View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by ScholarSpace at University of Hawai'i at Manoa

Chromosome Numbers of Hawaiian Angiosperms: New Records and Comments1

Michael Kiehn2

Abstract: In this paper chromosome counts for 90 collections representing 67 native Hawaiian angiosperm species and eight hybrids in 22 families are presented and discussed. Included are the first records for 26 species, two sub- specific taxa, eight natural hybrids, and the endemic Pteralyxia (Apocyna- ceae). In four families Hawaiian representatives have been investigated cytologically for the first time. For three species the investigations are the first on Hawaiian material. Seven counts differ from earlier reports in the literature. Implications of the results are discussed in the context of autochthonous chro- mosomal evolution and of colonization events for the Hawaiian Islands.

Chromosome numbers are available for growing root tips, shoot apices, or young nearly 40% of the native Hawaiian species flowers for mitotic counts, or young flower (Carr 1998). In most cases these numbers buds for meiotic investigations of cul- were obtained from counts of single or very tivated at the National Tropical Botanical few individuals. In this paper it is aimed Garden, La¯wa‘i, Kaua‘i, Hawai‘i (NTBG), to improve the knowledge on chromosome and of plants collected in the wild were made numbers for Hawaiian angiosperms, both in a freshly mixed 3:1 solution of ethanol with regard to taxa not previously investi- (96%) :glacial acetic acid. In each field fixa- gated and for additional collections of taxa tion, only one individual was sampled (except already counted. Data for 19 species pre- for degeneri, DL6453, collected and sented here were shared with G. Carr (Uni- fixed by D. Lorence, where several individuals versity of Hawai‘i at Ma¯noa, Honolulu) and may have been sampled in the same fixation). were included in his earlier survey (Carr Additional fixations of root tips were taken 1998, table 1.1) as ‘‘Kiehn unpubl.’’ These from plants cultivated at the Botanical Garden counts are documented and discussed further of the University of Vienna, Austria (HBV ), here and are footnoted ‘‘c’’ in Table 1. from seeds collected in Hawai‘i; the root tips were pretreated in 8-hydroxyquinoline materials and methods (0.002 m, ca. 6 hr at 8–10C). Each root tip fixation represented one individual. The In the course of eight visits to Hawai‘i be- fixed material was stored at 8–10C before tween 1989 and 2003, fixations of actively staining with Giemsa (after hydrolysis in 5 N HCl for 50 min. at 20C) or with Feulgen 1 Parts of this study were supported financially by the (Kiehn 1995). Hot aceto-carmine (2% solu- Austrian Science Foundation (FWF-proj. P-13107-Bio), tion in 45% acetic acid) was used to intensify the O¨ sterr. Forschungsgemeinschaft (proj. 06/3699), the staining if needed. Somatic chromosome Hochschuljubila¨umsstiftung der Stadt Wien (proj. Nr. numbers (2n) were established from any of H-21/94), and the National Tropical Botanical Garden, several tissue sources, including premeiotic La¯wa‘i, Kaua‘i, Hawai‘i. Manuscript accepted 20 August 2004. anthers, young flower buds, young ovaries, 2 Institute of Botany, University of Vienna, Rennweg wall, and root tips. Gametic number 14, A-1030 Vienna, Austria (e-mail: michael.kiehn@ determinations (n) were derived from meiotic univie.ac.at). divisions of microsporocytes. Collection data for each studied sample are listed in Table Pacific Science (2005), vol. 59, no. 3:363–377 1. Voucher specimens have been deposited : 2005 by University of Hawai‘i Press in the herbaria at the University of Vienna All rights reserved (wu), at the National Tropical Botanical Gar-

