A Cytogenetical Study of New Zealand Forms of Solanum nigrum L., S. nodiflorum Jacq. and S. gracile Otto
[Read by title and abstract before the Otago Branch on August 13, 1957.]
The form of Solanum nodiflorum collected by Banks and Solander appears to be extinct. The species is now represented by three apparently naturalized races which differ in alleles of a gene controlling distribution of purple pigment. They do not occur south of Wellington. The hexaploid S. nigrum L. is widely naturalized in both islands. S. gracile Otto is naturalized in Westland It closely resembles S. douglasii Dunal from which, however, it is genetically isolated. All species are strictly sexual and self-fertile, but probably none is exclusively self-poilinated In a test conducted within its natural climatic range S. nodiflorum produced 9.5% of outbred progeny.
The relationship of the Eurasian plant which Linnaeus named Solanum nigrum to similar forms in other countries has long been disputed. Most botanists, including Banks (Hooker, 1896, p. 227), Hooker (1864), Bentham (1869) and Cheeseman (1925), have regarded the Linnaean species as a variable one which achieved worldwide distribution without the assistance of European colonists, and have referred a wide range of material to it, including those collections made by Banks and Solander on the first occasion that Europeans landed in New Zealand and in New South Wales. However Dunal, in monographing the genus Solanum in 1852 “most extravagantly multiplied” (Bentham, 1869) the species, and recognized among many others a group of 54 termed “morellae verae” the greater number of which Bentham considered should be synonymized with S. nigrum (Bentham, 1869). In subsequent years there have been additions even to Dunal's total, suggesting much regional speciation (e.g., Polgar, 1940). All these descriptions have been ignored by most local botanists, though Cheel (1917) concluded that the true S. nigrum had not up to that time been recorded in Australia and applied the binomials S. opacum A. Br., S. pterocaulon Dunal and S. astroites Forst. to Australian and (in the case of S. astroites) to some New Zealand material; and Allan (1940) considered that S. nodiflorum Jacq. was naturalized in the Auckland-Hamilton area. Merrill (1954, p. 227) has applied this name to the plant present in New Zealand at the time of Cook's first voyage.
Recently it has become apparent that cytogenetical criteria can give greater precision to species limits in this section of the genus. The work of Jorgensen and Westergaard, still incompletely published (Westergaard, 1948), has shown that speciation involves three levels of polyploidy, and that forms of the same chromosome number may be isolated genetically by inability to cross or by the reduced vigour or fertility of their hybrids. Stebbins and Paddock (1949) have used chromosome number and degree of cross-compatibility as criteria for delimiting seven species in the Solanum nigrum complex of North America. The object of this investigation has been to consider the identity of the indigenous Solanum originally collected by Banks and Solander, and of the forms of black nightshade now widespread in this country. It is convenient at this point to anticipate the conclusions of the paper and state that the latter are considered to be S. nigrum L., S. nodiflorum Jacq. and S. gracile Otto. No attempt has been made to deal with locally established weed species that seem clearly to be of recent exotic origin.
Plants for study in cultivation at Dunedin were raised from seed or roots brought from all parts of New Zealand, and from seed obtained from Europe and America through the kindness of Professor C. A. Jorgensen, Dr. G. L. Stebbins and others. Herbarium material examined has included the nomenclatural types of S. nigrum L. and S. nodiflorum Jacq., and 55 sheets chiefly of North and Central American material kindly selected by Stebbins to illustrate his species concepts, and loaned by the Herbarium of the University of California. Specimens of most of the material studied, including the F1 hybrids raised, are preserved in the Otago University Herbarium. Essential references are given in the text in the form (O.U., etc.).
The Natural Breeding System
Nylon bags were used to protect all controlled pollinations. In these bags flowers emasculated in bud never set fruit without pollination, but produced normal fruit and seed when selfed. Normal fruit-setting also occurred when bags were placed over un-emasculated inflorescences. It thus appeared that flowers might usually be self-pollinated and self-fertilised, but it later became apparent that this is not exclusively the case:—
(1) Male-sterile hybrids S. gracile × S. douglasii which set no fruit when bagged produced occasional fruits when grown in the open, though they were more than 30 feet from plants with viable pollen. These fruits were more often grouped than solitary suggesting insect rather than wind pollination.
