Cytogenetics of Hibiscadelphus (Malvaceae): a Meiotic Analysis of Hybrids in Hawaii Volcanoes National Park L

Cytogenetics of Hibiscadelphus (Malvaceae): A Meiotic Analysis of Hybrids in Hawaii Volcanoes National Park l

GERALD D. CARR 2 AND JAMES K. BAKER 2

ABSTRACT: The chromosome number of 2n = 20 II reported here is the first determination for Hibiscadelphus giffardianus Rock and H. distans Bishop & Herbst. An earlier report of the same number for H. hualalaiensis Rock is substantiated. Cytogenetic analysis of H. x puakuahiwi Baker & Allen, a hybrid cross between H. giffardianus and H. hualalaiensis suggests that chromosome pairing and microsporogenesis are not affected in the first generation. However, although chromosome pairing is normal at diakinesis in the seven F 2S analysed, subsequent stages of meiosis and microsporogenesis are severely disturbed and result in the formation of a high percentage of abnormal meiospores in some of the plants. Nevertheless, the apparent lack of hybrid breakdown in two of the F 2 plants and the presence of probable F 3S in one area suggest that gene flow between the two taxa is possible, and, therefore, must be taken into consideration if these two endangered species are to be maintained as distinct entities.

Hibiscadelphus has been collected on the giffardianus. Subsequently, park personnel islands of Kauai, Hawaii, and Maui, but, collected seeds from F 1 trees with the same according to Bishop and Herbst (1973), intention ofpropagating more H giffardianus. all known wild populations of the five Accordingly, F 2 s were unknowingly culti species recognized by them are now extinct vated until the existence of the hybrids was or imminently threatened. Fortunately, detected in 1973 and described by Baker and Hibiscadelphus can be cultivated and has Allen (1976). been established in various localities includ The occurrence of these cultivated hybrids ing Kipuka Puaulu and Kipuka Ki in Hawaii (cf. Figure lA) stimulated the present inves Volcanoes National Park. tigation in which it was our intent to char One result of cultivation was the spon acterize the nature and degree of cytogenetic taneous production of hybrid seeds where differentiation between H. giffardianus and H. giffardianus Rock and H. hualalaiensis H. hualalaiensis. Rock were planted close together in Kipuka Puaulu. Park personnel unwittingly collected some ofthese hybrid seeds from H. giffardia nus and planted the propagated hybrid seed lings in various areas of the park with the MATERIALS AND METHODS intention of increasing the numbers of H. Floral bud material for meiotic analysis was fixed in a mixture ofchloroform, absolute

J This project was supported in part by contract ex ethanol, and glacial acetic acid (6:3:1). 800060031 from the National Park Service. Manuscript Anthers were squashed in acetocarmine and received 30 June 1976. mounted according to Beeks' permanent 2 University of Hawaii, Department of Botany, 3190 Maile Way, Honolulu, Hawaii 96822; and Hawaii squash method (Beeks 1955). Slides were Volcanoes National Park, Mauna Loa Field Station, examined and photographed with the aid of Hawaii 96718. phase contrast optics. 191 192 PACIFIC SCIENCE, Volume 31, April 1977

FIGURE I. Hibiscadelphus flowers and microsporogenesis. A, above-H. gi!fardianus, center-H. x puakuahiwi,

below-H. hualalaiensis; B, diakinesis in H. hualalaiensis, 2n = 20,,; C, diakinesis in KKHXF2 -1, 2n = 20,,; D, polyad from KPHXF2 -11, five microspores; E, polyad from KKHXF2 -1, six microspores.

TABLE I CHROMOSOME NUMBERS OF Hibiscadelphus

SPECIES CHROMOSOME NUMBER SOURCE

H. distans 2n = 20" Planting, Waimea Arboretum, Haleiwa, Hawaii-seed from holotype, Bishop & Herbst 1318 H. gi!fardianus 2n = 20" Planting, Kipuka Puaulu, Hawaii Volcanoes National Park H. hualalaiensis 2n = 20" Planting, Kipuka Puaulu, Hawaii Volcanoes National Park

