The Chromosome Complements of Some New Zealand
[Read before the Canterbury Branch, September, 1941; received by the Editor, October 31, 1941; issued separately, March, 1942.]
This investigation was undertaken with the object of finding a family of New Zealand plants suitable for experimental cytological research. For this reason, attention was directed to such families as have provided elsewhere most of the recent advances in the knowledge of the structure and behaviour of chromosomes, in particular the Liliaceae. It should be pointed out, however, that this family is not well represented in New Zealand. Of the ten genera, one is monotypic and four have only one New Zealand representative; of the remainder, the largest genus, Astelia (10 species) has been found to possess a large number of small chromosomes, while the balance have but few species, some difficult of cultivation, others flowering but rarely under domestication.
In addition, the chromosomes of 20 Epilobium species were counted. This was incidental to a proposed cytogenetic study of wider scope which was abandoned on account of technical difficulties.
Material and Methods.
Preliminary observations of pollen mother-cells were made in aceto-carmine, and in this medium also chromosome numbers were established where the material did not warrant further work, e.g., in Arthropodium. Pollen mother-cell material, of Chrysobactron hookeri was fixed in 2 BE, of Clematis indivisa, Dianella intermedia and Epilobium species in Allen's Bouin, following immersion for a few seconds in Carnoy.
Wherever possible, haploid counts were verified by somatic counts in root tips, all fixed in 2 BE.
Paraffin sections were cut at 24 μ in Chrysobactron, 18 μ in Clematis, 16 μ in all other cases.
All drawings were made at bench level with the aid of a camera lucida and are X 4500, excepting the somatic plates in Chrysobactron and Clematis, which are X 3270.
Table I lists the species for which chromosome numbers were established.
[Footnote] *Now Assistant Mycologist, Plant Diseases Division, Department of Scientific and Industrial Research.
|A. ccndidum var. purpuratum||Rough Peaks, Lake Wakatipu||11||22|
|Simpson and Thomson, MS.|
|Chrysobactron hookeri||Mount Torlesse||7||14|
|Section A: Similes.|
|Epilobium erectum||Queen Charlotte Sound||18||36|
|E. pubens||Governor's Bay||18||—|
|E. junceum||Governor's Bay||18||—|
|Section B: Microphyllae.|
|E. cockaynianum||Botanic Grdns., Christchurch||18||—|
|E. pictum||Hope Valley||18||—|
|Section C: Sparsiflorae.|
|Section D: Dermatophyllae.|
|E. crassum||Lake Sumner||18||—|
|E. glabellum||Mount Torlesse||18||—|
|E. sp., affinities with E. tasmanicum||Botanic Grdns., Christchurch||18||—|
|E. sp., affinities with E. orassifolium and E. brevipes||Botanic Grdns., Christchurch||18||—|
|E. sp., affinities with E. melanocaulon||Arthurs Pass||18||—|
|A-typical, possible Hybrid Forms.|
|E. nerterioides XE. pedunculare!||Cass||18||—|
|E. linnaeoidesXE. rotundifolium!||Cass||18||—|
|E. billardierianum, a-typical||Botanic Grdns., Christchurch||18||—|
Thelymitra longifolia, which occurs also in Australia and Tasmania, has 26 somatic chromosomes, varying in length from 3·2 to 4·9 μ. Two pairs are distinguished by the presence of trabants (Fig. 1). The centromere in all is median or nearly so.
Thelymitra falls into the systematic sub-group Neottieae. For other genera of this group, Gaiser (1930) records chromosome numbers as follows:—
|Epipactis||n = 12; 2n = 24.|
|Neottia||n = 16, 18.|
|Spiranthes||n = 12.|
|Gastrodia||n = 8–9; 2n = 16–18.|
Fig. 1—Thelymitra longifolia. Somatic metaphase; 2n = 26.
Fig. 2—Dianella intermedia. Somatic metaphase; 2n = 16.
Figs. 3–4—D. intermedia. Metaphase I, anaphase I; n = 8.
Fig. 5—Arthropodium candidum. Somatic metaphase; 2n = 22.
Fig. 6—A. candidum var. purpuratum. Somatic metaphase; 2n = 22.
Fig. 7—Arthropodium cirrhatum. Somatic metaphase; 2n = 44.
