The Different Forms of Flowers on Plants of the Same Species

Primula veris or the cowslip.

Differences in structure between the two forms.

Their degrees of fertility when legitimately and illegitimately united.

P. elatior, vulgaris, Sinensis, auricula, etc.

Summary on the fertility of the heterostyled species of Primula.

Homostyled species of Primula.

Hottonia palustris.

Androsace vitalliana.

(FIGURE 1.1. Primula veris.

Left: Long-styled form.

Right: Short-styled form.)

It has long been known to botanists that the common cowslip (Primula veris, Brit. Flora, var. officinalis, Lin.) exists under two forms, about equally numerous, which obviously differ from each other in the length of their pistils and stamens. (1/1. This fact, according to Von Mohl 'Botanische Zeitung' 1863 page 326, was first observed by Persoon in the year 1794.) This difference has hitherto been looked at as a case of mere variability, but this view, as we shall presently see, is far from the true one. Florists who cultivate the Polyanthus and Auricula have long been aware of the two kinds of flowers, and they call the plants which display the globular stigma at the mouth of the corolla, "pin-headed" or "pin-eyed," and those which display the anthers, "thrum-eyed." (1/2. In Johnson's Dictionary, "thrum" is said to be the ends of weavers' threads; and I suppose that some weaver who cultivated the Polyanthus invented this name, from being struck with some degree of resemblance between the cluster of anthers in the mouth of the corolla and the ends of his threads.) I will designate the two forms as the long-styled and short-styled.

The pistil in the long-styled form is almost exactly twice as long as that of the short-styled. The stigma stands in the mouth of the corolla or projects just above it, and is thus externally visible. It stands high above the anthers, which are situated halfway down the tube and cannot be easily seen. In the short-styled form the anthers are attached near the mouth of the tube, and therefore stand above the stigma, which is seated in about the middle of the tubular corolla. The corolla itself is of a different shape in the two forms; the throat or expanded portion above the attachment of the anthers being much longer in the long-styled than in the short-styled form. Village children notice this difference, as they can best make necklaces by threading and slipping the corollas of the long-styled flowers into one another. But there are much more important differences. The stigma in the long-styled form is globular; in the short-styled it is depressed on the summit, so that the longitudinal axis of the former is sometimes nearly double that of the latter. Although somewhat variable in shape, one difference is persistent, namely, in roughness: in some specimens carefully compared, the papillae which render the stigma rough were in the long- styled form from twice to thrice as long as in the short-styled. The anthers do not differ in size in the two forms, which I mention because this is the case with some heterostyled plants. The most remarkable difference is in the pollen- grains. I measured with the micrometer many specimens, both dry and wet, taken from plants growing in different situations, and always found a palpable difference. The grains distended with water from the short-styled flowers were about .038 millimetres (10 to 11/7000 of an inch) in diameter, whilst those from the long-styled were about .0254 millimetres (7/7000 of an inch), which is in the ratio of 100 to 67. The pollen-grains therefore from the longer stamens of the short-styled form are plainly larger than those from the shorter stamens of the long-styled. When examined dry, the smaller grains are seen under a low power to be more transparent than the larger grains, and apparently in a greater degree than can be accounted for by their less diameter. There is also a difference in shape, the grains from the short-styled plants being nearly spherical, those from the long-styled being oblong with the angles rounded; this difference disappears when the grains are distended with water. The long-styled plants generally tend to flower a little before the short-styled: for instance, I had twelve plants of each form growing in separate pots and treated in every respect alike; and at the time when only a single short-styled plant was in flower, seven of the long-styled had expanded their flowers.

We shall, also, presently see that the short-styled plants produce more seed than the long-styled. It is remarkable, according to Professor Oliver, that the ovules in the unexpanded and unimpregnated flowers of the latter are considerably larger than those of the short-styled flowers (1/3. 'Natural History Review' July 1862 page 237.); and this I suppose is connected with the long-styled flowers producing fewer seeds, so that the ovules have more space and nourishment for rapid development.

To sum up the differences:-The long-styled plants have a much longer pistil, with a globular and much rougher stigma, standing high above the anthers. The stamens are short; the grains of pollen smaller and oblong in shape. The upper half of the tube of the corolla is more expanded. The number of seeds produced is smaller and the ovules larger. The plants tend to flower first.

The short-styled plants have a short pistil, half the length of the tube of the corolla, with a smooth depressed stigma standing beneath the anthers. The stamens are long; the grains of pollen are spherical and larger. The tube of the corolla is of uniform diameter except close to the upper end. The number of seeds produced is larger.

I have examined a large number of flowers; and though the shape of the stigma and the length of the pistil both vary, especially in the short-styled form, I have never met with any transitional states between the two forms in plants growing in a state of nature. There is never the slightest doubt under which form a plant ought to be classed. The two kinds of flowers are never found on the same individual plant. I marked many cowslips and primroses, and on the following year all retained the same character, as did some in my garden which flowered out of their proper season in the autumn. Mr. W. Wooler, of Darlington, however, informs us that he has seen early blossoms on the Polyanthus, which were not long-styled, but became so later in the season. (1/4. I have proved by numerous experiments, hereafter to be given, that the Polyanthus is a variety of Primula veris.) Possibly in this case the pistils may not have been fully developed during the early spring. An excellent proof of the permanence of the two forms may be seen in nursery-gardens, where choice varieties of the Polyanthus are propagated by division; and I found whole beds of several varieties, each consisting exclusively of the one or the other form. The two forms exist in the wild state in about equal numbers: I collected 522 umbels from plants growing in several stations, taking a single umbel from each plant; and 241 were long-styled, and 281 short-styled. No difference in tint or size could be perceived in the two great masses of flowers.

We shall presently see that most of the species of Primula exist under two analogous forms; and it may be asked what is the meaning of the above-described important differences in their structure? The question seems well worthy of careful investigation, and I will give my observations on the cowslip in detail. The first idea which naturally occurred to me was, that this species was tending towards a dioecious condition; that the long-styled plants, with their longer pistils, rougher stigmas, and smaller pollen-grains, were more feminine in nature, and would produce more seed;-that the short-styled plants, with their shorter pistils, longer stamens and larger pollen-grains, were more masculine in nature. Accordingly, in 1860, I marked a few cowslips of both forms growing in my garden, and others growing in an open field, and others in a shady wood, and gathered and weighed the seed. In all the lots the short-styled plants yielded, contrary to my expectation, most seed. Taking the lots together, the following is the result:-

TABLE 1.1.

Column 1: Plant.

Column 2: Number of Plants.

Column 3: Number of Umbels Produced.

Column 4: Number of Capsules Produced.

Column 5: Weight of Seed In Grains.

Short-styled cowslips : 9 : 33 : 199 : 83.

Long-styled cowslips : 13 : 51 : 261 : 91.

If we compare the weight from an equal number of plants, and from an equal number of umbels, and from an equal number of capsules of the two forms, we get the following results:-

TABLE 1.2.

Column 1: Plant.

Column 2: Number of Plants.

Column 3: Weight of Seed in grains.

...

Column 4: Number of Umbels.

Column 5: Weight of Seed.

...

Column 6: Number of Capsules.

Column 7: Weight of Seed in grains.

