My dear Sir

I am very sorry to learn from your letter of August 17^th, that your health holds so weak again; but I hope and sincerely wish, that you will soon and fully recover.¹

The dimorphic plant, of which I enclosed dried flowers in my letter of June 17^th is an Aegiphila (Verbenaceæ).² It seems to be quite sterile with own-form pollen; I have lately seen numerous bushes on an excursion to the mouth of the Itajahy;³ all those which stood far from others, bore not a single fruit, though they had flowered richly; two bushes growing close together, were loaded with fruits.—

The family of Rubiaceæ continues to furnish me with dimorphic species; to the genera formerly mentioned (Borreria, Suteria, Hedyotis, Lipostoma, Manettia, Coccocypselum)⁴ I can now add Psychotria (3 or 4 species) and Faramea (These two genera have been named for me in Kew).⁵ There are now flowering in my forest many bushes of a very beautiful ⁠⟨⁠Faramea⁠⟩⁠ which is one of the most interesting dimorphic pla⁠⟨⁠nts I ever⁠⟩⁠ saw.— The difference in the length of the styli an⁠⟨⁠d stamens of⁠⟩⁠ the two forms is very considerable; in the long-styled form the antheræ are nearly sessile.

⁠⟨⁠pressed plant specimens and excised diagrams⁠⟩⁠⁶

The two rami of the stigma are long and slender in the short-styled,—short and much broader with the papillæ more crowded together in the long-styled form. The pollen-grains of the long-styled plants are about $\frac{1}{18}$ ^mm in diameter and have a smooth surface; those of the short-styled plants are much larger ($\frac{1}{12}$ ^mm) and covered with numerous short spines. This structure of the pollen-grains will be, I suppose, a very advantageous contrivance, by preventing the pollen-grains from being blown away by the wind and by causing them to adhere more easily to the hairy body of visiting insects, which will probably touch but slightly the widely projecting anthers of the short-styled plants, while they may rub strongly their proboscis against the enclosed anthers of the long-styled. I must not forget to add, that by a torsion of the long stamens the anthers of the short-styled form turn round so as to face the circumference of the flower. Sometimes the rotation is not perfect, some or all the anthers facing only obliquely towards the circumference. There are even anthers, which do not rotate at all. I think, it may be confidently expected, that in the long-run natural selection will wholly eliminate these imperfectly rotating anthers.—

In one of your letters you told me that all known species of Oxalis were trimorphic or monomorphic; now I have hardly any doubt, that one of our species is dimorphic.⁷ It is a small creeping plant with bright yellow flowers. I have gathered numerous specimens at several widely distant places, without finding a single short-styled plant.⁸ ⁠⟨⁠diagram excised⁠⟩⁠The length of the styli in the long-styled and mid styled forms is rather variable, they are often scarcely longer or shorter than the longer stamens; there are even plants, in which the styli reach exactly to the level of the longer stamens. It may be worth mentioning, that this dimorphic plant closely resembles a monomorphic species, in which the styli equal in length the longer stamens and which fertilized ⁠⟨⁠itself as⁠⟩⁠ I found by experiment, without insect-aid.—

⁠⟨⁠attached specimens⁠⟩⁠

⁠⟨⁠    ⁠⟩⁠ raised seedlings ⁠⟨⁠    ⁠⟩⁠ ⁠⟨⁠  ⁠⟩⁠ine white tri-⁠⟨⁠morphic⁠⟩⁠ Oxalis, which offer a curious instance of embryonic ressemblance. The leaves of the seedlings have the form typical in that genus, from which form the leaves of the adult plant deviate considerably.⁹

There is another difference, not visible in the dried leaves; those of the seedlings are reddish beneath, those of the adult plant are green on both sides.—

As far as I can judge from wild plants, our several trimorphic Oxalis are quite sterile with own-form pollen or at least with the plant’s own pollen. Two of our species propagate asexually with so extraordinary rapidity, that the descendants of the same mother-plant spread over large areas. I have seen in a large field (many acres) of young sugar-cane, the whole ground covered with the red blossoms of one of these species and all the flowers were of the same form.— Thus also all the white Oxalis growing on my own land are long-styled. Of these two species I have not yet seen a single pod on wild plants.— In my garden, where I have planted close together the three forms, the white Oxalis is now seeding freely.— Of two other trimorphic species (belonging to the section Thamnoxys Endl.)¹⁰ the three forms are commonly found mingled with each other and in this case they seed well, while isolated plants appear always to be sterile.—

After reading your paper on the illegitimate offspring of dimorphic and trimorphic plants, it occurred to me, that our sterile Oxalis, of which I formerly told you, might be an illegitimate plant, which had beaten in the struggle for life and exterminated its legitimate parents in consequence of a more luxuriant asexual reproduction;¹¹ for according to Gærtner (Bastarderzeugung pg. 546) “eine der ausgezeichnetsten und allgemeinsten Eigenschaften der Pflanzen bastarde ist die Luxuriation in allen ihren Theilen”,¹² and it seems not to be improbable that some illegitimate plants should also possess this luxuriating growth of hybrids.— On this view the more or less complete contabescence of the anthers, the variability of the length of the styli in some plants and their equalling in length the shorter stamens in one of the two forms, as well as the doubleness of the flowers, which I have repeatedly observed, would be explicable.—

As to our Pontederia, I have made lately an excursion to the Itajahy-mirim,¹³ where it grows abundantly, in order to see whether it was indeed trimorphic; but unfortunately it did not yet flower.¹⁴

The Eschscholtziæ, raised from your seeds, are now beginning to flower, so that in my next letter I hope to be able to give you the result of the experiments, which I am about to try.¹⁵

With profound respect, believe me, dear Sir | very faithfully yours | Fritz Müller