The best authorities consider all our grapes as the descendants of one species which now grows wild in western Asia, which grew wild during the Bronze age in Italy (10/1. Heer 'Pflanzen der Pfahlbauten' 1866 s. 28.), and which has recently been found fossil in a tufaceous deposit in the south of France. (10/2. Alph. De Candolle 'Geograph. Bot.' page 872; Dr. A.

Targioni-Tozzetti in 'Jour. Hort. Soc.' vol 9 page 133. For the fossil vine found by Dr. G. Planchon see 'Nat. Hist. Review' 1865 April page 224. See also the valuable works of M. de Saporta on the 'Tertiary Plants of France.') Some authors, however, entertain much doubt about the single parentage of our cultivated varieties, owing to the number of semi-wild forms found in Southern Europe, especially as described by Clemente (10/3.

G.o.dron 'De l'Espece' tome 2 page 100.) in a forest in Spain; but as the grape sows itself freely in Southern Europe, and as several of the chief kinds transmit their characters by seed (10/4. See an account of M.

Vibert's experiments by Alex. Jordan in 'Mem. de l'Acad. de Lyon' tome 2 1852 page 108.), whilst others are extremely variable, the existence of many different escaped forms could hardly fail to occur in countries where this plant has been cultivated from the remotest antiquity. That the vine varies much when propagated by seed, we may infer from the largely increased number of varieties since the earlier historical records. New hot-house varieties are produced almost every year; for instance (10/5.

'Gardener's Chronicle' 1864 page 488.) a golden-coloured variety has been recently raised in England from a black grape without the aid of a cross.

Van Mons (10/6. 'Arbres Fruitiers' 1836 tome 2 page 290.) reared a mult.i.tude of varieties from the seed of one vine, which was completely separated from all others, so that there could not, at least in this generation, have been any crossing, and the seedlings presented "les a.n.a.logues de toutes les sortes," and differed in almost every possible character both in the fruits and foliage.

The cultivated varieties are extremely numerous; Count Odart says that he will not deny that there may exist throughout the world 700 or 800, perhaps even 1000 varieties, but not a third of these have any value. In the catalogue of fruit cultivated in the Horticultural Gardens of London, published in 1842, 99 varieties are enumerated. Wherever the grape is grown many varieties occur: Pallas describes 24 in the Crimea, and Burnes mentions 10 in Cabool. The cla.s.sification of the varieties has much perplexed writers, and Count Odart is reduced to a geographical system; but I will not enter on this subject, nor on the many and great differences between the varieties. I will merely specify a few curious and trifling peculiarities, all taken from Odart's highly esteemed work (10/7. Odart 'Ampelographie Universelle' 1849.) for the sake of showing the diversified variability of this plant. Simon has cla.s.sed grapes into two main divisions, those with downy leaves, and those with smooth leaves, but he admits that in one variety, namely the Rebazo, the leaves are either smooth, or downy; and Odart (page 70) states that some varieties have the nerves alone, and other varieties their young leaves, downy, whilst the old ones are smooth. The Pedro-Ximenes grape (Odart page 397) presents a peculiarity by which it can be at once recognised amongst a host of other varieties, namely, that when the fruit is nearly ripe the nerves of the leaves or even the whole surface becomes yellow. The Barbera d'Asti is well marked by several characters (page 426), amongst others, "by some of the leaves, and it is always the lowest on the branches, suddenly becoming of a dark red colour." Several authors in cla.s.sifying grapes have founded their main divisions on the berries being either round or oblong; and Odart admits the value of this character; yet there is one variety, the Maccabeo (page 71), which often produces small round, and large oblong, berries in the same bunch. Certain grapes called Nebbiolo (page 429) present a constant character, sufficient for their recognition, namely, "the slight adherence of that part of the pulp which surrounds the seeds to the rest of the berry, when cut through transversely." A Rhenish variety is mentioned (page 228) which likes a dry soil; the fruit ripens well, but at the moment of maturity, if much rain falls, the berries are apt to rot; on the other hand, the fruit of a Swiss variety (page 243) is valued for well sustaining prolonged humidity. This latter variety sprouts late in the spring, yet matures its fruit early; other varieties (page 362) have the fault of being too much excited by the April sun, and in consequence suffer from frost. A Styrian variety (page 254) has brittle foot-stalks, so that the cl.u.s.ters of fruit are often blown off; this variety is said to be particularly attractive to wasps and bees. Other varieties have tough stalks, which resist the wind. Many other variable characters could be given, but the foregoing facts are sufficient to show in how many small structural and const.i.tutional details the vine varies. During the vine disease in France certain old groups of varieties (10/8. M. Bouchardat in 'Comptes Rendus'