Short-styled cowslips : 10 : 92 :: 100 : 251 :: 100 : 41.

Long-styled cowslips : 10 : 70 :: 100 : 178 :: 100 : 34.

So that, by all these standards of comparison, the short-styled form is the more fertile; if we take the number of umbels (which is the fairest standard, for large and small plants are thus equalised), the short-styled plants produce more seed than the long-styled, in the proportion of nearly four to three.

In 1861 the trial was made in a fuller and fairer manner. A number of wild plants had been transplanted during the previous autumn into a large bed in my garden, and all were treated alike; the result was:-

TABLE 1.3.

Column 1: Plant.

Column 2: Number of Plants.

Column 3: Number of Umbels.

Column 4: Weight of Seed in grains.

Short-styled cowslips : 47 : 173 : 745.

Long-styled cowslips : 58 : 208 : 692.

These figures give us the following proportions:-

TABLE 1.4.

Column 1: Plant.

Column 2: Number of Plants.

Column 3: Weight of Seed in grains.

...

Column 4: Number of Umbels.

Column 5: Weight of Seed in grains.

Short-styled cowslips : 100 : 1585 :: 100 : 430.

Long-styled cowslips : 100 : 1093 :: 100 : 332.

The season was much more favourable this year than the last; the plants also now grew in good soil, instead of in a shady wood or struggling with other plants in the open field; consequently the actual produce of seed was considerably larger. Nevertheless we have the same relative result; for the short-styled plants produced more seed than the long-styled in nearly the proportion of three to two; but if we take the fairest standard of comparison, namely, the product of seeds from an equal number of umbels, the excess is, as in the former case, nearly as four to three.

Looking to these trials made during two successive years on a large number of plants, we may safely conclude that the short-styled form is more productive than the long-styled form, and the same result holds good with some other species of Primula. Consequently my anticipation that the plants with longer pistils, rougher stigmas, shorter stamens and smaller pollen-grains, would prove to be more feminine in nature, is exactly the reverse of the truth.

In 1860 a few umbels on some plants of both the long-styled and short-styled form, which had been covered by a net, did not produce any seed, though other umbels on the same plants, artificially fertilised, produced an abundance of seed; and this fact shows that the mere covering in itself was not injurious. Accordingly, in 1861, several plants were similarly covered just before they expanded their flowers; these turned out as follows:-

TABLE 1.5.

Column 1: Plant.

Column 2: Number of Plants.

Column 3: Number of Umbels produced.

Column 4: Product of Seed.

Short-styled : 6 : 24 : 1.3 grain weight of seed, or about 50 in number.

Long-styled : 18 : 74 : Not one seed.

Judging from the exposed plants which grew all round in the same bed, and had been treated in the same manner, excepting that they had been exposed to the visits of insects, the above six short-styled plants ought to have produced 92 grains' weight of seed instead of only 1.3; and the eighteen long-styled plants, which produced not one seed, ought to have produced above 200 grains' weight. The production of a few seeds by the short-styled plants was probably due to the action of Thrips or of some other minute insect. It is scarcely necessary to give any additional evidence, but I may add that ten pots of Polyanthuses and cowslips of both forms, protected from insects in my greenhouse, did not set one pod, though artificially fertilised flowers in other pots produced an abundance. We thus see that the visits of insects are absolutely necessary for the fertilisation of Primula veris. If the corolla of the long-styled form had dropped off, instead of remaining attached in a withered state to the ovarium, the anthers attached to the lower part of the tube with some pollen still adhering to them would have been dragged over the stigma, and the flowers would have been partially self-fertilised, as is the case with Primula Sinensis through this means. It is a rather curious fact that so trifling a difference as the falling-off of the withered corolla, should make a very great difference in the number of seeds produced by a plant if its flowers are not visited by insects.

The flowers of the cowslip and of the other species of the genus secrete plenty of nectar; and I have often seen humble bees, especially B. hortorum and muscorum, sucking the former in a proper manner, though they sometimes bite holes through the corolla. (1/5. H. Muller has also seen Anthophora pilipes and a Bombylius sucking the flowers. 'Nature' December 10, 1874 page 111.) No doubt moths likewise visit the flowers, as one of my sons caught Cucullia verbasci in the act. The pollen readily adheres to any thin object which is inserted into a flower. The anthers in the one form stand nearly, but not exactly, on a level with the stigma of the other; for the distance between the anthers and stigma in the short-styled form is greater than that in the long-styled, in the ratio of 100 to 90. This difference is the result of the anthers in the long-styled form standing rather higher in the tube than does the stigma in the short-styled, and this favours their pollen being deposited on it. It follows from the position of the organs that if the proboscis of a dead humble-bee, or a thick bristle or rough needle, be pushed down the corolla, first of one form and then of the other, as an insect would do in visiting the two forms growing mingled together, pollen from the long-stamened form adheres round the base of the object, and is left with certainty on the stigma of the long-styled form; whilst pollen from the short stamens of the long-styled form adheres a little way above the extremity of the object, and some is generally left on the stigma of the other form. In accordance with this observation I found that the two kinds of pollen, which could easily be recognised under the microscope, adhered in this manner to the proboscides of the two species of humble-bees and of the moth, which were caught visiting the flowers; but some small grains were mingled with the larger grains round the base of the proboscis, and conversely some large grains with the small grains near the extremity of the proboscis. Thus pollen will be regularly carried from the one form to the other, and they will reciprocally fertilise one another. Nevertheless an insect in withdrawing its proboscis from the corolla of the long-styled form cannot fail occasionally to leave pollen from the same flower on the stigma; and in this case there might be self- fertilisation. But this will be much more likely to occur with the short-styled form; for when I inserted a bristle or other such object into the corolla of this form, and had, therefore, to pass it down between the anthers seated round the mouth of the corolla, some pollen was almost invariably carried down and left on the stigma. Minute insects, such as Thrips, which sometimes haunt the flowers, would likewise be apt to cause the self-fertilisation of both forms.

The several foregoing facts led me to try the effects of the two kinds of pollen on the stigmas of the two forms. Four essentially different unions are possible; namely, the fertilisation of the stigma of the long-styled form by its own-form pollen, and by that of the short-styled; and the stigma of the short-styled form by its own-form pollen, and by that of the long-styled. The fertilisation of either form with pollen from the other form may be conveniently called a LEGITIMATE UNION, from reasons hereafter to be made clear; and that of either form with its own-form pollen an ILLEGITIMATE UNION. I formerly applied the term "heteromorphic" to the legitimate unions, and "homomorphic" to the illegitimate unions; but after discovering the existence of trimorphic plants, in which many more unions are possible, these two terms ceased to be applicable. The illegitimate unions of both forms might have been tried in three ways; for a flower of either form may be fertilised with pollen from the same flower, or with that from a another flower on the same plant, or with that from a distinct plant of the same form. But to make my experiments perfectly fair, and to avoid any evil result from self-fertilisation or too close interbreeding, I have invariably employed pollen from a distinct plant of the same form for the illegitimate unions of all the species; and therefore it may be observed that I have used the term "own-form pollen" in speaking of such unions. The several plants in all my experiments were treated in exactly the same manner, and were carefully protected by fine nets from the access of insects, excepting Thrips, which it is impossible to exclude. I performed all the manipulations myself, and weighed the seeds in a chemical balance; but during many subsequent trials I followed the more accurate plan of counting the seeds. Some of the capsules contained no seeds, or only two or three, and these are excluded in the column headed "good capsules" in several of the following tables:-

TABLE 1.6. Primula veris.

