In the very first place, I declare that this is no scientific chapter. It is addressed to the thousands of men and women in the realm who tend a little group of orchids lovingly, and mark the wonders of their structure with as much bewilderment as interest. They read of hybridization, they see the result in costly specimens, they get books, they study papers on the subject. But the deeper their research commonly, the more they become convinced that these mysteries lie beyond their attainment. I am not aware of any treatise which makes a serious effort to teach the uninitiated. Putting technical expressions on one side—though that obstacle is grave enough—every one of those which have come under my notice takes the mechanical preliminaries for granted. All are written by experts for experts. My purpose is contrary. I wish to show how it is done so clearly that a child or the dullest gardener may be able to perform the operations—so very easy when you know how to set to work.

After a single lesson, in the genus Cypripedium alone, a young lady of my household amused herself by concerting the most incredible alliances—Dendrobium with Odontoglossum, Epidendrum with Oncidium, Oncidium with Odontoglossum, and so forth. It is unnecessary to tell the experienced that in every case the seed vessel swelled; that matter will be referred to presently. I mention the incident only to show how simple are these processes if the key be grasped.

Amateur hybridizers of an audacious class are wanted because, hitherto, operators have kept so much to the beaten paths. The names of Veitch and Dominy and Seden will endure when those of great savants are forgotten; but business men have been obliged to concentrate their zeal upon experiments that pay. Fantastic crosses mean, in all probability, a waste of time, space, and labour; in fact, it is not until recent years that such attempts could be regarded as serious. So much the more creditable, therefore, are Messrs. Veitch's exertions in that line.

But it seems likely to me that when hybridizing becomes a common pursuit with those who grow orchids—and the time approaches fast—a very strange revolution may follow. It will appear, as I think, that the enormous list of pure species—even genera—recognized at this date may be thinned in a surprising fashion. I believe—timidly, as becomes the unscientific—that many distinctions which anatomy recognizes at present as essential to a true species will be proved, in the future, to result from promiscuous hybridization through ?ons of time. "Proved," perhaps, is the word too strong, since human life is short; but such a mass of evidence will be collected that reasonable men can entertain no doubt. Of course the species will be retained, but we shall know it to be a hybrid—the offspring, perhaps, of hybrids innumerable.

I incline more and more to think that even genera may be disturbed in a surprising fashion, and I know that some great authorities agree with me outright, though they are unprepared to commit themselves at present. A very few years ago this suggestion would have been absurd, in the sense that it wanted facts in support. As our ancestors made it an article of faith that to fertilize an orchid was impossible for man, so we imagined until lately that genera would not mingle. But this belief grows unsteady. Though bi-generic crosses have not been much favoured, as offering little prospect of success, such results have been obtained already that the field of speculation lies open to irresponsible persons like myself. When Cattleya has been allied with Sophronitis, Sophronitis with Epidendrum, Odontoglossum with Zygopetalum, C?logene with Calanthe, one may credit almost anything. What should be stated on the other side will appear presently.

How many hybrids have we now, established, and passing from hand to hand as freely as natural species? There is no convenient record; but in the trade list of a French dealer those he is prepared to supply are set apart with Gallic precision. They number 416; but imagination and commercial enterprise are not less characteristic of the Gaul than precision.

In the excellent "Manual" of Messrs. Veitch, which has supplied me with a mass of details, I find ten hybrid Calanthes; thirteen hybrid Cattleyas, and fifteen L?lias, besides sixteen "natural hybrids"—species thus classed upon internal evidence—and the wondrous Sophro-Cattleya, bi-generic; fourteen Dendrobiums and one natural; eighty-seven Cypripediums—but as for the number in existence, it is so great, and it increases so fast, that Messrs. Veitch have lost count; Phajus one, but several from alliance with Calanthe; Chysis two; Epidendrum one; Miltonia one, and two natural; Masdevallia ten, and two natural; and so on. And it must be borne in mind that these amazing results have been effected in one generation. Dean Herbert's achievements eighty years ago were not chronicled, and it is certain that none of the results survive. Mr. Sander of St. Albans preserves an interesting relic, the only one as yet connected with the science of orchidology. This is Cattleya hybrida, the first of that genus raised by Dominy, manager to Messrs. Veitch, at the suggestion of Mr. Harris of Exeter, to the stupefaction of our grandfathers. Mr. Harris will ever be remembered as the gentleman who showed Mr. Veitch's agent how orchids are fertilized, and started him on his career. This plant was lost for years, but Mr. Sander found it by chance in the collection of Dr. Janisch at Hamburg, and he keeps it as a curiosity, for in itself the object has no value. But this is a digression.

