skip to content

Darwin Correspondence Project

Forms of flowers

Long and short styled form.jpg

Primula veris from Forms of flowers, p. 27
Primula veris from Forms of flowers, p. 27
Biodiversity Heritage Library

Darwin’s book The different forms of flowers on plants of the same species, published in 1877, investigated the structural differences in the sexual organs of flowers of the same species. It drew on and expanded five articles Darwin had published on the topic between 1861 and 1868. Two related questions inform all these papers: first, whether species could be defined by the fertility inter se of their offspring, and second, whether hybrid sterility was the inevitable result of crossing species. Thomas Huxley had stated in his review of Origin that a significant—if inconclusive—test of whether two individuals represented distinct ‘physiological’ species was to attempt to hybridise them, since distinct species would often either be infertile inter se or produce infertile offspring, whereas varieties of the same species would give rise to fertile progeny (T. H. Huxley 1860, pp.562-5). He later claimed, in a lecture at the Royal Institution, that Darwin’s theory, ‘falls short of being a satisfactory theory’, primarily because ‘there is as yet no proof that, by selection, modifications having the physiological character of species (i.e., whose offspring are incapable of propagation, inter se) have ever been produced from a common stock’ (T. H. Huxley 1860, p. 198). In Origin, p. 272, Darwin had contended that the sterility of interspecific hybrids, when contrasted with the fertility of crosses between different varieties, did not imply an essential distinction between species and varieties. He argued that the sterility of interspecific hybrids was not a special endowment but was gradually acquired through divergent modifications in the reproductive systems of the forms that were crossed.


‘How perplexing’

Darwin had first noticed differences in the flowers of cowslips in May 1860. ‘I have this morning been looking at my experimental Cowslips & I find some plants have all flowers with long stamens & short pistils which I will call “male plants”—others with short stamens & long pistils, which I will call “female plants”’, he told his close friend Joseph Hooker. Darwin suspected the flowers were dioecious (female and male flowers on separate plants) and also noted that their pollen differed in size and shape. He concluded by exclaiming, ‘I have now examined Primroses & find exactly same difference in size of pollen, correlated with same difference in length of style & roughness of stigma!’ At this early stage in his research, Darwin’s hypothesis was that the variations in the sexual organs were indicative of a transition from a hermaphrodite to a dioecious state.

After further study, Darwin discovered, contrary to his expectations, that ‘male plants’ produced seed capsules. He told the American botanist Asa Gray, ‘If it should prove that the so-called male plants produce less seed than the so-called females, what a beautiful case of gradation from hermaphrodite to unisexual condition it will be! If they produce about equal number of seed, how perplexing it will be.’ Darwin was surprised again in December 1860, as he reported to Hooker, ‘the other day at last I had time to weigh the seeds, & by Jove the Plants of Primrose & Cowslip with short pistils & large-grained pollen are rather more fertile than those with long pistils & small grained pollen. I find that they require the action of insects to set them, & I never will believe that these differences are without some meaning. Some of my experiments lead me to suspect that the large-grained pollen suits the long pistils & the small-grained pollen suits the short pistils. But I am determined to see, if I cannot make out the mystery next Spring.’ By March 1861, Darwin had begun to refer to the ‘dimorphous condition’ in Primula and planned to publish the results of his observations and experiments. His goal now was ‘making out the meaning of the dimorphism’. Two months later, he told Daniel Oliver, ‘I am surprised to find Cowslips utterly sterile without aid.— bears on origin of Oxlips.’ Having shown that cowslips required insect agency to effect fertilisation and set seed, Darwin became convinced that oxlips were produced among the progeny of cowslips only as a result of cross-fertilisation with primroses.