Likewise, an F 1 hybrid between H. giffar RESULTS dianus and H. hualalaiensis (H. x puakuahiwi Meiotic prophase in Hibiscadelphus giffar Baker & Allen) exhibited regular bivalent dianus, H. hualalaiensis, and H. distans Bishop formation during meiotic prophase prior to & Herbst is characterized by regular bivalent normal chromosome disjunction and micro formation, and all three species have 20 pairs spore formation (Table 2). of chromosomes (Table I, Figure IB). Later In all seven F 2 plants analysed, chromo stages of meiosis and microspore formation some pairing was essentially normal at in these species are also normal. diakinesis and each had 20 pairs of chromo- Cytogenetics of H ibiscadelphus-CARR AND BAKER 193 somes (Figure 1C), although in a few in TABLE 2 stances one of the larger pairs seemed loosely FREQUENCY OF ABNORMAL MEIOTIC POLYADS IN associated. One of these plants (KPHXF2-5, Hibiscadelphus HYBRIDS Table 2) also exhibited normal microspore development. However, in five other F2 FREQUENCY OF plants microsporogenesis resulted in the HYBRID ABNORMAL POLYADS (%) formation of various frequencies (1-13.1 H. x puakuahiwi-KKHX-l* OAt percent, Table 2) of abnormal polyads along KPHXF2 ·5 0.3 with apparently normal tetrads. Most of the KPHXF2 ·4 1.0 abnormal polyads in these plants contain KKHX1F2 -1 4.6 KPHXF2·8 6.5 five or six microspores of two size classes KPHXF -11 12.5t rather than four equal microspores. Four of 2 KPHXF2·14 13.1 these are generally large and almost equal in KKHXF2·1 53.8 size and one or two are much smaller (Figure * Permanent tag designation. ID). A seventh F2 plant (KKHXF2-1) pro t Levels of abnormality this high were detected in the parental species. duces a very high frequency of abnormal tBased on observation of 176 polyads; all others based on morc than polyads (53.8 percent) along with apparently 300 polyads. normal tetrads. These abnormal polyads usually contain two to eight microspores of its status as a genus apart from Hibiscus and various sizes (Figure IE). lessens the significance of Hutchinson's in Meiosis in all F2 plants appears to be dication that the two genera should be united normal through anaphase I. However, during (Hutchinson 1967). prophase preceeding metaphase II, some of Many other examples of interspecific F 1 the sister chromatids separate precociously in hybrids with normal chromosome pairing some microsporocytes, especially in those of and fertility have been described (e.g., Steb KKHXF2-1. In this plant a high percentage bins 1950: 234) that resemble the present case of microsporocytes exhibit two groups of 20 of Hibiscadelphus. Likewise, reduction of chromatids each at a stage corresponding to vigor and / or fertility of second-generation metaphase II. Whether this condition results hybrids whose F 1S are normal has been noted from disruption of the spindle or because of in several instances (Stebbins 1950: 227) and a timing problem is not yet clear. In any case, has been termed F2 or hybrid breakdown. chromatid movement to the poles is some This phenomenon is due ostensibly to for what haphazard, producing a number of mation of disharmonious gene combinations deficient nuclei that give rise to abnormal that result from meiotic recombination sub microspores of various sizes. sequent to the F 1 generation. Presumably, the more divergent the parental gene pools, the more intense this problem is likely to be. Most of the examples of F2 breakdown DISCUSSION appear to involve reduction of vigor in the The uniform chromosome number of second generation rather than reduction in 2n = 20 II reported here for Hibiscadelphus fertility only, as seems to be the case in distans, H. gifJardianus, and H. hualalaiensis Hibiscadelphus. One very similar situation is agrees with earlier reports in the genus (H. that described by Avers (1953) who found hualalaiensis, Niimoto 1966 and personal that certain interspecific Aster F 1 hybrids communication; Hibiscadelphus sp., Skovsted apparently were fully fertile, but that F2 and 1941). Although 2n = 20 II has also been backcross generations exhibited reduced fer reported for species of Hibiscus (Skovsted tility and thus constituted a barrier to gene 1941), this number has not yet been detected exchange between the two species involved. in Hawaiian members of the genus. Thus, Ofthe seven second-generation Hibiscadel together with other factors, the uniformly phus hybrids discussed here, two seem little unique chromosome number of Hibiscadel affected by hybrid breakdown, four are af phus among Hawaiian Malvaceae supports fected to a moderate degree, and one suffers 194 PACIFIC SCIENCE, Volume 31, April 1977 from severe meIOtIc perturbations. These LITERATURE CITED meiotic abnormalities are probably caused by a timing problem and perhaps by abnor AVERS, C. J. 1953. Biosystematic studies in malities of the spindle apparatus. Similar Aster. II. Isolating mechanisms and some but more completely documented instances phylogenetic considerations. Evolution of meiotic abnormalities resulting from genic 7:317-327. imbalance or other genetic factors include BAKER, K., and S. ALLEN. 1976. Hybrid multiple spindle formation (Vasek 1962), Hibiscadelphus (Malvaceae) from Hawaii. discontinuous or incompact spindle (Dar Phytologia 33: 276. lington and Thomas 1937), and inactivity of BEEKS, R. M. 1955. Improvements in the the spindle after the first telophase (Janaki squash technique for plant chromosomes. Ammal 1941). All of the foregoing meiotic Aliso 3: 131-134. maladies may result in formation ofabnormal BISHOP, L. E., and D. R. HERBST. 1973. A meiospores such as those seen in Figure IE. new Hibiscadelphus (Malvaceae) from However, it may be recalled that not all of Kauai. Brittonia 25: 290-293. the Hibiscadelphus hybrids have disturbed DARLINGTON, C. D., and P. T. THOMAS. 1937. meiosis. In view of the apparent high fertility The breakdown ofcell division in a Festuca ofsome of the second generation hybrids and folium derivative. Ann. Bot. I: 747-761. the probable occurrence of F 3S, it seems HUTCHINSON, J. 1967. The Genera ofFlower highly likely that gene flow between H. ing Plants. Vol. 2. Oxford University Press, giffardianus and H. hualalaiensis is possible. London. 659 pp. These factors should be carefully weighed in JANAKI-AMMAL, E. K. 1941. The breakdown any program designed to protect the genetic ofmeiosis in a male-sterile Saccarum. Ann. integrity of these highly endangered species. Bot. (London) 5: 83-87. NIIMOTO, D. H. 1966. Chromosome numbers of some Hibiscus species and other Mal vaceae.Baileya 14: 29-34. ACKNOWLEDGMENTS SKOVSTED, A. 1941. Chromosome numbers The senior author acknowledges the utility in the Malvaceae. II. C. R. Trav. Lab. of equipment made available through an Carlsberg, ser. physiol., 23(14): 195-242. Intramural Faculty Research Grant from the STEBBINS, G. L. 1950. Variation and Evolu University of Hawaii Research Council. He tion in Plants. Columbia University Press, is also grateful to Clifford Smith for support New York. 643 pp. and encouragement and to Derral Herbst and VASEK, F. C. 1962. "Multiple spindle"-a Yoneo Sagawa for reading a draft of the meiotic irregularity in Clarkia exilis. Amer. manuscript. J. Bot. 49: 536-539.