Dominy's earliest success, actually the very first of garden hybrids to flower—in 1856—was Calanthe Dominii, offspring of C. Masuca × C. furcata;—be it here remarked that the name of the mother, or seed parent, always stands first. Another interest attaches to C. Dominii. Both its parents belong to the Veratr?folia section of Calanthe, the terrestrial species, and no other hybrid has yet been raised among them. We have here one of the numberless mysteries disclosed by hybridization. The epiphytal Calanthes, represented by C. vestita, will not cross with the terrestrial, represented by C. veratr?folia, nor will the mules of either. We may "give this up" and proceed. In 1859 flowered C. Veitchii, from C. rosea, still called, as a rule, Limatodes rosea, × C. vestita. No orchid is so common as this, and none more simply beautiful. But although the success was so striking, and the way to it so easy, twenty years passed before even Messrs. Veitch raised another hybrid Calanthe. In 1878 Seden flowered C. Sedeni from C. Veitchii × C. vestita. Others entered the field then, especially Sir Trevor Lawrence, Mr. Cookson, and Mr. Charles Winn. But the genus is small, and they mostly chose the same families, often giving new names to the progeny, in ignorance of each other's labour.

The mystery I have alluded to recurs again and again. Large groups of species refuse to inter-marry with their nearest kindred, even plants which seem identical in the botanist's point of view. There is good ground for hoping, however, that longer and broader experience will annihilate some at least of the axioms current in this matter. Thus, it is repeated and published in the very latest editions of standard works that South American Cattleyas, which will breed, not only among themselves, but also with the Brazilian L?lias, decline an alliance with their Mexican kindred. But Baron Schroeder possesses a hybrid of such typical parentage as Catt. citrina, Mexican, and Catt. intermedia, Brazilian. It was raised by Miss Harris, of Lamberhurst, Kent, one single plant only; and it has flowered several times. Messrs. Sander have crossed Catt. guttata Leopoldii, Brazil, with Catt. Dowiana, Costa Rica, giving Catt. Chamberliana; L?lia crispa, Brazil, with the same, giving L?lio-Cattleya Pallas; Catt. citrina, Mexico, with Catt. intermedia, Brazil, giving Catt. citrina intermedia (Lamberhurst hybrid); L?lia flava, Brazil, with Catt. Skinneri, Costa Rica, giving L?lio-Catt. Marriottiana; L?lia pumila, Brazil, with Catt. Dowiana, Costa Rica, giving L?lio-Catt. Normanii; L?lia Digbyana, Central America, with Catt. Mossi?, Venezuela, giving L?lio-Catt. Digbyana-Mossi?; Catt. Mossi?, Venezuela, with L?lia cinnabarina, Brazil, giving L?lio-Catt. Phoebe. Not yet flowered and unnamed, raised in the Nursery, are Catt. citrina, Mexico, with L?lia purpurata, Brazil; Catt. Harrisoni?, Brazil, with Catt. citrina, Mexico; L?lia anceps, Mexico, with Epidendrum ciliare, U.S. Colombia. In other genera there are several hybrids of Mexican and South American parentage; as L. anceps × Epid. ciliare, Sophronitis grandiflora × Epid. radicans, Epid. xanthinum × Epid. radicans.

But among Cypripediums, the easiest and safest of all orchids to hybridize, East Indian and American species are unfruitful. Messrs. Veitch obtained such a cross, as they had every reason to believe, in one instance. For sixteen years the plants grew and grew until it was thought they would prove the rule by declining to flower. I wrote to Messrs. Veitch to obtain the latest news. They inform me that one has bloomed at last. It shows no trace of the American strain, and they have satisfied themselves that there was an error in the operation or the record. Again, the capsules secured from very many by-generic crosses have proved, time after time, to contain not a single seed. In other cases the seed was excellent to all appearance, but it has resolutely refused to germinate. And further, certain by-generic seedlings have utterly ignored one parent. Zygopetalum Mackayi has been crossed by Mr. Veitch, Mr. Cookson, and others doubtless, with various Odontoglossums, but the flower has always turned out Zygopetalum Mackayi pure and simple—which becomes the more unaccountable more one thinks of it.

