 44.12% for L, angustifolia, with an average of 34.13%. Pollen tube growth in L, elegans in liquid media ranged from 13.79% to 17.14% with an average of 15.69%. Neither L, haageana 2n or 4n nor any of the hybrids germinated under any of the conditions tried. Cytological Techniques The greatest success in obtaining dividing cells oc- curred when samples were taken in the late afternoon during the months of April through September. Examination of div- iding cells of the species plants confirmed the chromosome numbers reported in previously published literature. L, elegans has 12 chromosomes, as shown in Figure 14 and clar- ified in Figure 15. L, haageana (2n) also has 12 chromo— somes (Figures 16,17), while L, haageana (4n) has 24 chromo- somes (Figures 18,19). L, angustifolia has 11 chromosomes (Figures 20,21). Hybrid plants had a total of 23 chromosomes per cell (Figures 22,23,24,25,26,27). Dividing cells from all plants showed some disturbance in homologous pairing of chromo- somes. Other meiotic abnormalities were noted: univalents left on the metaphase plate after the chromosomes had moved to the opposite sides of the cell were seen in a cell of the 31 Figure 14. Dividing cell of L, elegans (500K). Figure 15. Chromosomes of L, elegans. 32 3. Figure 16. Dividing cell of L, haageana (2n) (1250K). «Is as. (gas? Figure 17. Chromosomes of L, haageang (2n). Figure 18. Figure 19. Dividing cell of L, haageana (4n) (1250X). Chromosomes of L, haageana (4n). 34 Figure 20. Dividing cell of L, anggstifolia (1250X). Figure 21. Chromosomes of L, angustifolia. 35 Figure 22. Dividing cell of the hybrid, L, angustifolia x L, elegans (500X). Figure 23. Chromosomes of the hybrid, L, angustifolia x L, elegans. 36 Figure 24. Dividing cell of the hybrid, L, elegans x L, angustifolia (1250X). Figure 25. Chromosomes of the hybrid, L, elegans x L, angustifolia. Figure 26. Dividing cell of the hybrid, L, anggstifolia x L, haageana (2n) (1250X). ::.:--- l e . .9 e 0 .: . Cs . Figure 27. Chromosomes of the hybrid, L, angustifolia x L, haageana (2n). 38 hybrid L, angustifolia x L, elegans; micronuclei and uni- valents were present in some cells of the hybrid L, angustifolia x L, haageana (2n). Where possible, the number of paired chromosomes per cell were counted for the hybrids. Four cells of the hybrid L, angustifolia x L, elegans showed a range of 3 to 5 bi- valents with an average of 4.25 bivalents. Four cells of the reciprocal hybrid, L, elegans x L, gngugtifolig, showed a range of 1 to 2 bivalents with an average of 1.25 bi- valents. Six cells of the hybrid L, angustifolia x L, haageana (2n) showed a range of 2 to 5 bivalents with an average of 3.67 bivalents. Powdery Mildew Inoculation Within each species and hybrid class, all ten plants reacted in a similar manner to the pathogen. Evaluation of leaf tissue of the test plants showed the following to be susceptible to powdery mildew: L, ele ans, L, haageana 2n and 4n (Figure 28). L, angustifolia and the hybrids, L, elegans x L, angustifolia, L, angustifolia x L, elegans, L, gngugtifolig x L, haageana (2n) were resistant. Evaluation of ray floret tissue gave some different results. Powdery mildew does not become evident on petal tissue until it is senescing. L, ele ns, L, haageana (2n and 4n) were suscep- tible (Figures 29,30), and L, angustifolia was resistant. Of the hybrids, L, angustifolia x L, elegans and L, angustifolia x L, haageana were resistant, and L, elegans x L, angustifolia was moderately susceptible (Figure 31). M & Figure 28. Powdery mildew on leaf tissue of L, elegans (left), L, haageana (2n) (center) and L, haageana (4n) (right). Figure 29. Powdery mildew on petal tissue of L, elegans. 40 Figure 30. Powdery mildew on petal tissue of L, haageana (4n). Figure 31. Powdery mildew on petal tissue of the hybrid, L, elegans x L, angustifolia. 