(2) Some S. nodiflorum hybrids heterozygous for alleles controlling pigmentation were grown in a glasshouse containing many plants homozygous for the dominant form of the gene. Seed set in unprotected flowers of these hybrids produced seedlings among which the recessive phenotype appeared significantly less often than it should have done with consistent selfing. But there were no significant departures from expected ratios when such hybrids were flowered in isolation, or when seed was taken from bagged inflorescences in which the flowers had selfed or been back-crossed.
(3) Two progeny tests designed to assess the extent of natural crossing in S. nodiflorum confirmed its occurrence. In both cases seed was taken from homozygous recessives closely interplanted with a variety carrying a dominant gene for purple pigmentation. Seed set in a garden at Dunedin, which is outside the natural range of the species, produced only three heterozygous purples in a population of 560 (0.54%) but that set in an Auckland garden yielded 40 purples in a population of 418 (9.5%).
Bumble-bees sometimes visit the flowers of S. gracile and S. douglasii but make only brief impatient passes at the smaller flowers of the other species, in which, however, lesser insects are not infrequently seen. Self-incompatibility appears to be characteristic of the tuberous species of Solanum, but predominant selfing has been recorded in the non-tuberous S. melongena (Fryxell, 1957).
All F1 plants raised in this work have been more vigorous than their parents, usually exceeding their height in pot culture by 25–50%. This applies both to interspecific hybrids which were substantially sterile and to fully fertile hybrids between parents differing little or not at all in appearance. But the crosses have always been between material of different geographic origin—no comparison has been made of the growth rate of inbred and outbred progeny from within a single natural population.
All chromosome counts have been made at microsporogenesis by squashing anthers in aceto-carmine. Only two numbers have been found in New Zealand material—namely, diploid—n = 12—and hexaploid—n = 36. The existence of tetraploids—n = 24—in this section of the genus is, however, confirmed in respect of a plant with orange-red fruits received from Portugal under the name S. nigrum var. miniatum.
Identity of the Hexaploid
Plants of hexaploid number were obtained from most parts of New Zealand. They are common on cultivated and waste land among exotic weeds, and also along tracks and in open places where the vegetation, though not undamaged, is essentially indigenous. The two collections selected for genetical work both came from the second type of habitat—namely, from a forest margin at Karori, Wellington (O.U. 4027) and from soil disturbed by nesting gulls in the Three Kings Islands (O.U. 4018). Both closely resembled plants raised from seed “e plantis spontaneis in loco natali” ex Portugal (O.U. 4028), and were readily crossed reciprocally with them. The F1 and F2 were vigorous and highly fertile, and the latter showed no segregation. The New Zealand hexaploid is therefore regarded as identical with that of Europe, a view endorsed by Jorgensen, to whom seed was sent. A hexaploid of the same appearance has been collected as an agricultural weed both at Sydney (O.U. 4012) and in North West Victoria (O.U. 4011, 4015).
The key characters of this species are tabled below (Table I). Stebbins and Paddock state that it occurs only sparingly in America as an introduction from Eurasia and Africa, and they assign the binomial S. nigrum L. to it. An examination of the Linnean Herbarium has confirmed this choice of name. Though the specimens are meagre, the large size of the pollen on the only localized sheet (248.18, bearing the sign for Central Asia) removes all reasonable doubt of its hexaploid nature. According to measurements kindly made at my request by Dr. C. R. Metcalfe, most grains expand when mounted in lacto-phenol and cotton blue to a diameter of 32μ, and one grain 44μ × 40μ was recorded.
Identity of Diploids
Diploids are plentiful in the North Island and in Westland, but there are differences in their appearance which are essentially regional. Diploids from coastal islands of the Auckland Province constantly have fresh green foliage and pure white flowers On agricultural land on the mainland, where they are commonly associated with hexaploids, their stems, leaves and, usually, their flowers are faintly purplish, or become so under appropriate conditions. Westland specimens are distinctive in flower and fruit (S. gracile, Table I).
North Island Specimens
Plants that differ only in pigmentation cross readily, are highly fertile in the F1, and segregate only for colour in the F2, the parental types emerging in 3:1 ratio. The extent to which purple pigment develops (other than in the ripe fruit) is thus governed fundamentally by alleles of a single gene. These are three in number—(a) full-purple enabling this colour to appear in stem, leaf and flower, (b) half-purple with which the flowers are not tinted even on wilting, and (c) non-purple with which purple pigment is confined to the nodes and stem-base. But purpling does not develop constantly to the limit of the plant's capacity, and its absence is only conclusive after some check to growth such as transplanting or cold weather. In these circumstances a full- or half-purple gene expresses itself in the colour of the leaf veins and sometimes of the lamina as well. Homozygotes are not as a class distinct from heterozygotes in depth of colour though the former provide the darkest and the latter the palest of the pigmented plants. Minor colour variations of this type seem unsuited for formal recognition in taxonomy, since no evidence of them, other than the collector's notes, can be preserved on a herbarium sheet.