Column 1: Nature of the Union.

Column 2: Number of Flowers fertilised.

Column 3: Number of Capsules produced.

Column 4: Number of good Capsules.

Column 5: Weight of Seed in grains.

Column 6: Calculated Weight of Seed from 100 good Capsules.

Long-styled by pollen of short-styled. Legitimate union : 22 : 15 : 14 : 8.8 : 62.

Long-styled by own-form pollen. Illegitimate union : 20 : 8 : 5 : 2.1 : 42.

Short-styled by pollen of long-styled. Legitimate union : 13 : 12 : 11 : 4.9 : 44.

Short-styled by own-form pollen. Illegitimate union : 15 : 8 : 6 : 1.8 : 30.

SUMMARY:

The two legitimate unions : 35 : 27 : 25 : 13.7 : 54.

The two illegitimate unions : 35 : 16 : 11 : 3.9 : 35.

The results may be given in another form (Table 1.7) by comparing, first, the number of capsules, whether good or bad, or of the good alone, produced by 100 flowers of both forms when legitimately and illegitimately fertilised; secondly, by comparing the weight of seed in 100 of these capsules, whether good or bad; or, thirdly, in 100 of the good capsules.

TABLE 1.7. Primula veris.

Column 1: Nature of the Union.

Column 2: Number of Flowers fertilised.

Column 3: Number of Capsules.

Column 4: Number of good Capsules.

Column 5: Weight of Seed in grains.

...

Column 6: Number of Capsules.

Column 7: Weight of Seed in grains.

...

Column 8: Number of good Capsules.

Column 9: Weight of Seed in grains.

The two legitimate unions : 100 : 77 : 71 : 39 :: 100 : 50 :: 100 : 54.

The two illegitimate unions : 100 : 45 : 31 : 11 :: 100 : 24 :: 100 : 35.

We here see that the long-styled flowers fertilised with pollen from the short- styled yield more capsules, especially good ones (i.e. containing more than one or two seeds), and that these capsules contain a greater proportional weight of seeds than do the flowers of the long-styled when fertilised with pollen from a distinct plant of the same form. So it is with the short-styled flowers, if treated in an analogous manner. Therefore I have called the former method of fertilisation a legitimate union, and the latter, as it fails to yield the full complement of capsules and seeds, an illegitimate union. These two kinds of union are graphically represented in Figure 1.2.

(FIGURE 1.2. Primula veris.

Graphic representation of two kinds of union between:

Left: Long-styled form.

Right: Short-styled form.)

If we consider the results of the two legitimate unions taken together and the two illegitimate ones, as shown in Table 1.7, we see that the former compared with the latter yielded capsules, whether containing many seeds or only a few, in the proportion of 77 to 45, or as 100 to 58. But the inferiority of the illegitimate unions is here perhaps too great, for on a subsequent occasion 100 long-styled and short-styled flowers were illegitimately fertilised, and they together yielded 53 capsules: therefore the rate of 77 to 53, or as 100 to 69, is a fairer one than that of 100 to 58. Returning to Table 1.7, if we consider only the good capsules, those from the two legitimate unions were to those from the two illegitimate in number as 71 to 31, or as 100 to 44. Again, if we take an equal number of capsules, whether good or bad, from the legitimately and illegitimately fertilised flowers, we find that the former contained seeds by weight compared with the latter as 50 to 24, or as 100 to 48; but if all the poor capsules are rejected, of which many were produced by the illegitimately fertilised flowers, the proportion is 54 to 35, or as 100 to 65. In this and all other cases, the relative fertility of the two kinds of union can, I think, be judged of more truly by the average number of seeds per capsule than by the proportion of flowers which yield capsules. The two methods might have been combined by giving the average number of seeds produced by all the flowers which were fertilised, whether they yielded capsules or not; but I have thought that it would be more instructive always to show separately the proportion of flowers which produced capsules, and the average number of apparently good seeds which the capsules contained.

Flowers legitimately fertilised set seeds under conditions which cause the almost complete failure of illegitimately fertilised flowers. Thus in the spring of 1862 forty flowers were fertilised at the same time in both ways. The plants were accidentally exposed in the greenhouse to too hot a sun, and a large number of umbels perished. Some, however, remained in moderately good health, and on these there were twelve flowers which had been fertilised legitimately, and eleven which had been fertilised illegitimately. The twelve legitimate unions yielded seven fine capsules, containing on an average each 57.3 good seeds; whilst the eleven illegitimate unions yielded only two capsules, of which one contained 39 seeds, but so poor, that I do not suppose one would have germinated, and the other contained 17 fairly good seeds.

From the facts now given the superiority of a legitimate over an illegitimate union admits of not the least doubt; and we have here a case to which no parallel exists in the vegetable or, indeed, in the animal kingdom. The individual plants of the present species, and as we shall see of several other species of Primula, are divided into two sets or bodies, which cannot be called distinct sexes, for both are hermaphrodites; yet they are to a certain extent sexually distinct, for they require reciprocal union for perfect fertility. As quadrupeds are divided into two nearly equal bodies of different sexes, so here we have two bodies, approximately equal in number, differing in their sexual powers and related to each other like males and females. There are many hermaphrodite animals which cannot fertilise themselves, but most unite with another hermaphrodite. So it is with numerous plants; for the pollen is often mature and shed, or is mechanically protruded, before the flower's own stigma is ready; and such flowers absolutely require the presence of another hermaphrodite for sexual union. But with the cowslip and various other species of Primula there is this wide difference, that one individual, though it can fertilise itself imperfectly, must unite with another individual for full fertility; it cannot, however, unite with any other individual in the same manner as an hermaphrodite plant can unite with any other one of the same species; or as one snail or earth-worm can unite with any other hermaphrodite individual. On the contrary, an individual belonging to one form of the cowslip in order to be perfectly fertile must unite with one of the other form, just as a male quadruped must and can unite only with the female.

I have spoken of the legitimate unions as being fully fertile; and I am fully justified in doing so, for flowers artificially fertilised in this manner yielded rather more seeds than plants naturally fertilised in a state of nature. The excess may be attributed to the plants having been grown separately in good soil. With respect to the illegitimate unions, we shall best appreciate their degree of lessened fertility by the following facts. Gartner estimated the sterility of the unions between distinct species, in a manner which allows of a strict comparison with the results of the legitimate and illegitimate unions of Primula. (1/6. 'Versuche uber die Bastarderzeugung' 1849 page 216.) With P. veris, for every 100 seeds yielded by the two legitimate unions, only 64 were yielded by an equal number of good capsules from the two illegitimate unions. With P. Sinensis, as we shall hereafter see, the proportion was nearly the same- -namely, as 100 to 62. Now Gartner has shown that, on the calculation of Verbascum lychnitis yielding with its own pollen 100 seeds, it yielded when fertilised by the pollen of Verbascum Phoeniceum 90 seeds; by the pollen of Verbascum nigrum, 63 seeds; by that of Verbascum blattaria, 62 seeds. So again, Dianthus barbatus fertilised by the pollen of D. superbus yielded 81 seeds, and by the pollen of D. japonicus 66 seeds, relatively to the 100 seeds produced by its own pollen. We thus see-and the fact is highly remarkable-that with Primula the illegitimate unions relatively to the legitimate are more sterile than crosses between distinct species of other genera relatively to their pure unions. Mr. Scott has given a still more striking illustration of the same fact: he crossed Primula auricula with pollen of four other species (P. palinuri, viscosa, hirsuta, and verticillata), and these hybrid unions yielded a larger average number of seeds than did P. auricula when fertilised illegitimately with its own-form pollen. (1/7. 'Journal of the Linnean Society Botany' volume 8 1864 page 93.)