41 Evaluation of the plants in the field where infection was allowed to occur naturally confirmed the above findings. L, anggstifolia and the hybrids, L, angustifolia x L, elegans and L, angustifolia x L, haageana (2n), were com- pletely resistant to powdery mildew. L, glggégg, L, haageana (2n and 4n) were completely susceptible to the fungus. The hybrid, L, elegans x L, angustifolia, had re- sistant leaves and moderately susceptible flowers. The disease always appeared first on plants of L, haageana (2n and 4n). Although initial development of the disease on plants of L, elegans occurred later, with time the extent of infection equalled that of L, haageana. DISCUSSION Zinnias proved to be a temperamental crop; the greatest success in all aspects of this research occurred during the year when temperatures and light intensities were high. A Mexican native, they thrive in hot, sunny weather and tend to decline in cool, cloudy weather, despite supplemental lighting and elevated greenhouse temperatures. This tenden- cy hindered research as growth, flowering and pollen produc- tion of L, elegans, L, haageana (2n and 4n) and some of the hybrids decreased from October through March. L, angustifolia and the hybrid, L, angustifolia x L, ele ans, appeared the least affected by seasonal variation. This seasonal effect could have an impact on potential commercial aspects of L, ele ans and L, haageana. Produc- tion of zinnias as a bedding plant may not be affected, considering that at the time of year when the crop is matur- ing, the light intensity and temperature are increasing. However, zinnias as a cut flower crop produced in the winter could be adversely affected by the seasonality exhibited by commercial cultivars. Measures could be taken to supplement their environment, but this may be expensive for the grower. Hybridization with L, angustifolia could result in develop- ment of commercial cultivars which continue to grow and 42 43 flower under suboptimal conditions, a trait observed among some of the interspecific hybrids. Hybridization is possible between L, angustifolia and L, haageana (2n) as well as between L, angustifolia and L, elegans. The partial homology seen between chromosomes of these three species could lend support to the theory of Raman, et a1. and Ramalingam, et al. that the haploid genome of each species is composed of two chromosome sets with all three species having one 6 chromosome set in common (21,22). However, the partial homology may not be sufficient for the production of viable pollen. Percentage of stained pollen among hybrids is very low and preliminary crossing of hybrids with parents has been unsuccessful. Possibly, an intensive program of backcrossing with an expanded number of plants implemented when the plants are growing and flowering vigorously might be more successful. It may be necessary to induce tetraploidy in the hy- brids to allow more complete pairing of chromosomes and production of viable pollen. This would necessitate breed- ing of hybrids and species at the tetraploid level with future reduction of the genome of the desired plant to the diploid level, if necessary, as few successful cultivars are tetraploids. The commercial floriculture industry could benefit from further research with these hybrids. Developing commercial cultivars through breeding with the interspecific hybrids would broaden the narrow germ plasm base of zinnias. There 44 are several desirable traits of L, angustifolia which have been exhibited by the hybrids and could be incorporated into comercial cultivars of L, elegans and L, haa eana. The narrow foliage allows light to penetrate to the leaf axils, making the plants responsive to growth regula- tors and photoperiod manipulation, a valuable trait for efficient coordination of bedding plant and cut flower pro- duction. The compact branching growth and constant free- flowering habit extends the plants’ decorative life in the garden and allows yearround growth in the greenhouse. The disease resistance would increase consumer satisfaction with garden performance and allow development of zinnia as a cut flower crop as the chronic problem of controlling powdery mildew during the fall and winter months would be eliminated. Such traits would further enhance the popularity of cultivars of L, elegans. Incorporation of these traits into commercial cultivars of the less popular L, haageana could increase the marketability of this species. Its erect, branching habit, disease resistance and continuous produc— tion of variegated flowers could be advertised to increase its appeal as a novelty type of zinnia. L, haageana has also been a source for development of cultivars with multi- colored flowers through hybridization with L, elegans. Development of such cultivars through breeding between dis- ease resistant lines of L, haageana and L, elegans would 45 insure that the product would have genetic resources of resistance as well as unusually colored flowers. This research has shown that there is potential for developing disease-resistant, culture-responsive cultivars of zinnias in the future. LIST OF REFERENCES 10. 