to the atmosphere. As its own pistil has dried up by this time, having expanded two days before, it cannot fertilize itself. There is no evidence that it would not be just as well if it could. This precision and uniformity as to time show that there are other considerations involved in the acts connected with fertilization, besides those usually suspected. This brings us to the question of Dichogamy as an agent.

Much stress is laid on the fact that in many flowers the pistil is mature before or after the stamens, and this is interpreted as an especial arrangement for cross fertilization. I pointed out some time ago that this difference in time varied with the season in many species. But the difference is striking in some closely allied species. Barbarea praecox and B. vulgaris, two cruciferous plants, are so nearly related that the difference can scarcely be defined. The former, however, has its pistil of about equal length with the stamens-all included in the petals. The stigma certainly receives its own pollen simultaneously with the expansion of the petals. But in B. vulgaris, the pistil protrudes beyond the closed petal, and in perfect condition to be fertilized by extraneous pollen before it can be served by its own. But both species make their way equally well through the world, and I think no better illustration could be offered of the fact that a dichogamous plant has no advantage in the struggle for life. This fact may, however, be illustrated in various ways. Supposing the Iris could not self-fertilize. Its next of kin, Sisyrinchium, is certainly a self-fertilizer; and who will say that it has not made its way proudly? Iris Virginica is comparatively local, but any student can get a specimen of Sisyrinchium Bermudianum at a few hours notice. You can find flowers which seem to forbid self-fertilization, it is true; but let us not close our eyes to those so constructed as to render insect aid impossible. There are some Scrophulariacous plants which have the pistil arranged above the stamens, so as to seem placed there in order that a visiting insect may rub its pollen-covered back against it on entering; but many Pentstemons (P. grandiflorus, P. Cobaea,) incline the pistil downwards, making any such insect fertilization very difficult, yet every flower perfects seeds. Browallia (B. elata), has a hairy cap over the stamens, and an insect would only aid in self-fertilization. But Browallia is not visited by insects, yet seeds abundantly. It might be argued, this is because it

has no fragrance; but there are some garden Verbenas which have fragrance as well as color. No insect visits them on my grounds, so far as I can discover, but both kinds seed equally well.

In fact this idea that color and fragrance are necessary to attract insects are given to plants for that purpose-receives a great blow from the fact that flowers with neither are thronged with insect patrons. Many species of Rhus are illustrations. But I have taken especial pains to note Rubus occidentalis, our native black-cap raspberry. It has not the faintest trace of odor. Its small, greenishwhite petals are so inconspicuous that it might as well be apetalous. But nothing can exceed the fondness of the honey-bee for it. It abounds in my vicinity, and from sunrise till far into the twilight of evening the honey-bees crowd on it. They neglect every other flower, even white clover, as long as it lasts. Surely there should be a necessity for insect fertilization in cases where insects are so assiduous! I have had this point suggested to me. Is it not a surprise, then, that although a gauze-bag was thrown over a cluster of flowers, yet a perfect fruit resulted to every blossom—as was the case with all the neglected clover flowers as well? As to clover flowers, so great is the faith in the necessity for insect fertilization, that humble-bees have been sent from England to New Zealand to help the clover along. But since last season, I have discovered that our humble-bees do not enter the mouth of the red clover— care nothing for the elaborate arrangements for cross-fertilization— but slit the tube and get at the honey from the outside! And yet the clover seeds abundantly; and, so far as I could see, every flower in the field where I saw the bees behaving so outrageously, bore its seed. Many flowers are served in this way; and unless one looks closely, he may be deceived. In the Persian lilac, if we follow the course of our friends of the insect-fertilization school, we see the stamens arranged above the pistil, and, as the pollen bursts simultaneously with the opening of the corolla, it ought to fall on the pistil, and the entrance of an insect would only aid this self-fertilization. But, with us, it never yields a solitary seed, and we may be asked to "behold the results of self-fertilization." But we see exactly the same arrangement in the common lilac, and that seeds abundantly. In both cases the humble-bee slits the tube, and the honey bee either follows in the slits made by its stronger friends, or else makes slits for itself, a point I have been unable positively to determine.

Indeed, one of the points I wish to insist on most strongly is, that the facts in this question have been but imperfectly observed, and then erroneously construed; and of this I will offer but one more illustration. It relates to dimorphous flowers, those with the pistils long in some flowers and short in others, as in Epigaea, Mitchella, Houstonia, and others. When we look at the allies of these plants, we notice that this behaviour is exceptional. It may be assumed that they have wandered from a condition where the separate sexual organs were nearer to a perfectly hermaphrodite character, and it is assumed that this wandering is in order to derive some benefit from cross-fertilization through insect agency. I have endeavored to test this assumption in Houstonia coerulea. I selected a number of plants of both forms, and marked them when in flower. In some clusters, aggregating about fifty flowers of the short-styled plants, and which I have no doubt were self-fertilized, forty-two perfected seed; but of fifty with long styles, and which would necessarily have more difficulty of availing themselves of their own pollen, only five matured seed. Thus we see that the self-fertilizer has at least the advantage of numbers; and in a battle for life, or for any purpose at all, that is surely an advantage of no mean importance.