The benefit which heterostyled dimorphic plants derive from the existence of the two forms is sufficiently obvious, namely, the intercrossing of distinct plants being thus ensured. (1/8. I have shown in my work on the 'Effects of Cross and Self-fertilisation' how greatly the offspring from intercrossed plants profit in height, vigour, and fertility.) Nothing can be better adapted for this end than the relative positions of the anthers and stigmas in the two forms, as shown in Figure 1.2; but to this whole subject I shall recur. No doubt pollen will occasionally be placed by insects or fall on the stigma of the same flower; and if cross-fertilisation fails, such self-fertilisation will be advantageous to the plant, as it will thus be saved from complete barrenness. But the advantage is not so great as might at first be thought, for the seedlings from illegitimate unions do not generally consist of both forms, but all belong to the parent form; they are, moreover, in some degree weakly in constitution, as will be shown in a future chapter. If, however, a flower's own pollen should first be placed by insects or fall on the stigma, it by no means follows that cross-fertilisation will be thus prevented. It is well known that if pollen from a distinct species be placed on the stigma of a plant, and some hours afterwards its own pollen be placed on it, the latter will be prepotent and will quite obliterate any effect from the foreign pollen; and there can hardly be a doubt that with heterostyled dimorphic plants, pollen from the other form will obliterate the effects of pollen from the same form, even when this has been placed on the stigma a considerable time before. To test this belief, I placed on several stigmas of a long-styled cowslip plenty of pollen from the same plant, and after twenty-four hours added some from a short-styled dark-red Polyanthus, which is a variety of the cowslip. From the flowers thus treated 30 seedlings were raised, and all these, without exception, bore reddish flowers; so that the effect of pollen from the same form, though placed on the stigmas twenty-four hours previously, was quite destroyed by that of pollen from a plant belonging to the other form.

Finally, I may remark that of the four kinds of unions, that of the short-styled illegitimately fertilised with its own-form pollen seems to be the most sterile of all, as judged by the average number of seeds, which the capsules contained. A smaller proportion, also, of these seeds than of the others germinated, and they germinated more slowly. The sterility of this union is the more remarkable, as it has already been shown that the short-styled plants yield a larger number of seeds than the long-styled, when both forms are fertilised, either naturally or artificially, in a legitimate manner.

In a future chapter, when I treat of the offspring from heterostyled dimorphic and trimorphic plants illegitimately fertilised with their own-form pollen, I shall have occasion to show that with the present species and several others, equal-styled varieties sometimes appear.

Primula elatior, Jacq.

Bardfield oxlip of English authors.

This plant, as well as the last or cowslip (P. veris, vel officinalis), and the primrose (P. vulgaris, vel acaulis) have been considered by some botanists as varieties of the same species. But they are all three undoubtedly distinct, as will be shown in the next chapter. The present species resembles to a certain extent in general appearance the common oxlip, which is a hybrid between the cowslip and primrose. Primula elatior is found in England only in two or three of the eastern counties; and I was supplied with living plants by Mr. Doubleday, who, as I believe, first called attention to its existence in England. It is common in some parts of the Continent; and H. Muller has seen several kinds of humble-bees and other bees, and Bombylius, visiting the flowers in North Germany. (1/9. 'Die Befruchtung der Blumen' page 347.)

The results of my trials on the relative fertility of the two forms, when legitimately and illegitimately fertilised, are given in Table 1.8.

TABLE 1.8. Primula elatior.

Column 1: Nature of the Union.

Column 2: Number of Flowers fertilised.

Column 3: Number of good Capsules produced.

Column 4: Maximum Number of Seeds in any one Capsule.

Column 5: Minimum Number of Seeds in any one Capsule.

Column 6: Average Number of Seeds per Capsule.

Long-styled by pollen of short-styled. Legitimate union : 10 : 6 : 62 : 34 : 46.5.

Long-styled by own-form pollen. Illegitimate union : 20 : 4 : 49* : 2 : 27.7. (*These seeds were so poor and small that they could hardly have germinated.)

Short-styled by pollen of long-styled. Legitimate union: 10 : 8 : 61 : 37 : 47.7.

Short-styled by own-form pollen. Illegitimate union : 17 : 3 : 19 : 9 : 12.1.

SUMMARY:

The two legitimate unions together : 20 : 14 : 62 : 37 : 47.1.

The two illegitimate unions together : 37 : 7 : 49* : 2 : 35.5. (*These seeds were so poor and small that they could hardly have germinated.)

If we compare the fertility of the two legitimate unions taken together with that of the two illegitimate unions together, as judged by the proportional number of flowers which when fertilised in the two methods yielded capsules, the ratio is as 100 to 27; so that by this standard the present species is much more sterile than P. veris, when both species are illegitimately fertilised. If we judge of the relative fertility of the two kinds of unions by the average number of seeds per capsule, the ratio is as 100 to 75. But this latter number is probably much too high, as many of the seeds produced by the illegitimately fertilised long-styled flowers were so small that they probably would not have germinated, and ought not to have been counted. Several long-styled and short- styled plants were protected from the access of insects, and must have been spontaneously self-fertilised. They yielded altogether only six capsules, containing any seeds; and their average number was only 7.8 per capsule. Some, moreover, of these seeds were so small that they could hardly have germinated.

Herr W. Breitenbach informs me that he examined, in two sites near the Lippe (a tributary of the Rhine), 894 flowers produced by 198 plants of this species; and he found 467 of these flowers to be long-styled, 411 short-styled, and 16 equal- styled. I have heard of no other instance with heterostyled plants of equal- styled flowers appearing in a state of nature, though far from rare with plants which have been long cultivated. It is still more remarkable that in eighteen cases the same plant produced both long-styled and short-styled, or long-styled and equal-styled flowers; and in two out of the eighteen cases, long-styled, short-styled, and equal-styled flowers. The long-styled flowers greatly preponderated on these eighteen plants,-61 consisting of this form, 15 of equal-styled, and 9 of the short-styled form.

Primula vulgaris (var. acaulis, Linn.)

The primrose of English Writers.

(FIGURE 1.3. Outlines of pollen-grains of Primula vulgaris, distended with water, much magnified and drawn under the camera lucida. The upper and smaller grains from the long-styled form; the lower and larger grains from the short- styled.)