363 364 PACIFIC SCIENCE . July 2005 den (ptbg), and/or in other herbaria as stated dra have been counted so far; both are tetra- in Table 1. Permanent slides for most of the ploids with n ¼ 24=2n ¼ 48 (Carr 1978, 1998, counts are in the collection of the author. Kiehn and Lorence 1996). The data reported here for Tetraplasandra bisattenuata (2n ¼ 46– results 48) fit well into the picture for this group. campanulaceae: The chromosome num- Chromosome counts for 67 species (90 ber determinations for seven species, five collections) of native Hawaiian angiosperms of which are new counts, are in accor- from 22 families are presented in Table dance with the earlier data for Hawaiian 1. They include the first chromosome num- species summarized by Lammers (1988) ber reports for 26 species, two subspecific and Carr (1998). All Hawaiian taxa of the taxa, eight natural hybrids, and the endemic family cytologically investigated to date ex- genus Pteralyxia (). In four fami- hibit 2n ¼ 28; Carr (1998) considered single lies (Droseraceae, Myrsinaceae, , deviating reports to be erroneous, a view cor- and Viscaceae) Hawaiian representatives have roborated by the results presented here. caryophyllaceae been investigated cytologically for the first : Of the seven Hawai- time. For three indigenous Hawaiian species ian species of Silene, S. alexanderi from Molo- the investigations are the first ones on Hawai- ka‘i reported here is the third to be counted ian material. Counts for seven species differ (see Carr 1998). All three species exhibit from earlier reports; 18 counts are in accor- 2n ¼ 24, in agreement with x ¼ 12 as the pre- dance with earlier published data. Some chro- dominant basic number for Silene (cf., e.g., mosome numbers are reported as a range. Goldblatt and Johnson 2000). cyperaceae This is the case in taxa with high chromo- : The chromosome number some numbers (e.g., in Hedyotis spp. or Cop- for Eleocharis obtusa is established for the rosma ernodeoides) where different counts on first time on Hawaiian material. The result several individuals led to differing results, or of 2n ¼ 10 is in accordance with earlier was caused by chromosomes sticking together counts for this species (e.g., from northeast- during fixation (e.g., in Alyxia, Labordia, Lipa- ern America [Gervais and Cayouette 1985]). droseraceae ris, or Korthalsella). : The first count for Dro- sera anglica on material from Hawai‘i revealed 2n ¼ 40. This result is in accordance with discussion earlier reports for this, by Murı´n and Ma´jov- sky´ (1987) from Europe and by Kondo Discussion of the new data in a taxonomic and Segawa (1988) from Asia. Thus there is context no apparent variation in the chromosome apocynaceae: The results reported number of this species across its whole range here for Alyxia stellata (¼ A. oliviformis, cf. of distribution. In contrast, the genus Drosera Middleton 2000) and for Pteralyxia kauaiensis in general is extremely variable in basic num- (first count for the genus) are in line with pre- bers (ranging from x ¼ 3tox ¼ 9, with addi- vious reports for the genus Alyxia (Skottsberg tional reports of x ¼ 13 and x ¼ 15 [Kondo 1955: 2n ¼ ca. 36, 2n ¼ ca. 39 for A. stellata; and Lavarack 1984, Kondo and Segawa Van der Laan and Arends 1985: 2n ¼ 36 for 1988, Sheikh and Kondo 1995]) and in ploidy two non-Hawaiian Alyxia species). The re- levels (from diploidy to dodecaploidy [Kondo ported range of 2n ¼ 34–36 is caused by the and Lavarack 1984, Sheikh and Kondo relatively large chromosomes being twisted 1995]). around each other in the available fixations. : Two of the three native Ha- araliaceae: According to molecular data, waiian species (V. calycinum, V. re- the Hawaiian genera Munroidendron, Reynold- ticulatum) have already been counted; both sia, and Tetraplasandra are closely related to exhibit 2n ¼ 24 (Carr 1998). The result of each other (Wen et al. 2001). The monotypic 2n ¼ 48 obtained here for Vaccinium cf. den- Munroidendron and one species of Tetraplasan- tatum from Moloka‘i indicates the presence Chromosome Numbers of Hawaiian Angiosperms . Kiehn 365 of tetraploidy in Hawaiian representatives of reports by Borgmann 1964 for taxa from the genus. [see Kiehn and Weber 1998 for The three native Hawaiian Vaccinium discussion]). The genus proves to be very uni- species, together with the southern Polyne- form karyologically. Some variation in chro- sian species V. cereum, form the monophyletic mosome length and size (as discussed by sect. Macropelma of Vaccinium (Vander Kloet Storey 1966) can be seen in Hawaiian mate- 1996). Vander Kloet (1996) hypothesized rial, but this is (at least partly) caused by that the Hawaiian taxa compose a coherent differences in fixation and pretreatments, be- evolutionary unit of occasionally anastomos- cause such variation was also observed in dif- ing selfing lineages. The production of such ferent fixations of the same (e.g., of C. lineages could explain the presence of hawaiensis and of C. paludosa). two ploidy levels among the Hawaiian taxa. Ranges of numbers given for C. grayi, C. The difficulties in identifying the investigated kalihii, C. lessoniana, and C. platyphylla are Vaccinium collection from Moloka‘i could also due to the poor quality of the respective fixa- be due to a hybrid origin of the investigated tions and most probably do not reflect a real collection. It would be desirable to study mei- range of chromosome numbers in these spe- otic divisions in this group or to use molecu- cies. In the case of the C. lessoniana collection, lar fingerprint methods to further evaluate this assumption is supported by the obser- the hypothesis of Vander Kloet (1996). vation of regular meiotic divisions. Regular geraniaceae: The genus Geranium is meiotic divisions are also present in collec- represented in Hawai‘i by six native species. tions of C. hawaiensis, C. kauaiensis, C. paludosa The chromosome number established here var. paludosa, C. sandwicensis, and in two puta- for Geranium cuneatum subsp. tridens from tive hybrids. However, meiotic irregularities East (2n ¼ 48–52) is in the same range were found in the investigated material of C. as the only earlier report available for a grandiflora and C. munroi. In spite of cytoplas- Hawaiian species (G. arboreum:2n ¼ ca. 50 mic bridges and the presence of abnormal [Carr 1978, 1998]) and similar to four reports pollen grains, C. grandiflora clearly revealed for species ranging from n ¼ 17 in regular divisions. However, the 2n ¼ 52 to 2n ¼ 56 (Dawson and Beuzenberg range given for C. munroi could reflect true 2000). A survey of chromosome counts for variation. Collection MK-990913-3/2 of this Geranium (http://www.rjb.csic.es/Geranium/ species exhibited meiotic irregularities, in- index_geranium.html) reports basic num- cluding anaphase bridges, lagging univalents, bers for the genus as x ¼ n ¼ 9, 10, 11, 12, cytoplasmic bridges between pollen mother 13, 14, 15, 16, 17, 21?, 23, 25, with a predom- cells (PMCs), and a high frequency of abnor- inance of x ¼ 14. Although chromosome mal pollen. morphology and numbers proved very useful All investigated (putative) hybrids and for the systematic treatment of the related their (hypothetical) parents are on the same genus Pelargonium (Gibby et al. 1996), there ploidy level and have the same chromosome are no published studies relating the chromo- numbers. Only one of the eight hybrids ana- some data of Geranium to a comprehensive lyzed showed meiotic irregularities (hybrid II taxonomic treatment of the genus. from Maui had degenerated pollen). Pollen gesneriaceae: Counts are now available viability tested by Luegmayr (1993) for 16 ac- for 24 of the 58 Hawaiian species of Cyrtan- cessions of 12 different interspecific hybrids dra and for eight naturally occurring hybrids of Hawaiian Cyrtandra was between 70 and (Table 1 and Storey 1966). The counts cover 99% for 15 of those accessions. Only one ac- all sections of the genus in Hawai‘i as recog- cession showed a very low viability (of 2%). nized by Wagner et al. (1999). Germination rates of seeds from six inter- All collections that allowed an exact count specific hybrids were all approximately 90%, revealed a chromosome number of n ¼ 17 similar to those of true species (unpubl. data). or 2n ¼ 34, as did all other Cyrtandra species Thus, it can be assumed that most natural hy- counted so far (with the exception of three brids between Hawaiian Cyrtandra species are TABLE 1 Chromosome Numbers of Hawaiian Angiosperms

Gametic Somatic Number Number Taxon (n) (2n) Island, Voucher (Herbariuma)