These North Island specimens agree closely with the type of Jacquin's Solanum nodiflorum which is preserved at the British Museum, and with the coloured plate apparently prepared from it (Jacquin, 1786–93). The mean pollen grain diameter of the type is 23μ (in blue lacto-phenol) which is within the range found in diploids and the plate shows fresh green foliage and pure white flowers. They have also been.
determined as S. nodiflorum by Jorgensen, to whom seed was sent, and they conform with the interpretation which Stebbins and Paddock place upon that species.
The identity of the New Zealand species with that of North America has been confirmed by crosses with material collected by Stebbins on the campus of the University of California (O.U. 3898, 3899). This seed produced both half-purple and full-purple plants, but none that were constantly a fresh green. Both forms were fully fertile in crosses with the non-purple North Auckland coastal plant (O.U. 3417, 3418), and in the F2 segregated only to give a 3:1 ratio for colour. A corresponding result was obtained when the half-purple American form was crossed with a full-purple garden weed from Auckland (O.U. 4021).
Judging by the limited herbarium material available (e.g., Bot. Div. 81060, 81205, 81612), the principal Solanaceous weed in Westland is unlike those found elsewhere in New Zealand. Roots received from Greymouth developed into plants (O.U. 4019, 4020) which exactly matched those raised from seed received from Professor Jorgensen under the name S. gracile Otto (O.U. 4022). Furthermore the two stocks crossed readily and a uniform and fertile F2 was raised. This species is not mentioned by Stebbins and Paddock, but it appeared to differ only in style length from certain Californian sheets determined by Stebbins as S. douglasii Dunal. However, attempts made to cross S. gracile with a stock of S. douglasii received indirectly from Dr. Paddock (O.U. 3895) consistently failed when the longer-styled S. douglasii was the female parent though the reciprocal hybrid was readily obtained. It was, however, self-sterile, 90% of the pollen having a collapsed appearance, but seed could be obtained by back-crossing to either parent. Later a second collection of S. douglasii was obtained direct from Santa Barbara (O.U. 4024), and this too failed to set fruit with pollen of S. gracile.
The morphological differences which separate S. gracile from S. nodiflorum are comparatively striking, but the two were crossed reciprocally without difficulty and only about 65% of the pollen of the F1 appeared defective. The seed set on open pollination produced a very heterogeneous population, including many plants that died young or were conspicuously abnormal.
The degree of genetical isolation thus demonstrated between S. gracile and S. douglasii and between S. gracile and S. nodiflorum is comparable with that which Stebbins and Paddock have recorded between S. douglasii, S. nodiflorum and S. americanum. It is therefore logical to maintain the species, but the name itself appears to need correction. Mr. Wm. T. Stearn, of the British Museum, has pointed out in correspondence that Otto's name was first validly published by Dunal (1852), by which time it had been invalidated by S. gracile Sendtner (1846). The task of locating the first valid name of this plant requires the attention of someone with ready access to the older herbaria. The ultimate source of Jorgensen's seed is unfortunately unknown, but according to Dunal the species is found in Central and Southern America. Since direct traffic once occurred between Californian and Westland gold-fields it might well be that California is the source of the New Zealand plant, and that it is included there in the range of S. douglasii.
The Indigenous Form of S. nodiflorum
Banks (Hooker, 1896) considered that the “garden nightshade” which he collected in New Zealand was “exactly the same as in England”, but his opinion is not confirmed by the herbarium material (there are well-filled sheets in the Auckland, Wellington and British Museums), or by the unpublished plate based on Parkinson's sketch (Cook's First Voyage, 1768–1771—Plants of New Zealand, Vol. 3, t. 133; British Museum—a photograph of this plate is deposited in the Otago University Herbarium). The mean pollen-grain diameter (21μ in blue lacto-phenol) is that of a diploid, and the general vegetative aspect is very like the Auckland coastal form of
The scale in Fig. 1 applies to Figs. 1, 2 and 3. 1—Ripe fruits of S. nodiflorum. 2—Ripe fruits of S. gracile. 3—Ripe fruits of S. nigrum. 4—S. nigrum pollen mother cell second metaphase × 2100. 5—S. nodiflorum pollen mother cell second metaphase × 2100. 6—Contents of single fruits: A, S. nigrum; B, S. douglasii; C. S. nodiflorum. 7—Flowers: A, S. nodiflorum; B. S. nigrum; C, S. gracile.