December 1, 1851 quoted in 'Gardener's Chronicle' 1852 page 435. See also C.V. Riley on the manner in which some few of the varieties of the American Labruscan Vine escape the attacks of the Phylloxera: 'Fourth Annual Report on the Insects of Missouri' 1872 page 63 and 'Fifth Report' 1873 page 66.) have suffered far more from mildew than others. Thus "the group of Cha.s.selas, so rich in varieties, did not afford a single fortunate exception;" certain other groups suffered much less; the true old Burgundy, for instance, was comparatively free from disease, and the Carminat likewise resisted the attack. The American vines, which belong to a distinct species, entirely escaped the disease in France; and we thus see that those European varieties which best resist the disease must have acquired in a slight degree the same const.i.tutional peculiarities as the American species.

WHITE MULBERRY (Morus alba).

I mention this plant because it has varied in certain characters, namely, in the texture and quality of the leaves, fitting them to serve as food for the domesticated silkworm, in a manner not observed with other plants; but this has arisen simply from such variations in the mulberry having been attended to, selected, and rendered more or less constant. M. de Quatref.a.ges (10/9. 'Etudes sur les Maladies actuelles du Ver a Soie' 1859 page 321.) briefly describes six kinds cultivated in one valley in France: of these the AMOUROUSO produces excellent leaves, but is rapidly being abandoned because it produces much fruit mingled with the leaves: the ANTOFINO yields deeply cut leaves of the finest quality, but not in great quant.i.ty: the CLARO is much sought for because the leaves can be easily collected: lastly, the ROSO bears strong hardy leaves, produced in large quant.i.ty, but with the one inconvenience, that they are best adapted for the worms after their fourth moult. MM. Jacquemet-Bonnefont, of Lyon, however, remark in their catalogue (1862) that two sub-varieties have been confounded under the name of the roso, one having leaves too thick for the caterpillars, the other being valuable because the leaves can easily be gathered from the branches without the bark being torn.

In India the mulberry has also given rise to many varieties. The Indian form is thought by many botanists to be a distinct species; but as Royle remarks (10/10. 'Productive Resources of India' page 130.), "so many varieties have been produced by cultivation that it is difficult to ascertain whether they all belong to one species;" they are, as he adds, nearly as numerous as those of the silkworm.

THE ORANGE GROUP.

We here meet with great confusion in the specific distinction and parentage of the several kinds. Gallesio (10/11. 'Traite du Citrus' 1811. 'Teoria della Riproduzione Vegetale' 1816. I quote chiefly from this second work.

In 1839 Gallesio published in folio 'Gli Agrumi dei Giard. Bot. di Firenze'

in which he gives a curious diagram of the supposed relationship of all the forms.), who almost devoted his life-time to the subject, considers that there are four species, namely, sweet and bitter oranges, lemons, and citrons, each of which has given rise to whole groups of varieties, monsters, and supposed hybrids. One high authority (10/12. Mr. Bentham 'Review of Dr. A. Targioni-Tozzetti, Journal of Hort. Soc.' volume 9 page 133.) believes that these four reputed species are all varieties of the wild Citrus medica, but that the shaddock (Citrus dec.u.mana), which is not known in a wild state, is a distinct species; though its distinctness is doubted by another writer "of great authority on such matters," namely, Dr.

Buchanan Hamilton. Alph. De Candolle (10/13. 'Geograph. Bot.' page 863.), on the other hand--and there cannot be a more capable judge--advances what he considers sufficient evidence of the orange (he doubts whether the bitter and sweet kinds are specifically distinct), the lemon, and citron, having been found wild, and consequently that they are distinct. He mentions two other forms cultivated in j.a.pan and Java, which he ranks undoubted species; he speaks rather more doubtfully about the shaddock, which varies much, and has not been found wild; and finally he considers some forms, such as Adam's apple and the bergamotte, as probably hybrids.