Apocynaceae Alyxia stellata ( J. R. Forst. & G. Forst.) Roem. & Schult. — 34–36 O‘ahu, Kiehn & Imada MK-940911-2/3 (bish, wu) þþþþPteralyxia kauaiensis Caum — 34–36 Kaua‘i, Wood 7701 (ptbg) ¼ Kiehn MK-031224-5 (wu), NTBG cult. # 990087 Araliaceae þþþTetraplasandra bisattenuata Sherff — 46–48 Kaua‘i, Lorence DL 8404 (ptbg) Campanulaceae þþþClermontia arborescens (H. Mann) Hillebr. subsp. — 28 Maui, Kiehn et al. MK-990914-1/1 (wu) waihiae (Wawra) Lammers þþþC. fauriei H. Le´v. — 28 Kaua‘i, Kiehn et al. MK-920831-1/8 (wu) þþþC. pallida Hillebr. — 28 Moloka‘i, Kiehn MK-940901-2/4 (bish) C. parviflora Gaud. ex A. Gray — 28 Hawai‘i, Kiehn & Cuddihy MK-920818-1/2a (wu) þþþCyanea asarifolia St. John — 28 Kaua‘i, Woods 1352 (ptbg) þþþC. fauriei H. Le´v. — 28 Kaua‘i, Kiehn & Flynn MK-900906-1/13 (wu) Trematolobelia kauaiensis (Rock) Skottsb. — 28 Kaua‘i, Kiehn & Luegmayr MK-920831-1/3 (ll, wu) Caryophyllaceae þþþSilene alexanderi Hillebr. — 24 Moloka‘i, Kiehn et al. MK-920816-5/1 (wu) Cyperaceae þEleocharis obtusa (Willd.) Schult. — 10 Kaua‘i, Lorence DL8433 ¼ Kiehn MK-990911-4/1 (wu) #Droseraceae þDrosera anglica Huds. — 40 Kaua‘i, Lorence DL8429 ¼ Kiehn MK-990911-2/1 (wu) Ericaceae þþþVaccinium cf. dentatum Sm. — 48b Moloka‘i, Kiehn et al. MK-920817-2/3 (wu) Geraniaceae þþþGeranium cuneatum Hook. subsp. tridens (Hillebr.) —48–52b Maui, Kiehn et al. MK-920822-2/1 (wu) Carlq. & Bissing Gesneriaceae Cyrtandra calpidicarpa (Rock) St. John & Storey — 34c O‘ahu, Kiehn & Obata MK-890809-2/6 (bish, wu) C. confertiflora (Wawra) C. B. Clarke var. confertiflora —34c Kaua‘i, Kiehn & Kiehn MK-890802-1/1 (wu) C. cordifolia Gaud. — 34c O‘ahu, Kiehn et al. MK-900822-1/1 (wu) þþþC. dentata St. John & Storey — 34 O‘ahu, Takahama et al. MK-990901-3/1 (bish, wu) — 34 O‘ahu, Takahama et al. MK-990901-5/1 (wu) C. grandiflora Gaud. 17c,d — O‘ahu, Kiehn et al. MK-900722-1/1 (ptbg, wu) C. grayana Hillebr. — 34c Maui, Kiehn et al. MK-890723-2/2 (bish, wu) —34c Maui, Smith 3915 (srp) þþþC. grayi C. B. Clarke — 34–36 Maui, Kiehn et al. MK-990913-1/1 (bish, wu) C. hashimotoi Rock — 34c Maui, Kiehn et al. MK-900803-3/3 (wu) —34c Maui, Kiehn et al. MK-990914-1/6 (wu) C. hawaiensis C. B. Clarke 17c 34c O‘ahu, Kiehn & Obata MK-890809-1/2 (bish, ptbg, wu) —34c O‘ahu, Kiehn & Obata MK-890809-1/3 (bish, wu) —34c O‘ahu, Kiehn & Obata MK-890809-1/5 (bish, wu) —34c O‘ahu, Kiehn & Obata MK-890809-1/7 (bish) —34c O‘ahu, Kiehn & Obata MK-890809-2/2 (bish) þþþC. kalihii Wawra — 32–34 O‘ahu, Takahama et al. MK-990902-4/1 (wu) C. kauaiensis Wawra 17c 34c Kaua‘i, Kiehn et al. MK-900724-6/1 (ptbg, wu) 34c Kaua‘i, MK-920825-3/3 (wu) C. kaulantha St. John & Storey — 34c O‘ahu, Kiehn & Obata MK-890809-1/4 (bish) C. laxiflora H. Mann — 34c O‘ahu, Kiehn & Obata MK-890809-1/6 (bish, ptbg) þþþC. lessoniana Gaud. — 32–34 O‘ahu, Takahama et al. MK-990902-2/6 (ptbg) 16–17 32–34 O‘ahu, Takahama et al. MK-990902-2/6a (wu) C. longifolia (Wawra) Hillebr. ex C. B. Clarke — 34c Kaua‘i, Kiehn & Kiehn MK-890720-1/1 (bish, ptbg, wu) C. lysiosepala (A. Gray) C. B. Clarke — 34c Hawai‘i, Kiehn & Cuddihy MK-900812-2/3 (us, wu) þþþC. munroi C. Forbes 15–17d 34 Maui, Kiehn et al. MK-990913-3/2 (wu) — 34 Maui, Kiehn et al. MK-990913-4/1 (bish, wu) C. oenobarba H. Mann — 34c Kaua‘i, Kiehn et al. MK-900823-1/3 (ptbg) þþC. paludosa Gaud. var. microcarpa Wawra — 34 Kaua‘i, Kiehn et al. MK-900823-2/11 (ptbg, us, wu) C. paludosa Gaud. var. paludosa 16–17c 34c Kaua‘i, Kiehn et al. MK-900823-1/13 (us) —34c Maui, Kiehn et al. MK-900803-7/3 (us, wu) —34c Maui, Kiehn et al. MK-990914-1/7 (wu) þþþC. platyphylla A. Gray — 34–36 Maui, Kiehn et al. MK-990914-1/9a (wu) C. propinqua C. Forbes — 34 O‘ahu, Kiehn et al. MK-920909-3/2 (bish, ll, us, wu) C. sandwicensis (H. Le`v.) St. John & Storey 17c 34c O‘ahu, Kiehn et al. MK-900722-2/1 (bish, ptbg, wu) þþþC. spathulata St. John — 34 Maui, Kiehn et al. MK-920823-1/1 (wu) Cyrtandra hybrids: þþC. cordifolia C. propinqua — 34 O‘ahu, Kiehn & Obata MK-890809-2/4 (bish, wu) þþC. hawaiensis C. calpidicarpa — 34 O‘ahu, Kiehn & Obata MK-890809-2/3 (bish, wu) þþC. cf. hawaiensis C. laxiflora — 34 O‘ahu, Kiehn & Obata MK-890809-1/1 (bish, wu) þþC. hawaiensis C. menziesii 17 — Hawai‘i, Kiehn et al. MK-920819-1/1 (wu) þþC. hashimotoi C. paludosa var. paludosa — 34 Maui, Kiehn et al. MK-900803-7/2 (us, wu) þþC. kauaiensis C. longiflora var. longiflora — 34 Kaua‘i, Kiehn & Luegmayr MK-920825-3/2 (wu) þþC. hybrid 5 17 — Maui, Kiehn et al. MK-990914-1/9 (wu) þþC. hybrid II — 32–34e Maui, Kiehn et al. MK-990914-1/10 (wu)

(continues) TABLE 1 (continued)

Gametic Somatic Number Number Taxon (n) (2n) Island, Voucher (Herbariuma)