Dianella intermedia. Eight chromosomes were counted in first metaphase and anaphase of meiosis (Figs. 3–4) and 16 chromosomes in mitotic metaphases (Fig. 2). The somatic complement consists of 8 pairs of M * chromosomes, with a range in length from approximately 4·5 to 9·3 μ.
Arthropodium. Aceto-carmine preparations of pollen mother-cells of A. candidum and the bronze-leaved variety of A. candidum (var. purpuratum Simpson and Thomson, MS.) showed n = 11 in
[Footnote] *M = chromosomes with median or submedian centromere.
metaphase I and II of meiosis (Figs. 8–9). A. cirrhatum was found to be relatively tetraploid with n = 22 (Fig. 10). It is a much larger plant than the former varieties, which fact agrees with their polyploid relationship. These findings were verified from somatic counts. The somatic chromosomes are too small for detailed study, but in both diploid forms there may be distinguished a larger pair of chromosomes (4·5 μ) with approximately median centromere, and a shorter pair (2·2 μ). The remainder of the complement ranges between these two extremes (Figs. 5–6). The centromeres are all median or nearly so.
The bronze-leaved variety of A. candidum has another long pair (3·6 μ). The tetraploid, A. cirrhatum, shows quite clearly 4 large chromosomes (4·5 μ), two very short pairs (1·8–2·3 μ), and with less certainty two pairs of intermediate length (3·6 μ). No quadrivalents were seen in meiosis, so that this species may be assumed to be an allo-tetraploid.
Chrysobactron hookeri. The limited amount of pollen mother-cell material available was poorly fixed, but side-views of metaphase I indicated the presence of 7 bivalents, with 2–4 chiasmata per bivalent. The chiasmata are random and unterminalised.
A study of somatic plates verified the presence of 7 pairs of chromosomes, 2n = 14 (Fig. 11). There are 5 M pairs, M1–M5, with a range in length from 12·5 to 18·5 μ; and 2 S* pairs, S1–S2, ranging from 10–11 μ in length. The S1 pair is further distinguished by the presence of a trabant; while one of the S2 chromosomes shows a lacuna, which in several cells was seen in both S2 chromosomes.
Epilobium. Representatives of all four systematic groups have n = 18 chromosomes (Figs. 13–16), and this applies also to 3 a-typical, possibly hybrid and to 3 undescribed, forms. It may be noted that the endemic division Dermatophyllae, in particular the very distinct E. crassum, conforms to the rule for the remaining divisions, and that all agree in this respect with species of the northern hemisphere: the 6 species and 4 hybrids recorded by Gaiser (1930) have also n = 18, 2n = 36 chromosomes. The small size of the chromosomes, with a range of only 1·3–2·7 μ, in root-tips (Fig. 12), did not invite a detailed study; no irregularities were observed in meiosis.
Clematis indivisa has 5 M chromosome pairs and 3 S pairs, making a somatic complement of 2n = 16 (Fig. 17). Those with median centromere cannot be paired off with certainty and vary in length from approximately 6·6 to 9 μ. In the S pairs, S1 and S2 are about 5 μ long, S3 slightly shorter, but the 3 pairs are further differentiated by a progressive decrease in the size of the short arm from S1 through S2 to S3, where the centromere is almost terminal.
C. indivisa is one of 9 endemic species. Its somatic chromosome number agrees with that established for Clematis species elsewhere by Langlet (Gaiser, 1930) and Meurman and Therman (1939). The chromosome complement of C. indivisa approximately corresponds in morphology also to that of the species studied by Meurman and Therman (1939). These authors record also tetraploid and hexaploid species.
No pollen mother-cell material was available.
I wish to acknowledge with thanks the helpful criticism and advice provided by Dr. O. H. Frankel throughout the course of this investigation.
Gaiser, L. O., 1930. Chromosome Numbers in Angiosperms, II, Bibliographic Genetica, vol. 6.
Meurman, O., and Therman, E., 1939. Studies in the Chromosome Morphology and Structural Hybridity in the Genus Clematis, Cytologia, vol. 10, 1 and 2.
Cheeseman, T. F., 1925. Manual of the New Zealand Flora, Wellington.
[Footnote] *S = chromosomes with sub-terminal centromere.