I have briefly abstracted these opinions for the sake of showing those who have never attended to such subjects, how perplexing they are. It would, therefore, be useless for my purpose to give a sketch of the conspicuous differences between the several forms. Besides the ever-recurrent difficulty of determining whether forms found wild are truly aboriginal or are escaped seedlings, many of the forms, which must be ranked as varieties, transmit their characters almost perfectly by seed. Sweet and bitter oranges differ in no important respect except in the flavour of their fruit, but Gallesio (10/14. 'Teoria della Riproduzione' pages 52-57.) is most emphatic that both kinds can be propagated by seed with absolute certainty. Consequently, in accordance with his simple rule, he cla.s.ses them as distinct species; as he does sweet and bitter almonds, the peach and nectarine, etc. He admits, however, that the soft-sh.e.l.led pine-tree produces not only soft-sh.e.l.led but some hard-sh.e.l.led seedlings, so that a little greater force in the power of inheritance would, according to this rule, raise a soft-sh.e.l.led pine-tree into the dignity of an aboriginally created species. The positive a.s.sertion made by Macfayden (10/15. Hooker 'Bot. Misc.' volume 1 page 302; volume 2 page 111.) that the pips of sweet oranges produced in Jamaica, according to the nature of the soil in which they are sown, either sweet or bitter oranges, is probably an error; for M.

Alph. De Candolle informs me that since the publication of his great work he has received accounts from Guiana, the Antilles, and Mauritius, that in these countries sweet oranges faithfully transmit their character. Gallesio found that the willow-leafed and the Little China oranges reproduced their proper leaves and fruit; but the seedlings were not quite equal in merit to their parents. The red-fleshed orange, on the other hand, fails to reproduce itself. Gallesio also observed that the seeds of several other singular varieties all reproduced trees having a peculiar physiognomy, partly resembling their parent-forms. I can adduce another case: the myrtle leaved orange is ranked by all authors as a variety, but is very distinct in general aspect: in my father's greenhouse, during many years, it rarely yielded any fruit, but at last produced one; and a tree thus raised was identical with the parent-form.

Another and more serious difficulty in determining the rank of the several forms is that, according to Gallesio (10/16. 'Teoria della Riproduzione'

page 53.) they largely intercross without artificial aid; thus he positively states that seeds taken from lemon-trees (C. lemonum) growing mingled with the citron (C. medica), which is generally considered as a distinct species, produced a graduated series of varieties between these two forms. Again, an Adam's apple was produced from the seed of a sweet orange, which grew close to lemons and citrons. But such facts hardly aid us in determining whether to rank these forms as species or varieties; for it is now known that undoubted species of Verbasc.u.m, Cistus, Primula, Salix, etc., frequently cross in a state of nature. If indeed it were proved that plants of the orange tribe raised from these crosses were even partially sterile, it would be a strong argument in favour of their rank as species. Gallesio a.s.serts that this is the case; but he does not distinguish between sterility from hybridism and from the effects of culture; and he almost destroys the force of this statement by another (10/17. Gallesio 'Teoria della Riproduzione' page 69.) namely, that when he impregnated the flowers of the common orange with the pollen taken from undoubted VARIETIES of the orange, monstrous fruits were produced, which included "little pulp, and had no seeds, or imperfect seeds."

In this tribe of plants we meet with instances of two highly remarkable facts in vegetable physiology: Gallesio (10/18. Ibid page 67.) impregnated an orange with pollen from a lemon, and the fruit borne on the mother tree had a raised stripe of peel like that of a lemon both in colour and taste, but the pulp was like that of an orange and included only imperfect seeds.

The possibility of pollen from one variety or species directly affecting the fruit produced by another variety of species, is a subject which I shall fully discuss in the following chapter.

The second remarkable fact is, that two supposed hybrids (10/19. Gallesio 'Teoria della Riproduzione' pages 75, 76.) (for their hybrid nature was not ascertained), between an orange and either a lemon or citron, produced on the same tree leaves, flowers, and fruit of both pure parent-forms, as well as of a mixed or crossed nature. A bud taken from any one of the branches and grafted on another tree produces either one of the pure kinds or a capricious tree reproducing the three kinds. Whether the sweet lemon, which includes within the same fruit segments of differently flavoured pulp (10/20. 'Gardener's Chronicle' 1841 page 613.), is an a.n.a.logous case, I know not. But to this subject I shall have to recur.

I will conclude by giving from A. Risso (10/21. 'Annales du Museum' tome 20 page 188.) a short account of a very singular variety of the common orange.