I believe I have shown that the facts are not wholly as they have been represented; and that even when they may exist as represented, they do not justify the deductions sought to be drawn from them.

Hasty generalizations as to the purposes of nature are dangerous. If, for instance, we examine swampy places, we find magnolias, willows, white cedars, red maples, cypresses and numerous others growing therein. We at once conclude that they grow there because they prefer the wet to the dryer land. But a wider acquaintance with these trees will show that all of them do better when, as we often find them, growing in dryer places. A suspicion then arises that there is something wrong with our reasoning, and we find at last that nature has a deeper purpose than merely an individual regard for these trees. Their seeds will only grow in wet soil; and of necessity, and not for individual benefit, have these trees to remain there. Again, I think there is nothing more certain than that effects will continue long after the causes which produced them have ceased to exist; so that actions which you see, may be associated with degradation instead of evolution-may be the last flickerings of

a dying light, and not an aurora indicating the birth of a new day. As to the present question, our reason will tell us that the phenomena we see may bear this interpretation as well as that given to them by the advocates of insect fertilization. In Europe, for instance, the common strawberry is almost universally hermaphrodite; but in this country the tendency to diœcism is well known. We know also that those parts of the world in which diœcism prevails are not as favorable to the existence of the strawberry as the others; and we may safely conclude that diœcism-a form of dimorphism—has no relation to any advantage to be derived through the sexes, but is an actual result of degrading conditions.

Then, physiologically, what good can result from cross-fertilization? It is asserted that probably most of the large order of composites are cross-fertilized, the flower in one head receiving the pollen of another flower in the same head by the aid of insects. This is contended from examination of the structure. After noting the behaviour of the parts, and in the absence of insects, I contend these plants are self-fertilizers. But admitting all that is claimed for them, compare one with an ordinary polypetalous flower, say Ranunculus-and where is the gain? The floral parts are all on the same common peduncle in both cases, and the stamens and pistils are as widely-nay, more widely-separated in a Ranunculus than in a dandelion. Practically, there is a wider separation of the sexes in the Ranunculus than in the dandelion, granting even all that or more than is asked for cross-fertilization in composites.

Physiological disturbances that aid the vital principle in the pistils, and interfere with that of the stamens, of course weaken the vital power of the pollen. In such cases, foreign pollen-pollen from flowers free from these disturbances, or where the disturbance favors the stamens instead of the pistil-would have more potency. It is, therefore, not surprising that some cases should offer, proving foreign pollen more potent than own pollen. It would be more surprising if there were none, for in every direction we find nature with overflowing abundance pushing beyond what we regard as the necessary mark. As the boy who, to jump across the stream, first goes back, and when he lands on the other side, goes farther than he wants to; so does nature in all things, or I have not read her story right.

I cannot refer in a brief paper like this to more than a few of the

observations I have made, nor do I think it necessary. I submit these propositions :

1. That cross-fertilization by insect agency does not exist nearly to the extent claimed for it.

2. Where it does exist, there is no evidence that it is of any material benefit to the race, but to the contrary.

3. Difficulties in self-fertilization result from physiological disturbances that have no relation to the general welfare of plants as species. THOMAS MEEHAN.

ΤΗ

THE POETRY OF THE PICTOR IGNOTUS.1

HE readers of Henry Crabbe Robinson's Diary will not easily forget the strange fragments of conversation he records as displaying the character and the thoughts of his artist friend, William Blake. And those who have the good fortune to own, or even to have looked into Gilchrist's Life of Blake, will carry away from that book and its weird illustrations impressions they would not gladly lose-impressions of the simple, pure-hearted, inspired child of genius, who beat about a prosaic world like a seraph that had lost its way, and then died misunderstood and slightly cared for. The pictor ignotus, as he loved to call himself, was the most un-English of Englishmen. More even than Milton, he reveals possibilities in the English character that no one had ever dreamed of. In one of his "Proverbs of Hell" he says, "The fool sees not the same tree that a wise man sees ;" and certainly he saw nothing with the same eyes that his contemporaries did. His measures of magnitude and of weight were so utterly different from theirs-so different, though perhaps in a less degree, even from ours! His piety and his seeming impieties, his ability and his limitations, were both as far as the poles from anything of the same sort to be found in England at the close of the eighteenth century. Only Hamann of Koenigsberg, or Blake's friend Flaxman, or Flaxman's master Swedenborg, comes into any sort of comparison with him. He reminds

1 THE POETICAL WORKS OF WILLIAM BLAKE, LYRICAL AND MISCELLANEOUS. Edited with a Prefatory Memoir by William Michael Rosetti. Pp. cxxxiii, 231. 12mo. Boston: Roberts Brothers.