As Darwin was finishing his paper on the dimorphic forms of Primula, he was able to inform Gray , ‘I think I have made out their good or meaning clearly. The pollen of A is fitted for stigma of B & conversely.’ He was eager to hear of analogous cases, telling Gray ‘This subject interests me much, so do help me if you can; for I have some very faint hopes that it may throw some light on Hybridisation’. Darwin echoed this assertion in the conclusion to his paper (‘Dimorphic condition in Primula’, p. 94): ‘The simple fact of two individuals of the same undoubted species, when homomorphically united, being as sterile as are many distinct species when crossed, will surprise those who look at sterility as a special endowment to keep created species distinct.’ What Darwin referred to here as a ‘homomorphic union’, he would later call ‘illegitimate union’; both terms referred to the fertilisation of one form with pollen from the same form. At this late stage in his Primula research, Darwin discovered that another plant, Linum grandiflorum, had two forms and that one of these was absolutely sterile with pollen from the same form, while the other was fertile with its own pollen. He included this case  in his paper, ‘On the two forms, or dimorphic condition, in the species of Primula, and on their remarkable sexual relations’ ('Dimorphic condition in Primula’), which he read at a meeting of the Linnean Society of London on 21 November 1861.

‘Dimorphic condition in Primula’, p. 78.


‘Truly wonderful’

It was not long before Darwin began to receive remarks on his Primula paper. George Bentham confessed, ‘Your paper on hearing it read threw quite a new light on the kind of semidioicality which had been so much observed of late and which in the course of our examinations for Genera Plantarum had become quite so common a thing that precise instances no longer attract much attention and I only mentioned a few that occurred on the spur of the moment’. He also noted that botanists rarely considered such features worth highlighting in taxonomic works. He closed by asking Darwin to study the case of Viola since several species had perfect flowers (those with both male and female parts) with apparently well-developed sexual organs, but also apetalous, closed flowers with seemingly rudimentary stamens. The curious thing was that the perfect flowers, which appeared in winter and spring, never seemed to set seed, while the closed flowers, which appeared later, always produced an abundance of seed. This was a topic Darwin would return to, but for the moment, his attention turned to Linum, which had been briefly mentioned in his Primula paper.

In July 1862, Darwin explained to Gray, ‘I have lately been putting the pollen of the two forms on the division of the stigma of the same flower; & it strikes me as truly wonderful, that the stigma distinguishes the pollen; & is penetrated by the tubes of the one & not by those of the other; nor are the tubes exserted.’ A month later he told Hooker, ‘In function, but not in appearance, the pollen of these two forms, as tested by their action may be said to be generically distinct.’ As well as performing his own experiments, Darwin received information on other species of Linum from trusted correspondents. Charles Crocker, a former head of the propagation department at Kew, reported similar results with two species, one of which had not been worked on by Darwin.

Before Darwin began writing his paper on dimorphism in Linum, he reformulated his ideas about the meaning of the existence of different forms. ‘I do not at present like the term “Diœcio-dimorphism”’, he explained to Gray, ‘for I think it gives quite false notion, that the phenomena are connected with a separation of the sexes. Certainly in Primula there is unequal fertility in the two forms, & I suspect this is case with Linum; & therefore I felt bound in Primula paper to state that it might be a step towards dioicous condition; though I believe there are no dioicous forms in Primulaceæ or Linaceæ. But the three forms in Lythrum convince me that the phenomenon is in no way necessarily connected with any tendency to separation of sexes.’ After slogging through another chapter of what would eventually be published as Variation, Darwin told Hooker, ‘I have just finished a long weary chapter on simple facts of variation of cultivated Plants; & am now refreshing myself with paper on Linum for Linn. Soc.y.’ His journal records that he wrote his paper, ‘On the existence of two forms, and on their reciprocal sexual relation, in several species of the genus Linum’, between 11 and 21 December 1862. The paper was read at a meeting of the Linnean Society on 3 February 1863.

Forms of flowers, p. 97.


‘Almost stark staring mad’

Darwin next moved on to Lythrum, a genus that he had begun researching in 1862 after Hooker had supplied him with seeds. Having concluded that the existence of three forms was evidence for the phenomenon not being an intermediary stage towards separate sexes, Darwin was keen to further his research on purple loosestrife (Lythrum salicaria), a species of Lythrum he had been working on since late July 1862. He told Oliver that, ‘as each form has two sets of anthers, 18 different crosses are practicable within the limits of this one species!! … A nice job, Heaven knows whether my patience will last; but I shd. like to make out this wonderfully complex case—' In August, he excitedly told Gray, ‘I am almost stark staring mad over Lythrum; if I can prove what I fully believe it is grand case of Trimorphism with 3 different pollens & 3 stigmas’. Darwin had hoped to publish the results of the crossing experiments immediately, but by October 1862, he admitted to Hooker, ‘I am rather disgusted to find I cannot publish this year on Lythrum salicaria: I must make 126 additional crosses!!’ Nevertheless, on telling Gray about the need for this further work, Darwin added, ‘The case, I think, is worth any labour’.