Loganiaceae Labordia degeneri Sherff — >40 < 60c Kaua‘i, Lorence DL6453 (ptbg) —58–62b Kaua‘i, Flynn 6633 ¼ Kiehn MK-990910-3/7 (wu) L. helleri Sherff — 40–44c Kaua‘i, Lorence DL6364 & 6365 (ptbg) L. waialealae Wawra 30–34b 59–61b Kaua‘i, Lorence DL8432 ¼ Kiehn MK-990911-3/3 (wu) L. waiolani Wawra c. 60b O‘ahu, Takahama et al. MK-990902-2/7 (wu) #Myrsinaceae þþþMyrsine sandwicensis A. DC. — 46 Maui, Kiehn et al. MK-990914-1/2 (bish, wu) Oleaceae Nestegis sandwicensis (A. Gray) Degener & al. — 44(–46) O‘ahu, Takahama et al. MK-990901-2/1a (wu) #Orchidaceae þþþLiparis hawaiensis H. Mann — 26–30 Maui, Kiehn et al. MK-990913-3/4 (wu) Pittosporaceae Pittosporum sp. I — 32–36 O‘ahu, Takahama et al. MK-990901-2/1 (wu) P. sp. II — 32–36 Maui, Kiehn et al. MK-990914-1/3 (wu) þþþP. gayanum Rock — 24 Kaua‘i, Flynn 6622 ¼ Kiehn MK-990910-2/3 (wu) princeps Cham. & Schlechtd. var. anomala Rock — 24c Kaua‘i, Kiehn et al. MK-900823-2/8 (wu) —24c Kaua‘i, Wood 2388 (ny, ptbg, us) P. princeps Cham. & Schlechtd. var. princeps —24b O‘ahu, Perlman 15594 (ny, ptbg) þHeteropogon contortus (L.) P. Beauv. ex Roem. & Schult. 18–20 — Kaua‘i, Flynn 3401 (ptbg) þþþPoa sandvicensis (Reichart) Hitchc. — 42–44 Kaua‘i, Perlman et al. 13334 (mo, ptbg) Primulaceae Lysimachia glutinosa Rock — 72 Kaua‘i, Flynn 6625a ¼ Kiehn MK-990910-2/1a (wu) Coprosma ernodeoides A. Gray — 254–264b Hawai‘i, Kiehn MK-900814-5/1 (wu) C. foliosa A. Gray — 44 Maui, Kiehn et al. MK-890723-1/2 (wu) C. ochracea W. Oliver — 40–44 O‘ahu, Takahama et al. MK-990902-2/3 (bish, wu) Hedyotis centrantoides (Hook. & Arn.) Steud. — ca. 100 O‘ahu, Takahama et al. MK-990902-5/1 (bish, wu) þþþH. schlechtendahliana Steud. subsp. remyi (Hillebr.) Fosb. — 94–100 La¯na‘i, Lorence 8850 (bish, ptbg, us) H. terminalis (Hook. & Arn.) W. L. Wagner & Herbst — 90–100 Kaua‘i, Kiehn MK-890720-1/2 (wu) Psychotria greenwelliae Fosb. — 66–88c Kaua‘i, Kiehn et al. MK-890804-2/1 (bish, wu, ptbg) P. mariniana (Cham. & Schltd.) Fosb. — 66 O‘ahu, Kiehn et al. MK-940910-3/1 (bish) Rutaceae þþþMelicope barbigera A. Gray — 36 Kaua‘i, Flynn 6632 ¼ Kiehn MK-990910-3/6 (canb, ny, wu) M. sp. I — 36 O‘ahu, Takahama et al. MK-990902-2/2 (wu) M. sp. II — 18 f,36g Kaua‘i, Flynn 6630 ¼ Kiehn MK-990910-3/4 (wu) Sapotaceae Pouteria sandwicensis (A. Gray) Baehni & Degener 12(13)b 24–26b Kaua‘i, Flynn 6623 ¼ Kiehn MK-990910-2/4 (wu) Violaceae þþþIsodendrion pyrifolium A. Gray — 16 Hawai‘i, Perlman 13048 (ptbg) # Viscaceae þþþKorthalsella cf. platycaula (Tiegh.) Engl. 14–16 28–32 Kaua‘i, Flynn 6629 ¼ Kiehn MK-990910-3/3 (wu) Note: #, First report for the family from Hawai‘i; þþþþ , first record for the genus; þþþ, first record for the species; þþ, first record for the subspecies, variety, or hybrid; þ, first record for the taxon from Hawai‘i. a Abbreviations follow Holmgren et al. (1990). b Number differs from earlier literature reports. c Reported as ‘‘Kiehn unpubl.’’ in Carr (1998). d Meiotic irregularities observed (see text). e Pollen degenerated. f Based on one seedling. g Based on three seedlings. 370 PACIFIC SCIENCE . July 2005 fully fertile. Of particular interest in this re- (1998). In the light of the divergent data for gard is Smith 2874, a collection of Cyrtandra L. helleri and the low number of counts avail- from O‘ahu that has been independently able, Carr (1998:32) stated that ‘‘additional identified as an F1 hybrid between C. cordifolia clarification and confirmation of the lower and C. laxiflora by W. L. Wagner (Smithso- numbers determined by Kiehn (unpublished) nian Institution) and myself. A seedling from . . . would strengthen the case of autochtho- a fruit of this hybrid regularly sets flowers at nous polyploidy in Labordia.’’ the Botanical Garden, University of Vienna. New data provided here give further Morphologically, this cannot be distin- evidence for the existence of different ploidy guished from one of the parents of the F1 hy- levels in the genus: besides tetraploidy and brid, C. laxiflora. More work on hybrids (in octoploidy (already revealed by earlier counts) particular with larger numbers of plants) is hexaploidy could be present in L. degeneri, L. desirable to further interpret this observation. waialealae, and L. waiolani. These new results, One explanation could be that species differ- however, also raise questions because there ences are based on only a few genes, and are now three species of Labordia (L. helleri, backcrosses of an F1 hybrid with one of the L. waialealae, L. waiolani) for which two dif- parental species yield progeny very similar to ferent ploidy levels are reported. The result the recurrent parental species. This would ex- of 2n > 40 < 60 for L. degeneri (Kiehn in plain the observation of individual F1 hybrid Carr 1998) may indicate another instance plants between sympatrically growing Cyrtan- of infraspecific polyploidy in Labordia. The dra species and the apparent lack of ‘‘hybrid range reported was due to the fact that swarms’’ in Cyrtandra in Hawai‘i. Carr (1995) the field fixation, Lorence DL6453, contained discussed the potential relevance of hybrid- some cells with 2n ¼ ca. 40 and others with ization to plant evolution in the Hawaiian Is- 2n ¼ ca. 60 chromosomes, possibly because lands based on an example in the Asteraceae- individuals at different ploidy levels were Madiinae. sampled in this fixation (this is the only fixa- (geniostomaceae): Re- tion where the number of collected individ- cent molecular work (Struwe et al. 1994) in- uals is not known). In general, the data dicates a close relationship of the genera available for Labordia seem to reflect ongoing Labordia (Hawaiian endemic) and polyploidization at the species level in this ge- (of Pacific distribution). A separation of the nus. No meiotic irregularities were observed two genera from Loganiaceae as Geniosto- in L. waialealae; the range of n ¼ 30–34 is maceae (proposed by Struwe et al. 1994), due to clumping of the chromosomes in the however, is not accepted by the latest Angio- field fixation. sperm Phylogeny Group (2003) classification. myrsinaceae: The count of 2n ¼ 46 for The possibility of autochthonous polyploid Myrsine sandwicensis is the first for the family evolution in the Geniostoma-Labordia relation- in Hawai‘i. The result is in accordance with ship was discussed by Carr (1998) and Motley most other data for the genus (Dawson 1995, and Carr (1998). They gave 2n ¼ 40 for Gen- Murray and DeLange 1999) and is not infor- iostoma rupestre J. R. & G. Forst., a possible mative for phylogenetic considerations. ancestor of Labordia. Based on published data oleaceae: The result obtained for Neste- for Labordia (2n ¼ 80 or 2n ¼ ca. 80 for L. gis sandwicensis is 2n ¼ 44(–46). The reason hirtella, L. waiolani [Carr 1978]; L. helleri, L. for the given range is the presence of two kaalae, L. tinifolia, L. fagraeoidea, L. hedyosmi- smaller elements, most likely satellites but al- folia, L. hirtella, L. hosakana, L. waialealae, L. ternatively representing two small chromo- waiolani [Motley and Carr 1998]) and on un- somes. Skottsberg (1955) reported 2n ¼ 44 published counts by me (Labordia degeneri: for N. sandwicensis. Counts made by Dawson 2n > 40 < 60, L. helleri:2n ¼ 40–44 [cited (1995) revealed 2n ¼ 46 for four Nestegis spe- in Carr 1998]), the occurrence of at least cies from New Zealand. More cytological in- two different ploidy levels in this Hawaiian vestigations are needed to clarify the situation endemic genus seemed apparent to Carr in this genus. Chromosome Numbers of Hawaiian Angiosperms . Kiehn 371