Mr. J. Scott examined 100 plants growing near Edinburgh, and found 44 to be long-styled, and 56 short-styled; and I took by chance 79 plants in Kent, of which 39 were long-styled and 40 short-styled; so that the two lots together consisted of 83 long-styled and 96 short-styled plants. In the long-styled form the pistil is to that of the short-styled in length, from an average of five measurements, as 100 to 51. The stigma in the long-styled form is conspicuously more globose and much more papillose than in the short-styled, in which latter it is depressed on the summit; it is equally broad in the two forms. In both it stands nearly, but not exactly, on a level with the anthers of the opposite form; for it was found, from an average of 15 measurements, that the distance between the middle of the stigma and the middle of the anthers in the short- styled form is to that in the long-styled as 100 to 93. The anthers do not differ in size in the two forms. The pollen-grains from the short-styled flowers before they were soaked in water were decidedly broader, in proportion to their length, than those from the long-styled; after being soaked they were relatively to those from the long-styled as 100 to 71 in diameter, and more transparent. A large number of flowers from the two forms were compared, and 12 of the finest flowers from each lot were measured, but there was no sensible difference between them in size. Nine long-styled and eight short-styled plants growing together in a state of nature were marked, and their capsules collected after they had been naturally fertilised; and the seeds from the short-styled weighed exactly twice as much as those from an equal number of long-styled plants. So that the primrose resembles the cowslip in the short-styled plants, being the more productive of the two forms. The results of my trials on the fertility of the two forms, when legitimately and illegitimately fertilised, are given in Table 1.9.

TABLE 1.9. Primula vulgaris.

Column 1: Nature of the Union.

Column 2: Number of Flowers fertilised.

Column 3: Number of good Capsules produced.

Column 4: Maximum Number of Seeds in any one Capsule.

Column 5: Minimum Number of Seeds in any one Capsule.

Column 6: Average Number of Seeds per Capsule.

Long-styled by pollen of short-styled. Legitimate union : 12 : 11 : 77 : 47 : 66.9.

Long-styled by own-form pollen. Illegitimate union : 21 : 14 : 66 : 30 : 52.2.

Short-styled by pollen of long-styled. Legitimate union:

8 : 7 : 75 : 48 : 65.0.

Short-styled by own-form pollen. Illegitimate union : 18 : 7 : 43 : 5 : 18.8 (This average is perhaps rather too low).

SUMMARY:

The two legitimate unions together : 20 : 18 : 77 : 47 : 66.0.

The two illegitimate unions together : 39 : 21 : 66 : 5 : 35.5 (This average is perhaps rather too low).

We may infer from this table that the fertility of the two legitimate unions taken together is to that of the two illegitimate unions together, as judged by the proportional number of flowers which when fertilised in the two methods yielded capsules, as 100 to 60. If we judge by the average number of seeds per capsule produced by the two kinds of unions, the ratio is as 100 to 54; but this latter figure is perhaps rather too low. It is surprising how rarely insects can be seen during the day visiting the flowers, but I have occasionally observed small kinds of bees at work; I suppose, therefore, that they are commonly fertilised by nocturnal Lepidoptera. The long-styled plants when protected from insects yield a considerable number of capsules, and they thus differ remarkably from the same form of the cowslip, which is quite sterile under the same circumstances. Twenty-three spontaneously self-fertilised capsules from this form contained, on an average, 19.2 seeds. The short-styled plants produced fewer spontaneously self-fertilised capsules, and fourteen of them contained only 6.2 seeds per capsule. The self-fertilisation of both forms was probably aided by Thrips, which abounded within the flowers; but these minute insects could not have placed nearly sufficient pollen on the stigmas, as the spontaneously self-fertilised capsules contained much fewer seeds, on an average, than those (as may be seen in Table 1.9.) which were artificially fertilised with their own-form pollen. But this difference may perhaps be attributed in part to the flowers in the table having been fertilised with pollen from a distinct plant belonging to the same form; whilst those which were spontaneously self-fertilised no doubt generally received their own pollen. In a future part of this volume some observations will be given on the fertility of a red-coloured variety of the primrose.

Primula Sinensis.

In the long-styled form the pistil is about twice as long as that of the short- styled, and the stamens differ in a corresponding, but reversed, manner. The stigma is considerably more elongated and rougher than that of the short-styled, which is smooth and almost spherical, being somewhat depressed on the summit; but the stigma varies much in all its characters, the result, probably, of cultivation. The pollen-grains of the short-styled form, according to Hildebrand, are 7 divisions of the micrometer in length and 5 in breadth; whereas those of the long-styled are only 4 in length and 3 in breadth. (1/10. After the appearance of my paper this author published some excellent observations on the present species 'Botanische Zeitung' January 1, 1864, and he shows that I erred greatly about the size of the pollen-grains in the two forms. I suppose that by mistake I measured twice over pollen-grains from the same form.) The grains, therefore, of the short-styled are to those of the long- styled in length as 100 to 57. Hildebrand also remarked, as I had done in the case of P. veris, that the smaller grains from the long-styled are much more transparent than the larger ones from the short-styled form. We shall hereafter see that this cultivated plant varies much in its dimorphic condition and is often equal-styled. Some individuals may be said to be sub-heterostyled; thus in two white-flowered plants the pistil projected above the stamens, but in one of them it was longer and had a more elongated and rougher stigma, than in the other; and the pollen-grains from the latter were to those from the plant with a more elongated pistil only as 100 to 88 in diameter, instead of as 100 to 57. The corolla of the long-styled and short-styled form differs in shape, in the same manner as in P. veris. The long-styled plants tend to flower before the short-styled. When both forms were legitimately fertilised, the capsules from the short-styled plants contained, on an average, more seeds than those from the long-styled, in the ratio of 12.2 to 9.3 by weight, that is, as 100 to 78. In Table 1.10 we have the results of two sets of experiments tried at different periods.

TABLE 1.10. Primula Sinensis.

Column 1: Nature of the Union.

Column 2: Number of Flowers fertilised.

Column 3: Number of good Capsules produced.

Column 4: Average Weight of Seeds per Capsule.

...

Column 5: Average Number of Seeds per Capsule as ascertained on a subsequent

occasion.

Long-styled by pollen of short-styled. Legitimate union : 24 : 16 : 0.58 :: 50.

Long-styled by own-form pollen. Illegitimate union : 20 : 13 : 0.45 :: 35.

Short-styled by pollen of long-styled. Legitimate union:

8 : 8 : 0.76 :: 64.

Short-styled by own-form pollen. Illegitimate union :

7 : 4 : 0.23 :: 25.

SUMMARY:

The two legitimate unions together : 32 : 24 : 0.64 :: 57.

The two illegitimate unions together : 27 : 17 : 0.40 :: 30.

The fertility, therefore, of the two legitimate unions together to that of the two illegitimate unions, as judged by the proportional number of flowers which yielded capsules, is as 100 to 84. Judging by the average weight of seeds per capsule produced by the two kinds of unions, the ratio is as 100 to 63. On another occasion a large number of flowers of both forms were fertilised in the same manner, but no account of their number was kept. The seeds, however, were carefully counted, and the averages are shown in the right hand column. The ratio for the number of seeds produced by the two legitimate compared with the two illegitimate unions is here 100 to 53, which is probably more accurate than the foregoing one of 100 to 63.