One of the more onerous aspects of this labour was the counting of seeds. Darwin’s eldest son William Erasmus Darwin volunteered his help. He had been working on seed pods collected from wild plants near his home in Southampton, recording his findings in a botanical notebook (DAR 117: 50). Darwin released William from counting in November 1862, telling him, ‘Next year I shall make 136 more crosses, so I shall have counting with a vengeance’. While he waited for the next flowering season, Darwin reflected on the meaning of the three forms. He revealed his ideas to Gray, ‘the three forms in Lythrum convince me that the phenomenon is no way necessarily connected with any tendency to separation of sexes. The case seems to me in result or function to be almost identical with what old C. K. Sprengel called “dichogamy”, & which is so frequent in truly hermaphrodite groups; namely the pollen & stigma of each flower being mature at different periods’.  Christian Konrad Sprengel’s 1793 study of flower fertilisation was little regarded by botanists when it was published, but the work was notable for its statement of two related doctrines, namely, that flowers were generally adapted to be cross-fertilised, and that floral structures were often adapted for insect visitation. In his reply to Darwin, Gray unsurprisingly revealed, ‘I know nothing of Sprengel’s ”dichogamy”. Where?


‘They did not believe in my results’

In July 1863, when Lythrum was flowering, Darwin resumed his crossing experiments. He also wrote to encourage the German botanist Friedrich Hildebrand who was working on Linum, ‘I much hope that you may publish the results of these experiments; because I was told that the most eminent French Botanists in Paris said that my paper on Primula was the work of imagination, & that the case was so improbable they did not believe in my results.’ Undeterred, Darwin reported to Gray at the beginning of August, ‘I have worked Lythrum like a Trojan & have just finished 134 crosses, no slight labour; but the case seems to me worth any labour, for I declare I think it about the oddest case of reproduction ever noticed.—a triple marriage between three hermaphrodite.’ Gray replied, ‘If your Lythrum-paper shall be at all equal in interest to that on Linum it will be a gem’.

During this time, Darwin also took up Bentham’s suggestion to study the small unopened flowers of Viola. His observations contradicted the received view that the open flowers were sterile, and he informed Oliver, ‘perfect flowers of violets except V. tricolor are fertile only when visited by insects: I marked flowers visited by Bees & prevented Bees visiting others &c— The imperfect flowers are of course fertile without insect-visits'. From the end of the summer, Darwin suffered a long spell of sickness, and it was only in February 1864 that he wrote up his results. ‘The first job which I shall do is to draw up result of Lythrum crosses & on movements of climbing plants’, he informed Gray. By April he resumed counting seed capsules of Lythrum and could tell Hooker, ‘I believe in a fortnight … I shall dictate my paper on Lythrum!!!!’ Darwin was eager to finish in order to have the paper read at the Linnean Society before the summer break. ‘I have almost finished my Lythrum paper’, he told Hooker in late May, ‘I fear it can be copied & sent only just before close of Session of Linn. Soc. & that the title alone will be read’. The paper, ‘On the sexual relations of the three forms of Lythrum salicaria’, was sent to the Society on 10 June 1864 and read six days later at the final meeting of the session. Darwin had asked Oliver to copy on the Society’s blackboard the diagram of the three flower-forms of the plant, with dotted lines indicating which pollen must be applied to each stigma to produce full fertility, noting, ‘Without the diagram my paper wd. be unintelligible’.

‘Three forms of Lythrum salicaria’, p. 171.

After receiving positive feedback from Hooker, Darwin  admitted, ‘I have done nothing which has interested me so much as Lythrum since making out the Complemental males of Cirripedes. I fear that I have dragged in too much miscellaneous matter into the paper’. Some of the ‘miscellaneous matter’ would later be expanded when Darwin returned to the subject in his book, but the next aspect he focused on was the character of offspring of the various crosses. For this, he turned to his earlier crossing experiments, which included some species only briefly discussed or not mentioned at all in the three earlier papers. He also relied on the results of similar work carried out by correspondents like John Scott. Scott had been studying the character of seedlings in Primula in parallel experiments to Darwin’s; at first, it seemed that homomorphic unions only produced seedlings of the same form, but in March 1863, Darwin told Scott that with regard to cowslips, ‘homorphic seedlings from short-styled parent have presented both forms, which disgusts me’.