It is the "citrus aurantium fructu variabili," which on the young shoots produces rounded-oval leaves spotted with yellow, borne on petioles with heart-shaped wings; when these leaves fall off, they are succeeded by longer and narrower leaves, with undulated margins, of a pale-green colour embroidered with yellow, borne on footstalks without wings. The fruit whilst young is pear-shaped, yellow, longitudinally striated, and sweet; but as it ripens, it becomes spherical, of a reddish-yellow, and bitter.

PEACH AND NECTARINE (Amygdalus persica).

The best authorities are nearly unanimous that the peach has never been found wild. It was introduced from Persia into Europe a little before the Christian era, and at this period few varieties existed. Alph. De Candolle (10/22. 'Geograph. Bot.' page 882.), from the fact of the peach not having spread from Persia at an earlier period, and from its not having pure Sanscrit or Hebrew names, believes that it is not an aboriginal of Western Asia, but came from the terra incognita of China. The supposition, however, that the peach is a modified almond which acquired its present character at a comparatively late period, would, I presume, account for these facts; on the same principle that the nectarine, the offspring of the peach, has few native names, and became known in Europe at a still later period.

(FIGURE 42.-PEACH AND ALMOND STONES, of natural size, viewed edgeways. 1.

Common English peach. 2. Double, crimson-flowered, Chinese Peach. 3.

Chinese Honey Peach. 4. English Almond. 5. Barcelona Almond. 6. Malaga Almond. 7. Soft-sh.e.l.led French Almond. 8. Smyrna Almond.)

Andrew Knight (10/23. 'Transactions of Hort. Soc.' volume 3 page 1 and volume 4 page 396 and note to page 370. A coloured drawing is given of this hybrid.), from finding that a seedling-tree, raised from a sweet almond fertilised by the pollen of a peach, yielded fruit quite like that of a peach, suspected that the peach-tree is a modified almond; and in this he has been followed by various authors. (10/24. 'Gardener's Chronicle' 1856 page 532. A writer, it may be presumed Dr. Lindley, remarks on the perfect series which may be formed between the almond and the peach. Another high authority, Mr. Rivers, who has had such wide experience, strongly suspects ('Gardener's Chronicle' 1863 page 27) that peaches, if left to a state of nature, would in the course of time retrograde into thick-fleshed almonds.) A first-rate peach, almost globular in shape, formed of soft and sweet pulp, surrounding a hard, much furrowed, and slightly flattened stone, certainly differs greatly from an almond, with its soft, slightly furrowed, much flattened, and elongated stone, protected by a tough, greenish layer of bitter flesh. Mr. Bentham (10/25. 'Journal of Hort. Soc.' volume 9 page 168.) has particularly called attention to the stone of the almond being so much more flattened than that of the peach. But in the several varieties of the almond, the stone differs greatly in the degree to which it is compressed, in size, shape, strength, and in the depth of the furrows, as may be seen in figure 42 (Nos. 4 to 8) of such kinds as I have been able to collect. With peach-stones also (Nos. 1 to 3) the degree of compression and elongation is seen to vary; so that the stone of the Chinese Honey-peach (No. 3) is much more elongated and compressed than that of the (No. 8) Smyrna almond. Mr. Rivers, of Sawbridgeworth, to whom I am indebted for some of the specimens above figured, and who has had such great horticultural experience, has called my attention to several varieties which connect the almond and the peach. In France there is a variety called the Peach-Almond, which Mr. Rivers formerly cultivated, and which is correctly described in a French catalogue as being oval and swollen, with the aspect of a peach, including a hard stone surrounded by a fleshy covering, which is sometimes eatable. (10/26. Whether this is the same variety as one lately mentioned ('Gardener's Chronicle' 1865 page 1154) by M. Carriere under the name of persica intermedia, I know not; this variety is said to be intermediate in nearly all its characters between the almond and peach; it produces during successive years very different kinds of fruit.) A remarkable statement by M. Luizet has recently appeared in the 'Revue Horticole' (10/27. Quoted in 'Gardener's Chronicle' 1866 page 800.), namely, that a Peach-almond, grafted on a peach, bore, during 1863 and 1864 almonds alone, but in 1865 bore six peaches and no almonds. M. Carriere, in commenting on this fact, cites the case of a double-flowered almond which, after producing during several years almonds, suddenly bore for two years in succession spherical fleshy peach-like fruits, but in 1865 reverted to its former state and produced large almonds.