11. LIST OF REFERENCES Agricultural Statistics Board. 1987. Floriculture Crops, 1986 Summary. Washington, DC, USDA, NASS. Apr. SpCr 6-1(87). pp. 3-10. Anderson, K. 1971. The behavior of powdery mildew conidia (Erysiphe cichoracearum) on the leaves of re- sistant and susceptible species of Zinnia. M.S. Thesis. Penn. State University. Anonymous. 1987. All America Selections 1933-1987. Seed World 125(1): 13-22. Baldwin, J.T.,Jr. 1938. Kalanchoe: the genus and its chromosomes. Amer. J. Bot. 25: 572-579. Bose, S. and U.C. Panigrahi. 1969. Studies on induced polyploidy in Zinnia linearis Benth. Cytologia 34: 103-111. Boyle, T.H. and D.P. Stimart. 1982. Interspecific hybrids of Zinnia elegans Jacq. and L, gngugtifolia HBK: embryology, morphology and powdery mildew resis- tance. Euphytica 31: 857-867. Boyle, T.H. and D.P. Stimart. 1983. Developmental responses of Linnia to photoperiod. J. Amer. Soc. Hort. Sci. 108(6): 1053-1059. Boyle, T.H., D.P. Stimart and M.S. McIntosh. 1986. Seasonal variation in vegetative and reproductive de- velopment in Zinnia elegans Jacq. J. Amer. Soc. Hort. Sci. 111(2): 260-266. Brewbaker, J.L. and B.H. Kwack. 1963. The essential role of calcium ion in pollen germination and pollen tube growth. Amer. J. Bot. 50(9): 859-865. Craig, R. 1976. Flower seed industry. In: Bedding Plants. 2nd ed. J. Mastalerz, ed. Penn. Flower Growers. University Park, PA. pp. 25-46. Crop Reporting Board. 1981. Floriculture Crops, Pro- duction Area and Sales, 1979 and 1980. Intentions for 1981. Washington DC, USDA, ESS. Mar. SpCr6-l(81). p. 10. 46 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 47 Darlington, C.D. and L.F. LaCour. 1976. The Handling of Chromosomes. 6th ed. John Wiley and Sons, NY. p. 117. Duffy, R.K. 1981. The inheritance of a male sterile, apetalous inflorescence and narrow leaf shape in Zinnia elegans Jacq. M.S. Thesis. Michigan State University. Ewart, L.C. 1984. Personal communication. Gupta, P.K. and R. Koak. 1976. Induced autotetra- ploidy in Zinnia elegans Jacq. Cytologia 41: 187-191. Jones, J.J. and D.L. Strider. 1979. Susceptibility of zinnia cultivars to bacterial leaf spot caused by Lanthomonas nigromaculaaa f. sp. zinniae. Plant Dis. Rptr. 63(6): 449-453. Metcalf, H.N. and J.N. Sharma. 1971. Germ plasm re- sources of the genus Zinnia L. Economic Botany 25: 169-216. Olorode, O. 1970. The evolutionary implications of interspecific hybridization among four species of Zinnia Sect. Mendezia (Compositae). Brittonia 22: 207-216. Powell, A.M. and B.L. Turner. 1963. Chromosome num- bers in the Compositae. VII. Additional species in the southwestern United States and Mexico. Madrono 17: 128-140. Power, P. 1986. Personal communication. Ramalingam, R.S., S.R.S. Rangasamy and V.S. Raman. 1971. The cytology of an interspecific hybrid in Zinnia. Cytologia 36: 522-528. Raman, V.S., S.R.S. Rangasamy and R.S. Ramalingam. 1976. Cytomorphology and stability of induced variants in Zinnia linearis. Cytologia 41: 201-206. Shahin, S.S., W.F. Campbell, L.H. Pollard and A.R. Hamson. 1971. Interspecific hybrids of Zinnia peruviana and L, elegans through embryo culture. J. Amer. Soc. Hort. Sci. 96(3): 365-367. Smith, L. 1947. The aceto-carmine smear technique. Stain Tech. 22: 17-31. Stimart, D.P. and D.J. Brown. 1982. Regulation of postharvest flower senescence in Zinnia elegans Jacq. Scientia Hort. 17: 391-396. 26. 27. 28. 29. 30. 31. 48 Terry-Lewandowski, V.M., G.R. Bauchan and D.P. Stimart. 1984. Cytology and breeding behavior of interspecific hybrids and induced amphiploids of Zinnia elegans and Zinnia angugtifolia. Can. J. Genet. Cytol. 26: 40-45. Terry-Lewandowski, V.M. and D.P. Stimart. 1983. Mul- tiple resistance in induced amphiploids of Zinnia elegans and L, angastifolia to three major pathogens. Plant Disease 67(12): 1387-1389. Terry-Lewandowski, V.M. and D.P. Stimart. 1985. The inheritance of resistance to powdery mildew in inter- specific hybrids and induced amphiploids of Zinnia elegans Jacq. and L, anguatifolia HBK. Euphytica 34: 483-487. Torres, A.M. 1963. Taxonomy of Zinnia. Brittonia 15: 1-25. Turner, B.L., M. Powell and R.M. King. 1962. Chromo- some numbers in the Compositae. VI. Additional Mexican and Guatemalan species. Rhodora 64: 251-271. Voight, A.O. 1987. The '86 bedding plant season judged 'average': expansion to slow somewhat in ’87. Penn. Flower Growers Bulletin 373(1): 1-12.