Scott had also informed Darwin of some unexpected results from crosses with differently coloured primroses, some of which had ‘not yielded me a single seed’. Darwin was dubious, replying, ‘Are you sure that your coloured Primroses are not descendants of coloured Polyanthus; so as to be hybrids’. On receiving seed of the different varieties from Scott, Darwin repeated the experiments in 1865, but with ‘widely different results’. Another plant to interest Darwin was Pulmonaria angustifolia, a species of lungwort also known as blue cowslip. He told Gray in October 1865 that with respect to its own pollen, the long-styled form was absolutely sterile and the short-styled almost perfectly fertile. He had also ascertained, as he told Gray, ‘that plants raised from Dimorphic species fertilized by their own pollen, are themselves generally sterile, & are often dwarfs, so that they offer the closest analogy with Hybrids; the first cross & the product both being more or less sterile; this seems to me a very curious fact’. Despite this, Darwin was unsure when he might publish these results as he was resolved on turning his attention to Variation and a fourth edition of Origin.


‘Much indirect light’

In light of his research on dimorphism and trimorphism, Darwin was able to revise his chapter on hybridity in the new edition of Origin. He encouraged Huxley to read it, noting, ‘Asa Gray & Fritz Müller (the latter especially) think that the new facts on illegitimate offspring of dimorphic plants throw much indirect light on the subject’. Müller had been collecting information on dimorphic and trimorphic plants since Darwin had asked him to observe whether Brazilian varieties of Oxalis exhibited different forms. With typical enthusiasm, Müller had reported finding several new species of dimorphic plants. Darwin told Müller that he would publish a final paper on these plants but worried, ‘I shd regret much if I prevented you from publishing on the subject’.

In March 1867, Darwin received a small book from Hildebrand on plant sexuality. Hildebrand suggested a more accurate term with which to refer to the sexual system of dimorphic and trimorphic plants: heterostyly (Hildebrand 1867a, p. 46). Even before he had read the book, Darwin replied, ‘From turning over the pages of your book I suspect that it is very like a long chapter which I have sketched out & intended to write, but which perhaps I never should have finished & certainly could not have done it nearly as well as you’. Darwin continued to make observations on the offspring of ‘illegitimate crosses’ as he now termed the offspring of unions of the same form, his work interrupted only by the need to correct proofs of Variation. By mid-November, he could tell Hooker that he would ‘soon prepare a paper or two for Linn. Soc.

 ‘I have just finished 2 papers on the fertilization of plants’, Darwin told Ernst Haeckel in February 1868. The first of these, ‘On the character and hybrid-like nature of the offspring from the illegitimate unions of dimorphic and trimorphic plants’, was read at a meeting of the Linnean Society on 20 February, while the second, ‘On the specific difference between Primula veris, Brit. Fl. (var. officinalis of Linn.), P. vulgaris, Brit. Fl. (var. acaulis, Linn.), and P. elatior, Jacq.; and on the hybrid nature of the common oxlip. With supplementary remarks on naturally produced hybrids in the genus Verbascum’, was read on 19 March. Both papers were published in the June 1868 issue of the Journal of the Linnean Society of London (Botany). In August 1868, Alfred Russel Wallace wrote to Darwin, puzzled about ‘the difficulty of how these forms, with their varying degrees of sterility, originated’, although he conceded that ‘the fact remains an excellent answer to the statement, that sterility of hybrids proves the absolute distinctness of the parents’. Fritz Müller was enthusiastic, commenting, ‘The results obtained by you as to the hybrid-like character of the illegitimate offspring of dimorphic and trimorphic plants appear to me so remarkable and important, that I have planted into my garden some of our dimorphic and trimorphic plants (Manettia bicolor and several species of Oxalis) in order to repeat your experiments’. Darwin was pleasantly surprised when a very favourable review of his work on illegitimate offspring appeared in Botanische Zeitung, and told Hildebrand, the author of the review, ‘The subject is so obscure that I did not expect that any one wd have noticed my paper, & I am accordingly very much pleased that you shd have brought the subject before the many excellent naturalists of Germany’.