S. nodiflorum. The small anthers, which do not exceed 1.5 mm in length, also justify its transfer to that species. But flexibility in the branching of the inflorescences is apparent in every piece, and this cannot be matched in local collections today. An umbellate or sub-umbellate inflorescence now predominates, and racemose arrangement of the flowers, or forking of the inflorescence axis are only rare abnormalities. The calyx of the immature fruit also appears to be larger in the Banks and Solander material than is usual. Solander's unpublished Primitiae Florae Novae-Zealandiae records this plant in six localities ranging from the Bay of Islands (Motuaro) to Queen Charlotte Sound (Totaranui), but the sheets are not themselves localized. However, the possibility that all the material came from one aberrant plant seems to be disposed of by the presence of these forms of inflorescence in some of the earlier collections made by local botanists—particularly material from Mt. Wellington lava fields obtained by Petrie in 1895 (Dom. Mus. 3401), an otherwise unlabelled sheet from Mokihinui I. (Dom. Mus. 3403) and an undated Auckland specimen collected by Kirk (Ak. Mus. 11607). It would appear, therefore, that the form of S. nodiflorum that was widespread in pre-European times has disappeared.
The fresh green form of S. nodiflorum is probably common today in the Pacific. Seed of it was received from Fiji (O.U. 4026), and its indigenous status in New South Wales is established by the Banks and Solander sheet collected in 1770 (British Museum), which can still be matched with plants growing on the shores of Sydney Harbour among other indigenous species (O. U. 3900. 3901, 4025). It predominates at present on the offshore islands of North Auckland, where a native herb of rock crevices and disturbed soil would be likely to persist, but since it is accompanied there by naturalized S. nigrum L. this is no proof that it is indigenous, and really early collections of it from New Zealand appear to be lacking. That genes for purple pigmentation are of recent introduction seems certain, for they are found at present almost exclusively in weeds of cultivation. But a full-purple plant was collected in the Three Kings Islands in 1955, and it may be that the fresh-green colour of the double-recessive will in future years disappear as the flexible pattern of the inflorescence seems earlier to have done.
Morphological Characters of Taxonomic Value
The vegetative growth of S. nodiflorum, S. nigrum, S. gracile and S. douglasii is too variable to provide convenient key characters for separation of the species. All flower within a few months of germination, but are perennial under favourable circumstances. Habit, leaf-size and form, and the extent to which stems are winged are so much affected by age and growing conditions that it is difficult to define limits within which they are genetically fixed. Though S. gracile and S. douglasii are normally finely tomentose and the other two species nearly glabrous even this difference is sometimes obscure. However, most features of flower and fruit are little altered by age or environment, so that they provide reliable though not always conspicuous means of discriminating between forms sufficiently isolated genetically to warrant their recognition as species. Though genetical criteria are not universally acceptable in determining what taxa are appropriate they should be given critical importance in this section of Solanum, since in it reproduction is strictly sexual with some degree of out-breeding, and the short life cycle is highly favourable to experiment.
Key characters of the four species dealt with are apparent from Table I, and some of them are illustrated in Plate 25.
Herbarium sheets in the principal New Zealand herbaria have been annotated, and there seems little point in citing those that have no special interest. The only additional type specimens that have been examined are the series on which S. allanii Polgar is based (Bot. Div. 19111, 19112, 19115, 19117, 19118) and the type of S. pachystylum Polgar (Bot. Div. 8954). Both should be synonymized with.