TABLE 1.11. Primula Sinensis (from Hildebrand).

Column 1: Nature of the Union.

Column 2: Number of Flowers fertilised.

Column 3: Number of good Capsules produced.

Column 4: Average Number of Seeds per Capsule.

Long-styled by pollen of short-styled. Legitimate union : 14 : 14 : 41.

Long-styled by own-form pollen, from a distinct plant. Illegitimate union : 26 : 26 : 18.

Long-styled by pollen from same flower. Illegitimate union : 27 : 21 : 17.

Short-styled by pollen of long-styled. Legitimate union: 14 : 14 : 44.

Short-styled by own-form pollen, from a distinct plant. Illegitimate union : 16 : 16 : 20.

Short-styled by pollen from the same flower. Illegitimate union : 21 : 11 : 8.

SUMMARY:

The two legitimate unions together : 28 : 28 : 43.

The two illegitimate unions together (own-form pollen): 42 : 42 : 18.

The two illegitimate unions together (pollen from the same flower ): 48 : 32 : 13.

Hildebrand in the paper above referred to gives the results of his experiments on the present species; and these are shown in a condensed form in Table 1.11. Besides using for the illegitimate unions pollen from a distinct plant of the same form, as was always done by me, he tried, in addition, the effects of the plant's own pollen. He counted the seeds.

It is remarkable that here all the flowers which were fertilised legitimately, as well as those fertilised illegitimately with pollen from a distinct plant belonging to the same form, yielded capsules; and from this fact it might be inferred that the two forms were reciprocally much more fertile in his case than in mine. But his illegitimately fertilised capsules from both forms contained fewer seeds relatively to the legitimately fertilised capsules than in my experiments; for the ratio in his case is as 42 to 100, instead of, as in mine, as 53 to 100. Fertility is a very variable element with most plants, being determined by the conditions to which they are subjected, of which fact I have observed striking instances with the present species; and this may account for the difference between my results and those of Hildebrand. His plants were kept in a room, and perhaps were grown in too small pots or under some other unfavourable conditions, for his capsules in almost every case contained a smaller number of seeds than mine, as may be seen by comparing the right hand columns in Tables 1.10 and 1.11.

The most interesting point in Hildebrand's experiments is the difference in the effects of illegitimate fertilisation with a flower's own pollen, and with that from a distinct plant of the same form. In the latter case all the flowers produced capsules, whilst only 67 out of 100 of those fertilised with their own pollen produced capsules. The self-fertilised capsules also contained seeds, as compared with capsules from flowers fertilised with pollen from a distinct plant of the same form, in the ratio of 72 to 100.

In order to ascertain how far the present species was spontaneously self- fertile, five long-styled plants were protected by me from insects; and they bore up to a given period 147 flowers which set 62 capsules; but many of these soon fell off, showing that they had not been properly fertilised. At the same time five short-styled plants were similarly treated, and they bore 116 flowers which ultimately produced only seven capsules. On another occasion 13 protected long-styled plants yielded by weight 25.9 grains of spontaneously self- fertilised seeds. At the same time seven protected short-styled plants yielded only half-a-grain weight of seeds. Therefore the long-styled plants yielded nearly 24 times as many spontaneously self-fertilised seeds as did the same number of short-styled plants. The chief cause of this great difference appears to be, that when the corolla of a long-styled plant falls off, the anthers, from being situated near the bottom of the tube are necessarily dragged over the stigma and leave pollen on it, as I saw when I hastened the fall of nearly withered flowers; whereas in the short-styled flowers, the stamens are seated at the mouth of the corolla, and in falling off do not brush over the lowly-seated stigmas. Hildebrand likewise protected some long-styled and short-styled plants, but neither ever yielded a single capsule. He thinks that the difference in our results may be accounted for by his plants having been kept in a room and never having been shaken; but this explanation seems to me doubtful; his plants were in a less fertile condition than mine, as shown by the difference in the number of seeds produced, and it is highly probable that their lessened fertility would have interfered with especial force with their capacity for producing self- fertilised seeds.

[Primula auricula. (1/11. According to Kerner our garden auriculas are descended from P. pubescens, Jacq., which is a hybrid between the true P. auricula and hirsuta. This hybrid has now been propagated for about 300 years, and produces, when legitimately fertilised, a large number of seeds; the long-styled forms yielding an average number of 73, and the short-styled 98 seeds per capsule: see his "Geschichte der Aurikel" 'Zeitschr. des Deutschen und Oest. Alpen-Vereins' Band 6 page 52. Also 'Die Primulaceen-Bastarten' 'Oest. Botanische Zeitschrift' 1835 Numbers 3, 4 and 5.)

This species is heterostyled, like the preceding ones; but amongst the varieties distributed by florists the long-styled form is rare, as it is not valued. There is a much greater relative inequality in the length of the pistil and stamens in the two forms of the auricula than in the cowslip; the pistil in the long-styled being nearly four times as long as that in the short-styled, in which it is barely longer than the ovarium. The stigma is nearly of the same shape in both forms, but is rougher in the long-styled, though the difference is not so great as between the two forms of the cowslip. In the long-styled plants the stamens are very short, rising but little above the ovarium. The pollen-grains of these short stamens, when distended with water, were barely 5/6000 of an inch in diameter, whereas those from the long stamens of the short-styled plants were barely 7/6000, showing a relative difference of about 71 to 100. The smaller grains of the long-styled plant are also much more transparent, and before distention with water more triangular in outline than those of the other form. Mr. Scott compared ten plants of both forms growing under similar conditions, and found that, although the long-styled plant produced more umbels and more capsules than the short-styled, yet they yielded fewer seeds, in the ratio of 66 to 100. (1/12. 'Journal of the Linnean Society Botany' volume 8 1864 page 86.) Three short-styled plants were protected by me from the access of insects, and they did not produce a single seed. Mr. Scott protected six plants of both forms, and found them excessively sterile. The pistil of the long-styled form stands so high above the anthers, that it is scarcely possible that pollen should reach the stigma without some aid; and one of Mr. Scott's long-styled plants which yielded a few seeds (only 18 in number) was infested by aphides, and he does not doubt that these had imperfectly fertilised it.

I tried a few experiments by reciprocally fertilising the two forms in the same manner as before, but my plants were unhealthy, so I will give, in a condensed form, the results of Mr. Scott's experiments. For fuller particulars with respect to this and the five following species, the paper lately referred to may be consulted. In each case the fertility of the two legitimate unions, taken together, is compared with that of the two illegitimate unions together, by the same two standards as before, namely, by the proportional number of flowers which produced good capsules, and by the average number of seeds per capsule. The fertility of the legitimate unions is always taken at 100.

By the first standard, the fertility of the two legitimate unions of the auricula is to that of the two illegitimate unions as 100 to 80; and by the second standard as 100 to 15.

Primula Sikkimensis.

According to Mr. Scott, the pistil of the long-styled form is fully four times as long as that of the short-styled, but their stigmas are nearly alike in shape and roughness. The stamens do not differ so much in relative length as the pistils. The pollen-grains differ in a marked manner in the two forms; "those of the long-styled plants are sharply triquetrous, smaller, and more transparent than those of the short-styled, which are of a bluntly triangular form." The fertility of the two legitimate unions to that of the two illegitimate unions is by the first standard as 100 to 95, and by the second standard as 100 to 31.