I believe it is of value’

At this point, Darwin returned to what had initially been intended to be the final chapter of Variation, his study of human evolution and sexual selection. This research would eventually result in two books, The descent of man, and selection in relation to sex (1871), and The expression of the emotions in man and animals (1872). On completion of these works, Darwin returned to botanical studies and considered the possibility of republishing his earlier papers on dimorphic and trimorphic plants. In early 1875, he briefly considered adding this work to his book on ‘the good effects of crossing’ (Cross and self fertilisation), telling his German translator, Julius Carus, ‘All my papers on Dimorphic plants will be republished corrected in the Book at which I am now at work’. By the end of the year, however, he had changed his mind.

Darwin revisited this earlier work with the subtly different focus of observing ‘plants in as many natural families as possible’, explaining to Gray that he had ‘become convinced that plants of this class cannot be recognized merely by the varying lengths of pistils & stamens in a few specimens. It is necessary to compare size of pollen grains & state of stigma’. Darwin had introduced the concept of functional dimorphism in in ‘Two forms in species of Linum’ (p. 82) and clarified the meaning to Fritz Müller in a letter in September 1866, ‘What I meant in my paper on Linum about plants being dimorphic in function alone was that they shd be divided into two equal bodies functionally but not structurally different’. Darwin here had only hinted at the possibility of functional differences, but by 1875 he had completed Insectivorous plants, his first botanical work that moved from the realm of morphology into that of physiology. In his next book, The effects of cross and self fertilisation in the vegetable kingdom, Darwin considered aspects of floral physiology that affected fertility, as well as morphological characteristics.

Once Cross and self fertilisation was published in November 1876, Darwin decided to rework his papers on forms of flowers into a book. By January 1877, Darwin informed Hooker, ‘…I am only working up & adding to old matter about Dimorphic & Trimorphic plants’. He wanted his son Francis to examine at Kew yet more genera where there were indications of dimorphism, but added, ‘I will rank no plant as dimorphic without comparing pollen-grains & stigmas’. When Hermann Müller wrote to say that, in his experiments on Hottonia, ‘it is striking that the difference of fertility between legitimately and illegitimately fertilised plants is much smaller in the long-styled than in the short-styled form’, Darwin annotated this letter, wondering, ‘Would it be worth while to give Table of relative fertility of the illegitimate union of each dimorphic species’. Darwin followed through on the plan, and the published work contains several pages of tables showing the fertility of illegitimate offspring of heterostyled plants. By late March 1877 Darwin told Carus that he was working on a new section of the book not based on an earlier paper. Carus had earlier pointed out a mistake in Darwin’s terminology while translating Cross and self fertilisation, but Darwin replied, ‘I think I saw the word cleistogene used by some one else & that made me forget the true term. I found out my error some time ago, for I am now writing on cleistogamic flowers’. These were the small apetalous unopened flowers, like those of Viola, that Bentham had pointed out fifteen years earlier. Darwin had also adopted Hildebrand’s term heterostyly in preference to di- or trimorphism. Gray objected and suggested his own neologism, heterogone (diverse genitalia), with its counterpart, homogone, to which Darwin countered, ‘I have been thinking about your proposed new terms, and I cannot for very shame change again … heterostyled seems to me more definite than heterogone, as the latter would apply to di & monœcious & to polygamous plants’.

It took Darwin only five months to write Forms of flowers. He contacted his publisher John Murray in early April 1877, telling him, ‘I wish the accompanying M.S. to be published; & though I believe it is of value, it is not likely that more than a few hundred copies wd. be sold, unless indeed those who possess my other books wish to complete the series’. He seemed unsure that Murray would publish the book on his usual terms (at the expense of the publisher, with a percentage of the profits to Darwin), so asked for it to be published on commission if Murray did not want to take the risk. Murray did. When The different forms of flowers on plants of the same species went to print in early July, Darwin was informed that a thousand copies would be made, as well as stereotypes for an American edition, but in August he heard that 1250 copies had been printed. By October, only 150 copies remained unsold. There was clearly an audience with an appetite for Darwin’s botanical approach. In the space of fewer than two decades, Darwin’s research into the forms of flowers had changed the study of plant sexuality from a mere cataloguing of organ morphology into a fascinating history of functional adaptation.