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
|S. nigrum||S. nodiflorum||S. douglasii||S. gracile|
|1 Present distribution in New Zealand||Three Kings Is. (O.U. 4018) to Otago (O.U. 4014)||Kermadec Is. (Ak. Mus. 7597) to Wellington (O.U. 3897)||Not present||Westland|
|Earliest collection||1895, Dom. Mus. 3400, Bay of Is., D. Petrie||1769, Banks and Solander||—||1941, Botany Division, 33276, Greymouth, 26/2/41, H. H. Allan|
|2 Pollen-diam. (μ) in lactophenol||27–38||20–27|
|Inflorescence (typically)||Racemose||3 Umbellate||Racemose|
|Flower diameter (mm)||c.10||c.8||c.15||c.15|
|4 Anther length (mm)||2.0–2.7||1.1–2.0||2.5–4.0|
|Style protrusion beyond anthers||V. slight||V. slight||1.5–2.5 mm||0.5–1.0 mm|
|Seed lengh (mm)||1.7–2.0||1.2–1.5|
|Fruiting peduncles||Spreading or ascending||Deflexed|
|Fruit calyx||Moderately reflexed||Strongly reflexed||Appiled or mod. reflexed||Applied membraneous|
|Fruit surface||Mod. gloss||5 High gloss||Semi-matt||Matt|
|6 Stone cell masses||None||0–4|
[Footnote] 1 The great paucity of herbarium material from the South Island, if it reflects lack of interest, makes it probable that S. gracile was established in Westland long before the earliest collection was made. On the other hand the existence (mainly in the Dominion Museum Herbarium) of about ten sheets of S. nodiflorum that actually or apparently pre-date Petrie's collection of S. nigrum indicates that in the North Island the latter species would have been picked up earlier had it been present.
[Footnote] 2 Pollen diameter has the same value in lacto-phenol for both fresh and dried material, but fresh grains swell considerably in water. The size of the pollen is normal in unopened flowers and in flowers produced during drought.
[Footnote] 3 The umbels commonly have a flower inserted shortly below them.
[Footnote] 4 Anther length is the most convenient reliable character for separation of the three common local species and flower size is usually proportional to it.
[Footnote] 5 The fruit surface loses its gloss as the fruits become over-ripe.
[Footnote] 6 Polgar (1940) sought to make the number of stone-cell masses in the fruit pulp a critical character, but it seems always inconstant, and small ones are easily overlooked unless the pulp is dried.
S. nodiflorum Jacq.—the former because it was undoubtedly based on the full-purple form of S. nodiflorum which is abundant in the type locality (Mt. Wellington lava fields, Auckland), the latter because it has been collected only once and there is no proof that it is other than a non-perpetuating variant.
On the basis of Cheel's illustrations (1917) and the recent collections from Sydney already cited, it seems clear that the plants he referred to S. pterocaulon were S. nodiflorum, while those placed under S. opacum were S. nigrum. As he noted, the former bears few fruits per umbel in this locality, commonly three—but this character is not genetically fixed since seed from Sydney, when sown at Dunedin, produced plants (O.U. 3900B) in no way different from the North Auckland coastal form.
I am indebted to the Directors of the Herbaria whose collections have been named for access to their material, to Mr. Wm. T. Stearn, of the British Museum, for help in consulting the herbarium and library of that Institution, to Mr. N. Sandwith and Dr. C. R. Metcalfe for assistance in connection with the Linnean Herbarium and to the Director of Coimbra Botanic Garden, Professor C. A. Jorgensen, Dr. J. L. Stebbins, Mr. T. Rawson, Mr. L. O. Simpson, Mrs. O. D. Suggate, Mr. B. I. Parham, Mr. Peter Raven, Mr. Robt. Ornduff and Mr. Alan Mark for special collections of roots or seed. The work was assisted by a grant towards travelling expenses from the Research Fund of the University of New Zealand.
Allan, H. H., 1940. Handbook of the Naturalized Flora of New Zealand, p. 197.
Bentham, G., 1869. Flora australiensis, 4, p. 443.
Cheel, E., 1917. Notes on the “common nightshade” (Solanum nigrum Linn) and some closely related forms and species that have been confused with it, Proc. Linn. Soc. N.S.W. 42, 583–602.
Cheeseman, T. F., 1925. Manual of the New Zealand Flora, p. 768.
Dunal, M. F., 1952. Solanaceae, D. C. Prodromus, 13, pp. 45–59.
Fryxell, P. A., 1957. Mode of Reproduction in Higher Plants, Bot. Rev., 23, p. 206.
Hooker, J. D., 1864. Handbook of the New Zealand Flora, p. 200.
—— 1896. Journal of the Right Hon. Sir Joseph Banks, p. 227.
Jacquin, N., 1786–93. Icones Plantarum Rariorum, 2, t. 326.
Polgar, S., 1940. Appendix to Notes on New Zealand Floristic Botany, No. 7, by H. H. Allan, Trans. Roy. Soc., N. Z., 69, 278–281.
Westergaard, M., 1948. The aspects of polyploidy in the genus Solanum III: Seed production in autopolyploid and allopolyploid Solanums, Dansk. Videns Selskab Biol. Meddel, 18, 3, 3–18.
Professor G. T. S. Baylis, Botany Department, University of Otago.