orchidaceae: No chromosome counts include diploids (mostly n ¼ 10 [e.g., Ahsan for native Hawaiian orchids were available up et al. 1994]) and tetraploids (either n ¼ 18 to now. The reason for the range given here [e.g., Bir and Chauhan 1990] or n ¼ 20 [e.g., for hawaiensis from Maui (2n ¼ 26–30) Gill et al. 1980, Sinha et al. 1990, Ahsan et al. is clumping of chromosomes in the field 1994]). The same range of tetraploid num- fixation. Variation of chromosome numbers bers was obtained from analyses of field fixa- ranging from 2n ¼ 20 to 2n ¼ 76 is docu- tions of Heteropogon contortus from Hawai‘i. mented for Liparis (Goldblatt and Johnson The uncertainty reported here is due to sev- 2000), and B chromosomes are also reported eral close associations of chromosomes not al- in the genus (Vij and Sood 1986). Thus the lowing a more exact count. It is not clear data reported here have no obvious syste- whether these associations reflect partial mei- matic implications. otic pairing (secondary associations) or are ar- pittosporaceae: Of the 11 Pittosporum tifacts of the fixations. Such associations could species endemic to Hawai‘i a previous count explain the variation in chromosome numbers existed only for Pittosporum glabrum (2n ¼ 24 reported for this species in the literature. [Carr 1978]). Outside Hawai‘i, x ¼ n ¼ 12 The report of 2n ¼ 42–44 for sandvi- is the only reported number in the genus censis is the first report for a Poa species native (Dawson 1995), and no polyploids have been to Hawai‘i. Based on available chromosome reported (Fedorov 1974, Carr and McPher- data, x ¼ 7 is the prevalent basic number son 1986). The new result for P. gayanum in the genus Poa (Goldblatt and Johnson (2n ¼ 24) matches the earlier results. How- 2003), indicating that Poa sandvicensis is hexa- ever, 2n ¼ 32–36 obtained for two collections ploid. The fixation analyzed was a part of a of Pittosporum, one from O‘ahu and the other young inflorescence of a single individual cul- from Maui (neither identified to species), in- tivated at the NTBG. Mitotic numbers were dicate the occurence of a new basic number obtained from filament tissue that was pre- in the genus. In this context, the report of sumably diploid (2n). Meiotic divisions of mi- n ¼ 18 for the genus Citriobatus by Gros crosporocytes were present in all stages but (1965, cited in Fedorov 1974) deserves atten- were characterized by numerous univalents tion. The basic number for the family and and multivalents and, therefore, did not allow the genus Pittosporum is probably x ¼ 6, and a count. These irregularities in meiosis could the two collections from O‘ahu and Maui are be the effect of the hexaploid state of the thus hexaploids. Further investigations, in- plant. cluding meiotic divisions and pollen viability, primulaceae: The native Hawaiian spe- are required to better understand the implica- cies of Lysimachia belong to two subgenera, tions of the new reports. subg. Palladia and subg. Lysimachiopsis. They plantaginaceae: Two ploidy levels, are assumed to be the result of two separate 2n ¼ 12 and 2n ¼ 24, have been reported for colonization events (Wagner et al. 1999). the Hawaiian endemic Plantago princeps (Carr This idea is supported by the cytological data: 1998). Three counts for this species are re- L. mauritiana, the only Hawaiian represen- ported here, one for P. princeps s.str., and two tative of subgenus Palladia, is diploid with for P. princeps var. anomala. All three reveal x ¼ 10 (2n ¼ 20 [Carr 1978, 1998]). In 2n ¼ 24 chromosomes. The second chromo- contrast, both Hawaiian species of subg. Lysi- somally known species from Hawai‘i, P. pa- machiopsis counted so far exhibit high chro- chyphylla, is also tetraploid with 2n ¼ 24 mosome numbers, most probably based on (Skottsberg 1955, Moore 1973). Thus, tetra- x ¼ 9; 2n ¼ ca. 54 for L. hillebrandii was re- ploidy seems to be prevalent in the genus in ported by Skottsberg (1955), and the count Hawai‘i. More counts are needed to verify given here for Lysimachia glutinosa (2n ¼ 72) the existence of diploidy in Hawaiian popula- matches an earlier report by Carr (1978). It tions of Plantago. would be desirable to sample additional Ha- poaceae: Based on previous reports, Het- waiian species of subg. Lysimachiopsis to assess eropogon contortus populations outside Hawai‘i the extent of polyploidy. Chromosome size in 372 PACIFIC SCIENCE . July 2005