Primula cortusoides.

The pistil of the long-styled form is about thrice as long as that of the short- styled, the stigma being double as long and covered with much longer papillae. The pollen-grains of the short-styled form are, as usual, "larger, less transparent, and more bluntly triangular than those from the long-styled plants." The fertility of the two legitimate unions to that of the two illegitimate unions is by the first standard as 100 to 74, and by the second standard as 100 to 66.

Primula involucrata.

The pistil of the long-styled form is about thrice as long as that of the short- styled; the stigma of the former is globular and closely beset with papillae, whilst that of the short-styled is smooth and depressed on the apex. The pollen- grains of the two forms differ in size and transparency as before, but not in shape. The fertility of the two legitimate to that of the two illegitimate unions is by the first standard as 100 to 72; and by the second standard as 100 to 47.

Primula farinosa.

According to Mr. Scott, the pistil of the long-styled form is only about twice as long as that of the short-styled. The stigmas of the two forms differ but little in shape. The pollen-grains differ in the usual manner in size, but not in form. The fertility of the two legitimate to that of the two illegitimate unions is by the first standard as 100 to 71, and by the second standard as 100 to 44.]

SUMMARY ON THE FOREGOING HETEROSTYLED SPECIES OF PRIMULA.

TABLE 1.12. Summary on the Fertility of the two Legitimate Unions, compared with that of the two Illegitimate Unions, in the genus Primula. The former taken at 100.

Column 1: Name of Species.

Column 2: Illegitimate Unions, Judged of by the Proportional Number of Flowers

which produced Capsules.

Column 3: Illegitimate Unions, Judged of by the Average Number (or Weight in

some cases) of Seeds per Capsule.

Primula veris : 69 : 65.

Primula elatior : 27 : 75 (Probably too high).

Primula vulgaris : 60 : 54 (Perhaps too low).

Primula Sinensis : 84 : 63.

Primula Sinensis (second trial) : ? : 53.

Primula Sinensis (after Hildebrand) : 100 : 42.

Primula auricula (Scott) : 80 : 15.

Primula Sikkimensis (Scott): 95 : 31.

Primula cortusoides (Scott): 74 : 66.

Primula involucrata (Scott): 72 : 48.

Primula farinosa (Scott): 71 : 44.

Average of the nine species : 88.4 : 61.8.

The fertility of the long-and short-styled plants of the above species of Primula, when the two forms are fertilised legitimately, and illegitimately with pollen of the same form taken from a distinct plant, has now been given. The results are seen in Table 1.12; the fertility being judged by two standards, namely, by that of the proportional number of flowers which yielded capsules, and by that of the average number of seeds per capsule. But for full accuracy many more observations, under varied conditions, would be requisite.

With plants of all kinds some flowers generally fail to produce capsules, from various accidental causes; but this source of error has been eliminated, as far as possible, in all the previous cases, by the manner in which the calculations have been made. Supposing, for instance, that 20 flowers were fertilised legitimately and yielded 18 capsules, and that 30 flowers were fertilised illegitimately and yielded 15 capsules, we may assume that on an average an equal proportion of the flowers in both lots would fail to produce capsules from various accidental causes; and the ratio of 18/20 to 15/30, or as 100 to 56 (in whole numbers), would show the proportional number of capsules due to the two methods of fertilisation; and the number 56 would appear in the left-hand column of Table 1.12, and in my other tables. With respect to the average number of seeds per capsule hardly anything need be said: supposing that the legitimately fertilised capsules contained, on an average, 50 seeds, and the illegitimately fertilised capsules 25 seeds; then as 50 is to 25 so is 100 to 50; and the latter number would appear in the right hand column.

It is impossible to look at the above table and doubt that the legitimate unions between the two forms of the above nine species of Primula are much more fertile than the illegitimate unions; although in the latter case pollen was always taken from a distinct plant of the same form. There is, however, no close correspondence in the two rows of figures, which give, according to the two standards, the difference of fertility between the legitimate and illegitimate unions. Thus all the flowers of P. Sinensis which were illegitimately fertilised by Hildebrand produced capsules; but these contained only 42 per cent of the number of seeds yielded by the legitimately fertilised capsules. So again, 95 per cent of the illegitimately fertilised flowers of P. Sikkimensis produced capsules; but these contained only 31 per cent of the number of seeds in the legitimate capsules. On the other hand, with P. elatior only 27 per cent of the illegitimately fertilised flowers yielded capsules; but these contained nearly 75 per cent of the legitimate number of seeds. It appears that the setting of the flowers, that is, the production of capsules whether good or bad, is not so much influenced by legitimate and illegitimate fertilisation as is the number of seeds which the capsules contain. For, as may be seen at the bottom of Table 1.12, 88.4 per cent of the illegitimately fertilised flowers yielded capsules; but these contained only 61.8 per cent of seeds, in comparison, in each case, with the legitimately fertilised flowers and capsules of the same species. There is another point which deserves notice, namely, the relative degree of infertility in the several species of the long-styled and short-styled flowers, when both are illegitimately fertilised. The data may be found in the earlier tables, and in those given by Mr. Scott in the Paper already referred to. If we call the number of seeds per capsule produced by the illegitimately fertilised long-styled flowers 100, the seeds from the illegitimately fertilised short- styled flowers will be represented by the following numbers (Table 1.a.):-

TABLE 1.a.

Primula veris : 71.

Primula elatior : 44 (Probably too low).

Primula vulgaris : 36 (Perhaps too low).

Primula Sinensis : 71.

Primula auricula : 119.

Primula Sikkimensis : 57.

Primula cortusoides : 93.

Primula involucrata : 74.

Primula farinosa : 63.

We thus see that, with the exception of P. auricula, the long-styled flowers of all nine species are more fertile than the short-styled flowers, when both forms are illegitimately fertilised. Whether P. auricula really differs from the other species in this respect I can form no opinion, as the result may have been accidental. The degree of self-fertility of a plant depends on two elements, namely, on the stigma receiving its own pollen and on its more or less efficient action when placed there. Now as the anthers of the short-styled flowers of several species of Primula stand directly above the stigma, their pollen is more likely to fall on it, or to be carried down to it by insects, than in the case of the long-styled form. It appears probable, therefore, at first sight, that the lessened capacity of the short-styled flowers to be fertilised with their own pollen, is a special adaptation for counteracting their greater liability to receive their own pollen, and thus for checking self-fertilisation. But from facts with respect to other species hereafter to be given, this view can hardly be admitted. In accordance with the above liability, when some of the species of Primula were allowed to fertilise themselves spontaneously under a net, all insects being excluded, except such minute ones as Thrips, the short-styled flowers, notwithstanding their greater innate self-sterility, yielded more seed than did the long-styled. None of the species, however, when insects were excluded, made a near approach to full fertility. But the long-styled form of P. Sinensis gave, under these circumstances, a considerable number of seeds, as the corolla in falling off drags the anthers, which are seated low down in the tube, over the stigma, and thus leaves plenty of pollen on it.

HOMOSTYLED SPECIES OF PRIMULA.