Hybrids partake of the nature of both parents, but they incline generally, as in the extreme cases mentioned, to resemble one much more strongly than the other. When a Cattleya or L?lia of the single-leaf section is crossed with one of the two-leaf, some of the offspring, from the same capsule, show two leaves, others one only; and some show one and two alternately, obeying no rule perceptible to us at present. So it is with the charming L?lia Maynardii from L. Dayana × Cattleya dolosa, just raised by Mr. Sander and named after the Superintendent of his hybridizing operations. Catt. dolosa has two leaves, L. Dayana one; the product has two and one alternately. Sepals and petals are alike in colour, rosy crimson, veined with a deeper hue; lip brightest crimson-lake, long, broad and flat, curving in handsomely above the column, which is closely depressed after the manner of Catt. dolosa.

The first bi-generic cross deserves a paragraph to itself if only on that account; but its own merits are more than sufficient. Sophro-Cattleya Batemaniana was raised by Messrs. Veitch from Sophronitis grandiflora × Catt. intermedia. It flowered in August, 1886; petals and sepals rosy scarlet, lip pale lilac bordered with amethyst and tipped with rosy purple.

But one natural hybrid has been identified among Dendrobes—the progeny doubtless of D. crassinode × D. Wardianum. Messrs. J. Laing have a fine specimen of this; it shows the growth of the latter species with the bloom of the former, but enlarged and improved. Several other hybrid crosses are suspected. Of artificial we have not less than fifty.

Phaius—it is often spelt Phajus—is so closely allied with Calanthe that for hybridizing purposes at least there is no distinction. Dominy raised Ph. irroratus from Ph. grandifolius × Cal. vestita; Seden made the same cross, but, using the variety Cal. v. rubro-occulata, he obtained Ph. purpureus. The success is more interesting because one parent is evergreen, the other, Calanthe, deciduous. On this account probably very few seedlings survive; they show the former habit. Mr. Cookson alone has yet raised a cross between two species of Phajus—Ph. Cooksoni from Ph. Wallichii × Ph. tuberculosus. One may say that this is the best hybrid yet raised, saving Calanthe Veitchii, if all merits be considered—stateliness of aspect, freedom in flowering, striking colour, ease of cultivation. One bulb will throw up four spikes—twenty-eight have been counted in a twelve-inch pot—each bearing perhaps thirty flowers.

Seden has made two crosses of Chysis, both from the exquisite Ch. bractescens, one of the loveliest flowers that heaven has granted to this world, but sadly fleeting. Nobody, I believe, has yet been so fortunate as to obtain seed from Ch. aurea. This species has the rare privilege of self-fertilization—we may well exclaim, Why! why?—and it eagerly avails itself thereof so soon as the flower begins to open. Thus, however watchful the hybridizer may be, hitherto he has found the pollen masses melted in hopeless confusion before he can secure them.

One hybrid Epidendrum has been obtained—Epi. O'Brienianum from Epi. evectum × Epi. radicans; the former purple, the latter scarlet, produce ×a bright crimson progeny.

Miltonias show two natural hybrids, and one artificial—Mil. Bleuiana from Mil. vexillaria × Mil. Roezlii; both of these are commonly classed as Odontoglots, and I refer to them elsewhere under that title. M. Bleu and Messrs. Veitch made this cross about the same time, but the seedlings of the former flowered in 1889, of the latter, in 1891. Here we see an illustration of the advantage which French horticulturists enjoy, even so far north as Paris; a clear sky and abundant sunshine made a difference of more than twelve months. When Italians begin hybridizing, we shall see marvels—and Greeks and Egyptians!

Masdevallias are so attractive to insects, by striking colour, as a rule, and sometimes by strong smell—so very easily fertilized also—that we should expect many natural hybrids in the genus. They are not forthcoming, however. Reichenbach displayed his scientific instinct by suggesting that two species submitted to him might probably be the issue of parents named; since that date Seden has produced both of them from the crosses which Reichenbach indicated.