Lysimachia glutinosa is 3–5 mm in nearly fully metaphase), and it is very difficult to obtain contracted prometaphase, the same size range clearly spread metaphase plates. Thus, the as given for other Lysimachia species (Ko et ranges of the counts reported here do not al. 1986). necessarily reflect a real range of chromo- rubiaceae: Coprosma is represented in some number variation between individuals. Hawai‘i by 13 species, one of which, C. erno- First counts for Hawaiian Psychotria deoides, has black fruits and is considered to (Kiehn and Lorence 1996) indicated the exis- represent an introduction independent from tence of two ploidy levels (6x,8x) in this the remaining species, which have orange genus in Hawai‘i. The new count for P. mar- fruits (Wagner et al. 1999). The cytological iniana corroborates the earlier result for this situation in Coprosma has been discussed by species (Kiehn and Lorence 1996). Unfortu- Kiehn (1996). New counts for orange-fruited nately all other attempts to obtain results for species (2n ¼ 44, 40–44) are in accordance Hawaiian Psychotria species have failed up to with all earlier reports for Hawaiian taxa, now due to ‘‘self-tanning’’ effects in the field which exclusively were tetraploid (5 of 12 fixations. Tannic acids present in tissues in- species counted). A new count for C. erno- terfered with the field fixations in P. fauriei, deoides from Hawai‘i Volcanoes National Park P. greenwelliae, P. hathewayi, P. hawaiiensis, P. revealed 2n ¼ 254–264 chromosomes in each hexandra, and P. mauiensis, resulting in a pre- of six analyzed cells from two seedlings. This fixation of the chromosomes, which then is the highest chromosome number reported clumped together, making counts virtually in Rubiaceae so far. It also indicates the pres- impossible (see Greilhuber 1987, Kiehn 1995 ence of two different ploidy levels in C. erno- for discussions of this problem). Fixation of deoides: an earlier count of this species (Kiehn root tips of very young seedlings, which are 1996) based on material from East Maui and nearly free of tannins, may provide a means earlier data from the literature (Skottsberg to obtain accurate chromosome counts in 1955) showed 2n ¼ ca. 220. More cytological these species. studies are needed to further elucidate this rutaceae: Previous counts for Hawaiian situation (e.g., whether the two ploidy levels species of Melicope established two ploidy distinguish populations of C. ernodeoides from levels: 2n ¼ 36 for M. elliptica (Carr 1978) of East Maui and Hawai‘i Island). Evidence for sect. Apocarpa,and2n ¼ 72 for M. wawraeana apomixis in the tetraploid C. waima from of sect. Megacarpa (Guerra 1984, as Pelea wa- New Zealand (Heenan et al. 2002) suggests wraeana). Counts from outside Hawai‘i are the possible occurrence of this mechanism in available for the two New Zealand species of C. ernodeoides and could explain the regular Melicope: M. simplex and M. ternata are both fruit set in this species despite meiotic irregu- reported to have n ¼ 18 and 2n ¼ 36 chro- larities that are to be expected with such mosomes (Guerra 1984, Dawson and Beu- high chromosome numbers. Further work is zenberg 2000). New results reported here for needed to obtain additional chromosome M. barbigera (sect. Apocarpa) and for an un- data and to test for the possible occurrence identified species from O‘ahu (both 2n ¼ 36) of apomixis in C. ernodeoides. match the earlier data. However, a third For Hedyotis, three new results are re- ploidy level is tentatively identified based ported here: a first count for H. schlechten- on the observations made on four seedlings dahliana subsp. remyi (2n ¼ 94–100), a new of an unidentified species from Kaua‘i (Flynn island report for H. centranthoides from 6630): one seedling clearly exhibited 2n ¼ 18, O‘ahu (2n ¼ ca. 100), and an additional count but the three others showed 2n ¼ 36. The for H. terminalis from Kaua‘i (2n ¼ 90–100). implications of these results can only be eval- The new data fully support the earlier reports uated when further counts become available. of very high chromosome numbers by Skotts- sapotaceae: The report by Skottsberg berg (1955) and Kiehn (1996). The chromo- (1955) of 2n ¼ 48 for Pouteria sandwicensis somes of Hawaiian Hedyotis are dotlike and could be interpreted as tetraploidy on a basic very small (not exceeding 0.5 mm in mitotic number of x ¼ 12 ( Johnson 1991). Chromo- Chromosome Numbers of Hawaiian Angiosperms . Kiehn 373 some data are also available for four Pouteria Melicope (Rutaceae), Carr (1978) stated that species from outside Hawai‘i, including P. only one determination was available. With aningeri, P. laevigata, and P. pariry (2n ¼ 28), chromosome numbers for five taxa now and P. obovata (2n ¼ 26) ( Johnson 1991). The known (ca. 10% of the Hawaiian species) counts for Pouteria sandwicensis presented here and indications for at least two, if not three, (n ¼ 12–13 and 2n ¼ 24–26) differ from the different ploidy levels, the need for further in- report of 2n ¼ 48 by Skottsberg (1955) for vestigation of this group is obvious. the same species. It appears that there may chromosome numbers of non- be both diploids and tetraploids in this spe- endemic hawaiian taxa: Carr (1998:6) cies, with x ¼ 12 or 13 as the basic number. mentioned that counts for 38 nonendemic This result fits well with other cytological Hawaiian species are available only for non- reports for tribe Chrysophylleae of the Sapo- Hawaiian material. In this paper the first taceae ( Johnson 1991). More cytological reports of chromosome numbers based on investigations on Pouteria sandwicensis are de- Hawaiian material for three of these taxa rep- sirable to clarify this situation. resenting three different genera are provided violaceae: The new count for Isoden- (Drosera anglica, Eleocharis obtusa, Heteropogon drion pyrifolium (2n ¼ 16) is in accordance contortus). The results obtained for all three with the earlier results given by Carr (1985) species are in accord with earlier counts on for I. longifolium and I. subsessilifolium (n ¼ 8). extra-Hawaiian material; no change in chro- Three of the four species of this endemic Ha- mosome number has occurred in these taxa waiian genus are now known cytologically. after their arrival in the Hawaiian Islands. viscaceae: The data presented here autochthonous chromosome evolu- for the Hawaiian Korthalsella cf. platycaula tion in hawai‘i: Carr (1998:26 and table (n ¼ 14–16, 2n ¼ 28–32) match the results 1.1) noted reports indicating interspecific or for the three Korthalsella species from New interpopulational variation of the chromo- Zealand (2n ¼ 28) investigated by Beuzen- some number in at least 38 genera of Hawai- berg and Groves (1974). The range of the ian angiosperms. Some of the reports of such counts reported here was due to the large variation (e.g., in the Campanulaceae, and in chromosomes (ca. 8–10 mm in mitotic meta- the Asteraceae genera Argyroxiphium, Bidens, phase) that were not spread well enough to Tetramolopium, and Wilkesia) were considered allow more exact counts. by Carr (1998) ‘‘to be almost certainly the result of faulty cytological determinations,’’ a statement that, for the Campanulaceae, is Discussion of general aspects supported by the results reported here. The status of chromosome studies in additional new data for Alyxia and Silene give hawai‘i: Before this study, there were no no indication for chromosomal variation in chromosome data available for 53 genera na- these two genera in the Hawaiian Islands. tive to Hawai‘i (Carr 1998). In this paper the There are some Hawaiian plant groups first data for three nonendemic genera and mentioned by Carr (1998) as potentially ex- species are provided (see next sections). In hibiting autochthonous chromosomal evolu- addition, chromosome numbers were deter- tion. In this context Carr (1998:26) stated mined for one endemic species in each of that ‘‘it would be especially gratifying to two genera (Myrsine, Pteralyxia) not previ- have more data for Acacia, Coprosma, Gouania, ously studied in Hawai‘i. However, 48 genera Hedyotis, Hibiscus, Lysimachia, Phyllostegia, with species native to Hawai‘i still remain un- Plantago, Psychotria, Ranunculus, and Rumex.’’ known cytologically. In this paper new data for four of those 11 Knowledge about the largest genus of Ha- genera are included. New evidence for varia- waiian angiosperms, Cyrtandra, is increased tion (different ploidy levels) is provided for by one-third (chromosomes of 24 of 58 spe- the genera Coprosma, Hedyotis, and Psychotria, cies and eight hybrids now counted). For the but the new data increase the doubts about third largest genus in the Hawaiian Islands, the existence of two ploidy levels in the genus 374 PACIFIC SCIENCE . July 2005