It has now been shown that nine of the species in this genus exist under two forms, which differ not only in structure but in function. Besides these Mr. Scott enumerates 27 other species which are heterostyled (1/13. H. Muller has given in 'Nature' December 10, 1874 page 110, a drawing of one of these species, viz. The alpine P. villosa, and shows that it is fertilised exclusively by Lepidoptera.); and to these probably others will be hereafter added. Nevertheless, some species are homostyled; that is, they exist only under a single form; but much caution is necessary on this head, as several species when cultivated are apt to become equal-styled. Mr. Scott believes that P. Scotica, verticillata, a variety of Sibirica, elata, mollis, and longiflora, are truly homostyled; and to these may be added, according to Axell, P. stricta. (1/14. Koch was aware that this species was homostyled: see "Treviranus uber Dichogamie nach Sprengel und Darwin" 'Botanische Zeitung' January 2, 1863 page 4.) Mr. Scott experimented on P. Scotica, mollis, and verticillata, and found that their flowers yielded an abundance of seeds when fertilised with their own pollen. This shows that they are not heterostyled in function. P. Scotica is, however, only moderately fertile when insects are excluded, but this depends merely on the coherent pollen not readily falling on the stigma without their aid. Mr. Scott also found that the capsules of P. verticillata contained rather more seed when the flowers were fertilised with pollen from a distinct plant than when with their own pollen; and from this fact he infers that they are sub- heterostyled in function, though not in structure. But there is no evidence that two sets of individuals exist, which differ slightly in function and are adapted for reciprocal fertilisation; and this is the essence of heterostylism. The mere fact of a plant being more fertile with pollen from a distinct individual than with its own pollen, is common to very many species, as I have shown in my work 'On the Effects of Cross and Self-fertilisation.'

Hottonia palustris.

This aquatic member of the Primulaceae is conspicuously heterostyled, as the pistil of the long-styled form projects far out of the flower, the stamens being enclosed within the tube; whilst the stamens of the short-styled flower project far outwards, the pistil being enclosed. This difference between the two forms has attracted the attention of various botanists, and that of Sprengel, in 1793, who, with his usual sagacity, adds that he does not believe the existence of the two forms to be accidental, though he cannot explain their purpose. (1/15. 'Das entdeckte Geheimniss der Nature' page 103.) The pistil of the long-styled form is more than twice as long as that of the short-styled, with the stigma rather smaller, though rougher. H. Muller gives figures of the stigmatic papillae of the two forms, and those of the long-styled are seen to be more than double the length, and much thicker than the papillae of the short-styled form. (1/16. 'Die Befruchtung' etc. page 350.) The anthers in the one form do not stand exactly on a level with the stigma in the other form; for the distance between the organs is greater in the short-styled than in the long-styled flowers in the proportion of 100 to 71. In dried specimens soaked in water the anthers of the short-styled form are larger than those of the long-styled, in the ratio of 100 to 83. The pollen-grains, also, from the short-styled flowers are conspicuously larger than those from the long-styled; the ratio between the diameters of the moistened grains being as 100 to 64, according to my measurements, but according to the measurements of H. Muller as 100 to 61; and his are probably the more accurate of the two. The contents of the larger pollen-grains appear more coarsely granular and of a browner tint, than those in the smaller grains. The two forms of Hottonia thus agree closely in most respects with those of the heterostyled species of Primula. The flowers of Hottonia are cross-fertilised, according to Muller, chiefly by Diptera.

Mr. Scott made a few trials on a short-styled plant, and found that the legitimate unions were in all ways more fertile than the illegitimate (1/17. 'Journal of the Linnean Society Botany' volume 8 1864 page 79.); but since the publication of his paper H. Muller has made much fuller experiments, and I give his results in Table 1.13, drawn up in accordance with my usual plan:-

TABLE 1.13. Hottonia palustris (from H. Muller).

Column 1: Nature of the Union.

Column 2: Number of Capsules examined.

Column 3: Average Number of Seeds per Capsule.

Long-styled by pollen of short-styled. Legitimate union: 34 : 91.4.

Long-styled by own-form pollen, from a distinct plant. Illegitimate union: 18 : 77.5.

Short-styled by pollen of long-styled. Legitimate union: 30 : 66.2.

Short-styled by own-form pollen, from a distinct plant. Illegitimate union: 19 : 18.7.

SUMMARY:

The two legitimate unions together: 64 : 78.8.

The two illegitimate unions together: 37 : 48.1.

The most remarkable point in this table is the small average number of seeds from the short-styled flowers when illegitimately fertilised, and the unusually large average number of seeds yielded by the illegitimately fertilised long- styled flowers, relatively in both cases to the product of the legitimately fertilised flowers. (1/18. H. Muller says 'Die Befruchtung' etc. page 352, that the long-styled flowers, when illegitimately fertilised, yield as many seeds as when legitimately fertilised; but by adding up the number of seeds from all the capsules produced by the two methods of fertilisation, as given by him, I arrive at the results shown in Table 1.13. The average number in the long-styled capsules, when legitimately fertilised, is 91.4, and when illegitimately fertilised, 77.5; or as 100 to 85. H. Muller agrees with me that this is the proper manner of viewing the case.) The two legitimate unions compared with the two illegitimate together yield seeds in the ratio of 100 to 61.

H. Muller also tried the effects of illegitimately fertilising the long-styled and short-styled flowers with their own pollen, instead of with that from another plant of the same form; and the results are very striking. For the capsules from the long-styled flowers thus treated contained, on an average, only 15.7 seeds instead of 77.5; and those from the short-styled 6.5, instead of 18.7 seeds per capsule. The number 6.5 agrees closely with Mr. Scott's result from the same form similarly fertilised.

From some observations by Dr. Torrey, Hottonia inflata, an inhabitant of the United States, does not appear to be heterostyled, but is remarkable from producing cleistogamic flowers, as will be seen in the last chapter of this volume.

Besides the genera Primula and Hottonia, Androsace (vel Gregoria, vel Aretia) vitalliana is heterostyled. Mr. Scott fertilised with their own pollen 21 flowers on three short-styled plants in the Edinburgh Botanic Gardens, and not one yielded a single seed; but eight of them which were fertilised with pollen from one of the other plants of the same form, set two empty capsules. (1/19. See also Treviranus in 'Botanische Zeitung' 1863 page 6 on this plant being dimorphic.) He was able to examine only dried specimens of the long-styled forms. But the evidence seems sufficient to leave hardly a doubt that Androsace is heterostyled. Fritz Muller sent me from South Brazil dried flowers of a Statice which he believed to be heterostyled. In the one form the pistil was considerably longer and the stamens slightly shorter than the corresponding organs in the other form. But as in the shorter-styled form the stigmas reached up to the anthers of the same flower, and as I could not detect in the dried specimens of the two forms any difference in their stigmas, or in the size of their pollen-grains, I dare not rank this plant as heterostyled. From statements made by Vaucher I was led to think that Soldanella alpina was heterostyled, but it is impossible that Kerner, who has closely studied this plant, could have overlooked the fact. So again from other statements it appeared probable that Pyrola might be heterostyled, but H. Muller examined for me two species in North Germany, and found this not to be the case.