Again, as I hear from Mr. Rivers, the double-flowering Chinese peaches resemble almonds in their manner of growth and in their flowers; the fruit is much elongated and flattened, with the flesh both bitter and sweet, but not uneatable, and it is said to be of better quality in China. From this stage one small step leads us to such inferior peaches as are occasionally raised from seed. For instance, Mr. Rivers sowed a number of peach-stones imported from the United States, where they are collected for raising stocks, and some of the trees raised by him produced peaches which were very like almonds in appearance, being small and hard, with the pulp not softening till very late in the autumn. Van Mons (10/28. Quoted in 'Journal de La Soc. Imp. d'Horticulture' 1855 page 238.) also states that he once raised from a peach-stone a peach having the aspect of a wild tree, with fruit like that of the almond. From inferior peaches, such as these just described, we may pa.s.s by small transitions, through clingstones of poor quality, to our best and most melting kinds. From this gradation, from the cases of sudden variation above recorded, and from the fact that the peach has not been found wild, it seems to me by far the most probable view, that the peach is the descendant of the almond, improved and modified in a marvellous manner.

One fact, however, is opposed to this conclusion. A hybrid, raised by Knight from the sweet almond by the pollen of the peach, produced flowers with little or no pollen, yet bore fruit, having been apparently fertilised by a neighbouring nectarine. Another hybrid, from a sweet almond by the pollen of a nectarine, produced during the first three years imperfect blossoms, but afterwards perfect flowers with an abundance of pollen. If this slight degree of sterility cannot be accounted for by the youth of the trees (and this often causes lessened fertility), or by the monstrous state of the flowers, or by the conditions to which the trees were exposed, these two cases would afford a good argument against the peach being the descendant of the almond.

Whether or not the peach has proceeded from the almond, it has certainly given rise to nectarines, or smooth peaches, as they are called by the French. Most of the varieties, both of the peach and nectarine, reproduce themselves truly by seed. Gallesio (10/29. 'Teoria della Riproduzione Vegetale' 1816 page 86.) says he has verified this with respect to eight races of the peach. Mr. Rivers (10/30. 'Gardener's Chronicle' 1862 page 1195.) has given some striking instances from his own experience, and it is notorious that good peaches are constantly raised in North America from seed. Many of the American sub-varieties come true or nearly true to their kind, such as the white-blossom, several of the yellow-fruited freestone peaches, the blood clingstone, the heath, and the lemon clingstone. On the other hand, a clingstone peach has been known to give rise to a freestone.

(10/31. Mr. Rivers 'Gardener's Chronicle' 1859 page 774.) In England it has been noticed that seedlings inherit from their parents flowers of the same size and colour. Some characters, however, contrary to what might have been expected, often are not inherited; such as the presence and form of the glands on the leaves. (10/32. Downing 'The Fruits of America' 1845 pages 475, 489, 492, 494, 496. See also F. Michaux 'Travels in N. America'

English translation page 228. For similar cases in France see G.o.dron 'De l'Espece' tome 2 page 97.) With respect to nectarines, both cling and freestones are known in North America to reproduce themselves by seed.

(10/33. Brickell 'Nat. Hist. of N. Carolina' page 102 and Downing 'Fruit Trees' page 505.) In England the new white nectarine was a seedling of the old white, and Mr. Rivers (10/34. 'Gardener's Chronicle' 1862 page 1196.) has recorded several similar cases. From this strong tendency to inheritance, which both peach and nectarine trees exhibit,--from certain slight const.i.tutional differences (10/35. The peach and nectarine do not succeed equally well in the some soil: see Lindley 'Horticulture' page 351.) in their nature,--and from the great difference in their fruit both in appearance and flavour, it is not surprising, notwithstanding that the trees differ in no other respects and cannot even be distinguished, as I am informed by Mr. Rivers, whilst young, that they have been ranked by some authors as specifically distinct. Gallesio does not doubt that they are distinct; even Alph. De Candolle does not appear perfectly a.s.sured of their specific ident.i.ty: and an eminent botanist has quite recently (10/36.

G.o.dron 'De l'Espece' tome 2 1859 page 97.) maintained that the nectarine "probably const.i.tutes a distinct species."

Hence it may be worth while to give all the evidence on the origin of the nectarine. The facts in themselves are curious, and will hereafter have to be referred to when the important subject of bud-variation is discussed. It is a.s.serted (10/37. 'Transact. Hort. Soc.' volume 6 page 394.) that the Boston nectarine was produced from a peach-stone, and this nectarine reproduced itself by seed. (10/38. Downing's 'Fruit Trees' page 502.) Mr.