We have three natural hybrids among Phal?nopsis. Ph. intermedia made its appearance in a lot of Ph. Aphrodite, imported 1852. M. Porte, a French trader, brought home two in 1861; they were somewhat different, and he gave them his name. Messrs. Low imported several in 1874, one of which, being different again, was called after Mr. Brymer. Three have been found since, always among Ph. Aphrodite; the finest known is possessed by Lord Rothschild. That these were natural hybrids could not be doubted; Seden crossed Ph. Aphrodite with Ph. rosea, and proved it. Our garden hybrids are two: Ph. F.L. Ames, obtained from Ph. amabilis × Ph. intermedia, and Ph.Harriett? from Ph. amabilis × Ph. violacea, named after the daughter of Hon. Erastus Corning, of Albany, U.S.A.

Oncidiums yield only two natural hybrids at present, and those uncertain; others are suspected. We have no garden hybrids, I believe, as yet. So it is with Odontoglossums, as has been said, but in the natural state they cross so freely that a large proportion of the species may probably be hybrids. I allude to this hereafter.

I have left Cypripediums to the last, in these hasty notes, because that supremely interesting genus demands more than a record of dry facts. Darwin pointed out that Cypripedium represents the primitive form of orchid. He was acquainted with no links connecting it with the later and more complicated genera; some have been discovered since that day, but it is nevertheless true that "an enormous extinction must have swept away a multitude of intermediate forms, and left this single genus as the record of a former and more simple state of the great orchidacean order." The geographical distribution shows that Cypripedium was more common in early times—to speak vaguely—and covered an area yet more extensive than now. And the process of extermination is still working, as with other primitive types.

Messrs. Veitch point out that although few genera of plants are scattered so widely over the earth as Cypripedium, the species have withdrawn to narrow areas, often isolated, and remote from their kindred. Some are rare to the degree that we may congratulate ourselves upon the chance which put a few specimens in safety under glass before it was too late, for they seem to have become extinct even in this generation. Messrs. Veitch give a few striking instances. All the plants of Cyp. Fairieanum known to exist have sprung from three or four casually imported in 1856. Two bits of Cyp. superbiens turned up among a consignment of Cyp. barbatum; none have been found since, and it is doubtful whether the species survives in its native home. Only three plants of Cyp. Marstersianium have been discovered. They reached Mr. Bull in a miscellaneous case of Cypripediums forwarded to him by the Director of the Botanic Gardens at Buitzenzorze, in Java; but that gentleman and his successors in office have been unable to find another plant. These three must have reached the Gardens by an accident—as they left it—presented perhaps by some Dutchman who had been travelling.

Cyp. purpuratum is almost extinct at Hong Kong, and is vanishing fast on the mainland. It is still found occasionally in the garden of a peasant, who, we are told, resolutely declines to sell his treasure. This may seem incredible to those who know the Chinaman, but Mr. Roebelin vouches for the fact; it is one more eccentricity to the credit of that people, who had quite enough already. Collectors expect to find a new habitat of Cyp. purpuratum in Formosa when they are allowed to explore that realm. Even our native Cyp. calceolus has almost disappeared; we get it now from Central Europe, but in several districts where it abounded the supply grows continually less. The same report comes from North America and Japan. Fortunate it is, but not surprising to the thoughtful observer, that this genus grows and multiplies with singular facility when its simple wants are supplied. There is no danger that a species which has been rescued from extinction will perish under human care.

This seems contradictory. How should a plant thrive better under artificial conditions than in the spot where Nature placed it? The reason lies in that archaic character of the Cypriped which Darwin pointed out. Its time has passed—Nature is improving it off the face of the earth. A gradual change of circumstances makes it more and more difficult for this primitive form of orchid to exist, and, conscious of the fate impending, it gratefully accepts our help.

One cause of extermination is easily grasped. Cypripeds have not the power of fertilizing themselves, except a single species, Cyp. Schlimii, which—accordingly, as we may say—is most difficult to import and establish; moreover, it flowers so freely that the seedlings are always weak. In all species the sexual apparatus is so constructed that it cannot be impregnated by accident, and few insects can perform the office. Dr. Hermann Muller studied Cyp. calceolus assiduously in this point of view. He observed only five species of insect which fertilize it. Cyp. calceolus has perfume and honey, but none of the tropical species offer those attractions. Their colour is not showy. The labellum proves to be rather a trap than a bait. Large insects which creep into it and duly bear away the pollen masses, are caught and held fast by that sticky substance when they try to escape through the lateral passages, which smaller insects are too weak to force their way through.