Plantago in Hawai‘i. The new counts provided Melicope. They are also not contradicting the here give further evidence for chromosomal hypothesis of reticulate speciation in Erica- variation in four of the 16 genera mentioned ceae (Vander Kloet 1996). The cytological by Carr (1998) as examples of real variation: picture is not yet clear for Psychotria. Coprosma, Labordia, Lysimachia, and Psychotria. More than 80% of the cytologically inves- conclusions tigated Hawaiian taxa are polyploids (Carr 1998). Chromosome numbers of Hawaiian The new data support the conclusions by species are often among the highest records Kiehn and Lorence (1996) and Carr (1998) for the corresponding genera (Kiehn and that the degree of chromosomal variation Lorence 1996, Carr 1998). Most polyploid in autochthonous Hawaiian plant groups in species are considered to be paleopolyploids, most cases reflects the variation of their con- with polyploidy evolved before dispersal of tinental ancestors—in relationships variable the ancestors to the Hawaiian archipelago outside Hawai‘i (such as Psychotria) the prob- (Carr 1998). Cases of potentially autochtho- ability of variation being found in Hawai‘i is nous polyploid evolution on Hawai‘i were also higher. Insular variation might be the re- listed and discussed by Carr (1998). Based on sult of different colonization events by taxa the findings reported here, there are addi- with different chromosome numbers (e.g., in tions to this list: Coprosma ernodeoides (Rubia- Lysimachia and Coprosma). However, autoch- ceae), Melicope (Rutaceae), and Vaccinium thonous insular chromosome evolution, con- (Ericaceae). Further investigations are re- sidered by Stuessy and Crawford (1998) to quired to evaluate the observed variation in be rare, seems to be found preferentially in Pittosporum. groups with continental ancestors already chromosome stasis in autochtho- possessing cytological variability (e.g., Hedyo- nous hawaiian plant groups: In two tis [see also Kiehn and Lorence 1996, Carr groups the existence of chromosome stasis in 1998]). Hawai‘i is further supported. In the genus The role of time in autochthonous chro- Cyrtandra, the new karyological data (espe- mosomal evolution events cannot be eval- cially for the interspecific hybrids) show that uated in most cases because of the lack of the enormous morphological variation of the data related to time of colonization of the lin- 58 endemic species of this genus in Hawai‘i, eage in question. However, in at least two which most probably are the result of a single groups exhibiting chromosome stasis in Ha- colonization event (Samuel et al. 1997, Kiehn wai‘i (Campanulaceae and Cyrtandra) there 2001), is not paralleled by cytological varia- are indications for recent speciation in Ha- tion. The same holds true for the very large wai‘i (Lammers 1988; H. K., unpubl. data). number of endemic species of Campanula- The newly observed cases of chromosomal ceae, composing more than 10% of the native variation in groups with little or no previous flowering plants of Hawai‘i. data (Coprosma ernodeoides, Melicope, Vacci- chromosome data and colonization nium, and possibly Pittosporum) clearly in- events: Besides Cyrtandra, there are indi- dicate the need for further chromosomal cations of single colonization events for the studies of the Hawaiian flora. Hawaiian species of the family Ericaceae (Vander Kloet 1996) and of the genus Labor- acknowledgments dia. Two colonization events are suggested for Coprosma, Psychotria, Lysimachia, and Meli- I thank all institutions in Hawai‘i that facili- cope (Wagner et al. 1999) and a single to very tated the work, especially the National Trop- few colonization events for all ca. 110 species ical Botanical Garden, La¯wa‘i, Kaua‘i; the of Hawaiian Campanulaceae (Givnish et al. Bishop Museum, Honolulu; and the Maui 1995). The available chromosome data sup- Land and Pineapple Company, Maui. Col- port these hypotheses for the Campanulaceae lecting permits obtained from the Ko¯ke‘e and for Cyrtandra, Coprosma, Lysimachia, and State Park, Kaua‘i; the National Park Service Chromosome Numbers of Hawaiian Angiosperms . Kiehn 375

(Volcanoes National Park, Big Island); The evolution in the Hawaiian Islands. Am. J. Nature Conservancy (Moloka‘i); and the U.S. Bot. 82:1574–1581. Army (O‘ahu) are gratefully acknowledged. ———. 1998. Chromosome evolution and Thanks are due to friends and colleagues speciation in Hawaiian flowering plants. who helped in the field and with identifica- Pages 5–47 in T. F. Stuessy and M. Ono, tions of collections, especially L. (Cuddihy) eds. Evolution and speciation of island spe- Pratt, T. Flynn, R. Hobdy, C. Imada, T. cies. Cambridge University Press, Cam- Lammers, D. H. Lorence, J. Obata, H. Op- bridge. penheimer, S. Perlman, C. Puttock, J. Smith, Carr, G. D., and G. McPherson. 1986. Chro- T. Takahama, W. L. Wagner, and K. Wood. mosome numbers of New Caledonian Monika Kiehn and N. Lux (Vienna) were plants. Ann. Mo. Bot. Gard. 73:486–489. most helpful in organizing and carrying out Dawson, M. I. 1995. Contributions to a chro- the excursions and the work in the field. mosome atlas of the New Zealand flora. D. H. Lorence (NTBG, Kaua‘i) and M. 34. Miscellaneous species. N. Z. J. Bot. Moeller (Royal Botanic Gardens, Edinburgh) 33:477–487. are thanked for suggestions and critical com- Dawson, M. I., and E. J. Beuzenberg. 2000. ments on the manuscript. Contributions to a chromosome atlas of the New Zealand flora. 36. Miscellaneous Literature Cited families. N. Z. J. Bot. 38:1–23. Fedorov, A. A., ed. 1974. Chromosome Ahsan, S. M. N., A. A. Vahidy, and S. I. Ali. numbers of flowering plants. Academy of 1994. Chromosome numbers and inci- Sciences of the USSR, Leningrad. dence of polyploidy in Panicoideae (Poa- Gervais, C., and J. Cayouette. 1985. Liste an- ceae) from Pakistan. Ann. Mo. Bot. Gard. note´e de nombres chromosomiques de la 81:775–783. flore vasculaire du nord-est de l’Amerique Angiosperm Phylogeny Group. 2003. An up- IV. Nat. Can. (Que.) 112:319–331. date of the Angiosperm Phylogeny Group Gibby, M., S. Hinnah, E. M. Marais, and classification for the orders and families of F. Albers. 1996. Cytological variation flowering plants: APG II. Bot. J. Linn. and evolution within Pelargonium section Soc. 141:399–436. Hoarea (Geraniaceae). Plant Syst. Evol. Beuzenberg, E. J., and B. E. Groves. 1974. 203:111–142. Contributions to a chromosome atlas of Gill, B. S., S. S. Bir, V. K. Singhal, and Y. S. the New Zealand flora. 16. Miscellaneous Bedi. 1980. Cytological studies on some families. N. Z. J. Bot. 12:211–217. grasses from Pachmarhi forest (central In- Bir, S. S., and H. S. Chauhan. 1990. SOCGI dia). J. Cytol. Genet. 15:51–57. plant chromosome number reports X. J. Givnish, T. J., K. J. Sytsma, J. F. Smith, and Cytol. Genet. 25:322–323. W. J. Hahn. 1995. Molecular evolution, Borgmann, E. 1964. Anteil der Polyploiden , and geographic specia- in der Flora des Bismarckgebirges von tion in Cyanea (Campanulaceae, Lobelioi- Ostneuguinea. Z. Bot. 52:118–173. deae). Pages 288–337 in W. L. Wagner Carr, G. D. 1978. Chromosome numbers of and V. A. Funk, eds. Hawaiian biogeogra- Hawaiian flowering plants and the signifi- phy: Evolution on a hot spot archipelago. cance of cytology in selected taxa. Am. J. Smithsonian Institution Press, Washing- Bot. 65:236–242. ton, D.C. ———. 1985. Additional chromosome num- Goldblatt, P., and D. E. Johnson, eds. 1990. bers of Hawaiian flowering plants. Pac. Index to plant chromosome numbers Sci. 39:302–306. 1986–1987. Monogr. Syst. Bot. 30. ———. 1995. A fully fertile intergeneric ———. 2000. Index to plant chromosome hybrid derivative from Argyroxiphium sand- numbers 1996–1997. Monogr. Syst. Bot. wicense ssp. macrocephalum x Dubautia men- 81. ziesii (Asteraceae) and its relevance to plant ———. 2003. Index to plant chromosome 376 PACIFIC SCIENCE . July 2005