Rivers states (10/39. 'Gardener's Chronicle' 1862 page 1195.) that from stones of three distinct varieties of the peach he raised three varieties of nectarine; and in one of these cases no nectarine grew near the parent peach-tree. In another instance Mr. Rivers raised a nectarine from a peach, and in the succeeding generation another nectarine from this nectarine.

(10/40. 'Journal of Horticulture' February 5, 1866 page 102.) Other such instances have been communicated to me, but they need not be given. Of the converse case, namely, of nectarine-stones yielding peach-trees (both free and clingstones), we have six undoubted instances recorded by Mr. Rivers; and in two of these instances the parent nectarines had been seedlings from other nectarines. (10/41. Mr. Rivers in 'Gardener's 'Chronicle' 1859 page 774, 1862 page 1195; 1865 page 1059; and 'Journal of Hort.' 1866 page 102.)

With respect to the more curious case of full-grown peach-trees suddenly producing nectarines by bud-variation (or sports as they are called by gardeners), the evidence is superabundant; there is also good evidence of the same tree producing both peaches and nectarines, or half-and-half fruit; by this term I mean a fruit with the one-half a perfect peach, and the other half a perfect nectarine.

Peter Collinson in 1741 recorded the first case of a peach-tree producing a nectarine (10/42. 'Correspondence of Linnaeus' 1821 pages 7, 8, 70.) and in 1766 he added two other instances. In the same work, the editor, Sir J.E.

Smith, describes the more remarkable case of a tree in Norfolk which usually bore both perfect nectarines and perfect peaches; but during two seasons some of the fruit were half and half in nature.

Mr. Salisbury in 1808 (10/43. 'Transact. Hort. Soc.' volume 1 page 103.) records six other cases of peach-trees producing nectarines. Three of the varieties are named; viz., the Alberge, Belle Chevreuse, and Royal George.

This latter tree seldom failed to produce both kinds of fruit. He gives another case of a half-and-half fruit.

At Radford in Devonshire (10/44. Loudon's 'Gardener's Mag.' 1826 volume 1.

page 471.) a clingstone peach, purchased as the Chancellor, was planted in 1815, and in 1824, after having previously produced peaches alone, bore on one branch twelve nectarines; in 1825 the same branch yielded twenty-six nectarines, and in 1826 thirty-six nectarines, together with eighteen peaches. One of the peaches was almost as smooth on one side as a nectarine. The nectarines were as dark as, but smaller than, the Elruge.

At Beccles a Royal George peach (10/45. Loudon's 'Gardener's Mag.' 1828 page 53.) produced a fruit, "three parts of it being peach and one part nectarine, quite distinct in appearance as well as in flavour." The lines of division were longitudinal, as represented in the woodcut. A nectarine- tree grew five yards from this tree.

Professor Chapman states (10/46. Ibid 1830 page 597.) that he has often seen in Virginia very old peach-trees bearing nectarines.

A writer in the 'Gardener's Chronicle' says that a peach tree planted fifteen years previously (10/47. 'Gardener's Chronicle' 1841 page 617.) produced this year a nectarine between two peaches; a nectarine-tree grew close by.

In 1844 (10/48. 'Gardener's Chronicle' 1844 page 589.) a Vanguard peach- tree produced, in the midst of its ordinary fruit, a single red Roman nectarine.

Mr. Calver is stated (10/49. 'Phytologist' volume 4 page 299.) to have raised in the United States a seedling peach which produced a mixed crop of both peaches and nectarines.

Near Dorking (10/50. 'Gardener's Chronicle' 1856 page 531.) a branch of the Teton de Venus peach, which reproduces itself truly by seed (10/51. G.o.dron, 'De l'Espece' tome 2 page 97.), bore its own fruit "so remarkable for its prominent point, and a nectarine rather smaller but well formed and quite round."

The previous cases all refer to peaches suddenly producing nectarines, but at Carclew (10/52. 'Gardener's Chronicle' 1856 page 531.) the unique case occurred, of a nectarine-tree, raised twenty years before from seed and never grafted, producing a fruit half peach and half nectarine; subsequently bore a perfect peach.