Natural hybrids occur so rarely, that their existence is commonly denied. The assertion is not quite exact; but when we consider the habits of the genus, it ceases to be extraordinary that Cypripeds rarely cross in their wild state. Different species of Cattleya, Odontoglots, and the rest live together on the same tree, side by side. But those others dwell apart in the great majority of cases, each species by itself, at a vast distance perhaps from its kindred. The reason for this state of things has been mentioned—natural laws have exterminated them in the spaces between, which are not so well fitted to maintain a doomed race.

Doubtless Cypripeds rarely fertilize—by comparison, that is, of course—in their native homes. The difficulty that insects find in performing that service has been mentioned. Mr. Godseff points out to me a reason far more curious and striking. When a bee displaces the pollen masses of a Cattleya, for instance, they cling to its head or thorax by means of a sticky substance attached to the pollen cases; so, on entering the next flower, it presents the pollen outwards to the stigmatic surface. But in the case of a Cypriped there is no such substance, the adhesive side of the pollen itself is turned outward, and it clings to any intruding substance. But this is the fertilizing part. Therefore, an insect which by chance displaces the pollen mass carries it off, as one may say, the wrong side up. On entering the next flower, it does not commonly present the surface necessary for impregnation, but a sterile globule which is the backing thereof. We may suppose that in the earlier age, when this genus flourished as the later forms of orchid do now, it enjoyed some means of fertilization which have vanished.

Under such disadvantages it is not to be expected that seed capsules would be often found upon imported Cypripeds. Messrs. Veitch state that they rarely observed one among the myriads of plants that have passed through their hands. With some species, however, it is not by any means so uncommon. When Messrs. Thompson, of Clovenfords, bought a quantity of the first Cyp. Spicerianum which came upon the market, they found a number of capsules, and sowed them, obtaining several hundred fine plants. Pods are often imported on Cyp. insigne full of good seed.

In the circumstances enumerated we have the explanation of an extraordinary fact. Hybrids or natural species of Cypripediums artificially raised are stronger than their parents, and they produce finer flowers. The reason is that they get abundance of food in captivity, and all things are made comfortable for them; whilst Nature, anxious to be rid of a form of plant no longer approved, starves and neglects them.

The same argument enables us to understand why Cypripeds lend themselves so readily to the hybridizer. Darwin taught us to expect that species which can rarely hope to secure a chance of reproduction will learn to make the process as easy and as sure as the conditions would admit—that none of those scarce opportunities may be lost. And so it proves. Orchidaceans are apt to declare that "everybody" is hybridizing Cypripeds nowadays. At least, so many persons have taken up this agreeable and interesting pursuit that science has lost count of the less striking results. Briefly, the first hybrid Cypripedium was raised by Dominy, in 1869, and named after Mr. Harris, who, as has been said, suggested the operation to him. Seden produced the next in 1874—Cyp. Sedeni from Cyp. Schlimii × Cyp. longiflorum; curious as the single instance yet noted in which seedlings turn out identical, whichever parent furnish the pollen-masses. In every other case they vary when the functions of the parents are exchanged.

For a long time after 1853, when serious work begun, Messrs. Veitch had a monopoly of the business. It is but forty years, therefore, since experiments commenced, in which time hundreds of hybrids have been added to our list of flowers; but—this is my point—Nature has been busy at the same task for unknown ages, and who can measure the fruits of her industry? I do not offer the remark as an argument; our observations are too few as yet. It may well be urged that if Nature had been thus active, the "natural hybrids" which can be recognized would be much more numerous than they are. I have pointed out that many of the largest genera show very few; many none at all. But is it impossible that the explanation appears to fail only because we cannot yet push it far enough? When the hybridizer causes by force a fruitful union betwixt two genera, he seems to triumph over a botanical law. But suppose the genera themselves are artificial, only links in a grand chain which Nature has forged slowly, patiently, with many a break an............