numbers 1998–2000. Monogr. Syst. Bot. cies of Drosera L. (Droseraceae). Bot. J. 94. Linn. Soc. 88:317–333. Greilhuber, J. 1987. ‘‘Self-tanning’’—a new Kondo, K., and M. Segawa. 1988. A cytotaxo- and important source of stoichiometric nomic study in artificial hybrids between error in cytophotometric determination of Drosera anglica Huds. and its certain nuclear DNA content in plants. Plant Syst. closely related species in series Drosera, Evol. 158:87–96. section Drosera, subgenus Drosera, Drosera. Gros, J.-P. 1965. Contribution a` l’e´tude cyto- Kromosomo (Tokyo) 51–52:1697–1709. taxonomique des Pittosporace´es. Mem. Lammers, T. G. 1988. Chromosome num- Mus. Natl. Hist. Nat. Ser. B Bot. 16:61– bers and their systematic implications in 90. Hawaiian Lobelioideae (Campanulaceae). Guerra, M. dos S. 1984. New chromosome Am. J. Bot. 75:1130–1134. numbers in Rutaceae. Plant Syst. Evol. Luegmayr, E. 1993. Pollen of Hawaiian 146:13–30. Cyrtandra (Gesneriaceae) including notes Heenan, P. B., M. I. Dawson, and R. A. Bick- on Southeast Asian taxa. Blumea 38:25–38. nell. 2002. Evidence for apomictic seed Middleton, D. J. 2000. Revision of Alyxia formation in Coprosma waima (Rubiaceae). (Apocynaceae). Part 1: Asia and Malesia. N. Z. J. Bot. 40:347–355. Blumea 45:1–146. Holmgren, P. K., N. H. Holmgren, and L. C. Moore, R. J., ed. 1973. Index to plant chro- Barnett. 1990. Index herbariorum. Part 1: mosome numbers 1967–1971. Regnum The herbaria of the world. 8th ed. New Veg. 90. York Botanical Garden, Bronx, New York. Motley, T. J., and G. D. Carr. 1998. Artificial Johnson, M. 1991. Cytology. Pages 15–22 hybridization in the Hawaiian endemic ge- in T. D. Pennington, ed. The genera of nus Labordia (Loganiaceae). Am. J. Bot. Sapotaceae. Royal Botanic Gardens, Kew, 85:654–660. U.K. Murı´n, A., and J. Ma´jovsky´. 1987. Karyologi- Kiehn, M. 1995. Chromosome survey of the cal study of the Slovak flora XIX. Acta Fac. Rubiaceae. Ann. Mo. Bot. Gard. 82:398– Rerum Nat. Univ. Comenianae Bot. 34:3– 408. 20. ———. 1996. Chromosomes of Rubiaceae Murray, B. G., and P. J. DeLange. 1999. occurring in Malesia, the Philippines, New Contributions to a chromosome atlas of Guinea, and the Pacific. Opera Bot. Belg. the New Zealand flora. 35. Miscellaneous 7:249–260. families. N. Z. J. Bot. 37:511–521. ———. 2001. South Pacifc and Hawaiian Samuel, R., W. Pinsker, and M. Kiehn. Cyrtandra: Molecular and micromorpho- 1997. Phylogeny of some species of Cyr- logical studies. Malay. Nat. J. 55:21–27. tandra (Gesneriaceae) inferred from the Kiehn, M., and D. H. Lorence. 1996. Chro- atp/rbcL cpDNA intergene region. Bot. mosome counts on angiosperms cultivated Acta 110:503–510. at the National Tropical Botanical Gar- Sheikh, S. A., and K. Kondo. 1995. Differen- den, Kaua‘i, Hawai‘i. Pac. Sci. 50:317–323. tial staining with orcein, Giemsa, CMA, Kiehn, M., and A. Weber. 1998. Chro- and DAPI for comparative chromosome mosome numbers of Malayan and other study of 12 species of Australian Drosera paleotropical Gesneriaceae. II. Tribes Tri- (Droseraceae). Am. J. Bot. 82:1278–1286. chosporeae, Cyrtandreae and Epithema- Sinha, R. R. P., A. K. Bhardwaj, and R. K. teae. Beitr. Biol. Pflanz. 70 (1997): 445–470. Singh. 1990. SOCGI plant chromosome Ko, S. C., I. T. Im, and Y. S. Kim. 1986. A number reports IX. J. Cytol. Genet. 25: cytotaxonomic study on the genus Lysima- 140–143. chia in Korea. Korean J. Plant Taxon. Skottsberg, C. 1955. Chromosome numbers 16:187–197. in Hawaiian flowering plants. Ark. Bot. Kondo, K., and P. S. Lavarack. 1984. A cyto- 59:63–70. taxonomic study of some Australian spe- Storey, W. B. 1966. Chromosome numbers Chromosome Numbers of Hawaiian Angiosperms . Kiehn 377

in Cyrtandra. Pages 31–33 in H. St. John. Cytotaxonomy of Apocynaceae. Genetica Monograph of Cyrtandra (Gesneriaceae) (The Hague) 68:3–35. on , Hawaiian Islands. Bernice P. Vij, S. P., and R. Sood. 1986. Cytology in re- Bishop Mus. Bull. 229. lation to morphology in Spiranthes lancea Struwe, L., V. A. Albert, and B. Bremer. (Sw.) Backer. Pages 147–161 in S. P. Vij, 1994 [1995]. and family level ed. Biology, conservation, and culture of classification of the . Cladistics orchids. Affiliated East-West Press, New 10:175–206. Delhi. (Cited in Goldblatt and Johnson Stuessy, T. F., and D. J. Crawford. 1998. 1990.) Chromosomal stasis during speciation Wagner, W. L., D. R. Herbst, and S. H. in angiosperms of oceanic islands. Pages Sohmer. 1999. Manual of the flowering 307–324 in T. F. Stuessy and M. Ono, plants of Hawai‘i. Rev. ed. Bishop Mus. eds. Evolution and speciation of island spe- Spec. Publ. 97. University of Hawai‘i cies. Cambridge University Press, Cam- Press/Bishop Museum Press, Honolulu. bridge. Wen, J., G. M. Plunkett, A. D. Mitchell, Vander Kloet, S. P. 1996. of and S. J. Wagstaff. 2001. The evolution of Vaccinium Section Macropelma (Ericaceae). Araliaceae: A phylogenetic analysis based Syst. Bot. 21:355–364. on ITS sequences of nuclear ribosomal Van der Laan, F. M., and J. C. Arends. 1985. DNA. Syst. Bot. 26:144–167.