To sum up the foregoing facts; we have excellent evidence of peach-stones producing nectarine-trees, and of nectarine-stones producing peach-trees,-- of the same tree bearing peaches and nectarines,--of peach-trees suddenly producing by bud-variation nectarines (such nectarines reproducing nectarines by seed), as well as fruit in part nectarine and in part peach,- -and, lastly, of one nectarine-tree first bearing half-and-half fruit, and subsequently true peaches. As the peach came into existence before the nectarine, it might have been expected from the law of reversion that nectarines would have given birth by bud-variation or by seed to peaches, oftener than peaches to nectarines; but this is by no means the case.

Two explanations have been suggested to account for these conversions.

First, that the parent trees have been in every case hybrids (10/53. Alph.

De Candolle 'Geograph. Bot.' page 886.) between the peach and nectarine, and have reverted by bud-variation or by seed to one of their pure parent forms. This view in itself is not very improbable; for the Mountaineer peach, which was raised by Knight from the red nutmeg-peach by pollen of the violette hative nectarine (10/54. Thompson in Loudon's 'Encyclop. of Gardening' page 911.), produces peaches, but these are said SOMETIMES to partake of the smoothness and flavour of the nectarine. But let it be observed that in the previous list no less than six well-known varieties and several unnamed varieties of the peach have once suddenly produced perfect nectarines by bud variation: and it would be an extremely rash supposition that all these varieties of the peach, which have been cultivated for years in many districts, and which show not a vestige of a mixed parentage, are, nevertheless, hybrids. A second explanation is, that the fruit of the peach has been directly affected by the pollen of the nectarine: although this certainly is possible, it cannot here apply; for we have not a shadow of evidence that a branch which has borne fruit directly affected by foreign pollen is so profoundly modified as afterwards to produce buds which continue to yield fruit of the new and modified form.

Now it is known that when a bud on a peach-tree has once borne a nectarine the same branch has in several instances gone on during successive years producing nectarines. The Carclew nectarine, on the other hand, first produced half-and-half fruit, and subsequently pure peaches. Hence we may confidently accept the common view that the nectarine is a variety of the peach, which may be produced either by bud-variation or from seed. In the following chapter many a.n.a.logous cases of bud-variation will he given.

The varieties of the peach and the nectarine run in parallel lines. In both cla.s.ses the kinds differ from each other in the flesh of the fruit being white, red, or yellow; in being clingstones or freestones; in the flowers being large or small, with certain other characteristic differences; and in the leaves being serrated without glands, or crenated and furnished with globose or reniform glands. (10/55. 'Catalogue of Fruit in Garden of Hort.

Soc.' 1842 page 105.) We can hardly account for this parallelism by supposing that each variety of the nectarine is descended from a corresponding variety of the peach; for though our nectarines are certainly the descendants of several kinds of peaches, yet a large number are the descendants of other nectarines, and they vary so much when thus reproduced that we can scarcely admit the above explanation.

The varieties of the peach have largely increased in number since the Christian era, when from two to five varieties were known (10/56. Dr. A.

Targioni-Tozzetti 'Journal Hort. Soc.' volume 9 page 167. Alph. de Candolle 'Geograph. Bot.' page 885.) and the nectarine was unknown. At the present time, besides many varieties said to exist in China, Downing describes, in the United States, seventy-nine native and imported varieties of the peach; and a few years ago Lindley (10/57. 'Transact. Hort. Soc.' volume 5 page 554. See also Carriere 'Description et Cla.s.s. des Varietes de Pechers.') enumerated one hundred and sixty-four varieties of the peach and nectarine grown in England. I have already indicated the chief points of difference between the several varieties. Nectarines, even when produced from distinct kinds of peaches, always possess their own peculiar flavour, and are smooth and small. Clingstone and freestone peaches, which differ in the ripe flesh either firmly adhering to the stone, or easily separating from it, also differ in the character of the stone itself; that of the freestones or melters being more deeply fissured, with the sides of the fissures smoother than in clingstones. In the various kinds the flowers differ not only in size, but in the larger flowers the petals are differently shaped, more imbricated, generally red in the centre and pale towards the margin: whereas in the smaller flowers the margin of the petal is usually more darkly coloured. One variety has nearly white flowers. The leaves are more or less serrated, and are either dest.i.tute of glands, or have globose or reniform glands (10/58. Loudon's 'Encyclop. of Gardening' page 907.) and some few peaches, such as the Brugnen, bear on the same tree both globular and kidney-shaped glands. (10/59. M. Carriere in 'Gardener's Chronicle'

1865 page 1154.) According to Robertson (10/60. 'Transact. Hort. Soc.'