miemiffVmmaimtti^^v'i'iif'nwtmimiiwmfiimf'mfr'nii'm'nitmmmiitmffmH' Xf/ iMftfia'-HaWMJfrVA"** ';f«ilW*«W*l«-W*»Jti,\« «-.«|K*w»i(!ft-A'»«J f ., r;:S{ if 1 1 1 V''1,T1 tes on the History of Plants and Geographical Botany from Chinese Sources, in 8vo, 51 pp., -with illustrations, Foochoo, without date, but the preface bears the date Dec. 1870. iiotes on Some Botanical QucstionSj in 8vo, 14 pp., 1880. METHODS FOR PROVING ORIGIN OF SPECIES. 25 make mention here and there of plants, with epithets or reflections on their mode of flowering, their ripening, their use, etc., which allow their names to be divined, and to be referred to modern botanical nomenclature. With the added help of a knowledge of the flora of the country, and of the common names in the languages derived from the dead language, it is possible to discover approximately the sense of some words. This is the case with Sanskrit,^ Hebrew,^ and Armenian.^ Lastly, a third category of dead languages offers no certainty, but merely presumptions or hypothetical and rare indications. It comprehends those tongues in which there is no written work, such as Keltic, with its dialects, the ancient Sclavonic, Pelasgic, Iberian, the speech of the primitive Aryans, Turanians, etc. It is possible to guess certain names or their approximate form in these dead languages by two methods, both of which should be employed with caution. The flrst and best is to consult the languages derived, or which we believe to be derived, directly from the ancient tongues, as Basque for the Iberian language, Albanian for the Pelasgic, Breton, Erse, and Gaelic for Keltic. The danger lies in the possibility of mistake in the filiation of the languages, and especially in a mistaken belief in the antiquity of a plant-name which may have * Wilson's dictionary contains names of plants, but botanists have more confidence in the names indicated by Roxburgh in his Flora Indica (edit, of 1832, 3 vols, in 8vo), and in Piddington's English Index to the Plants of India, Calcutta, 1832. Scholars fiud a greater number of words in the texts, but they do not give sufficient proof of the sense of these words. As a rule, we have not in Sanskrit what we have in Hebrew, Greek, and Chinese — a quotation of phrases concerning each word translated into a modern language. * The best work on the plant-names in the Old Testament is that of Rosenmiiller, Handhuch der biblischen Alterkunde, in 8vo, vol. iv., Leipzig, 1830. A good short work, in French, is La Botanique de la Bible, by Fred. Hamilton, in Svo, Nice, 1871. ' Reynier, a Swiss botanist, who had been in Egypt, has given the sense of many plant-names in the Talmud. See his volumes entitled Economie Publigue et Eurale des Arabes et des Juifs, in Svo, 1820 ; and Economie FuhJique et Rurale des Egrjptiens et des Carthaginois, in Svo, Lausanne, 1823. The more recent works of Duschak and Low are not based upon a knowledge of Eastern plants, and are unintelligible to botanists because of names in Syriac and Hebrew characters. 26 ORIGIN OF CULTIVATED PLANTS. been introduced by another people. Thus the Basque language contains many words which seem to have been taken from the Latin at the time of the Roman rule. Berber is full of Arab words, and Persian of words of every origin, which probably did not exist in Zend. The other method consists in reconstructing!: a dead language which had no literature, by means of those which are derived from it ; for instance, the speech of the Western Aryans, by means of the words common to several European languages which have sprung from it. Fick's dictionary will hardly serve for the words of ancient Aryan languages, for he gives but few plant- names, and his arrangement renders it unintellio'ible to those who have no knowledge of Sanskrit. Adolphe Pictet's work ^ is far more important to naturalists, and a second edition, augmented and improved, has been published since the author's death. Plant-names and agricultural terms are explained and discussed in this work, in a manner all the more satisfactory that an accurate knowledge of botany is combined with philology. If the author attributes perhaps too much importance to doubtful etymologies, be makes up for it by other knowledge, and by his excellent method and lucidity. The plant-names of the Euskarian or Basque language have been considered from the point of view of their probable etymology by the Comte de Charencey, in Les Ades de la Societe Pldlologiqiie (vol. i. No. 1, 1860). I shall have occasion to quote this work, of which the difficulties were great, in the absence of all literature and of all derived lang^uagres. 6. The necessity for combining the different methods. The various methods of which I have spoken are of unequal value. It is clear that when we have archa?o- h)gical records about a given species, like those of the Egyptian monuments, or of the Swiss lake-dwellings, these are facts of remarkable accuracy. Then come the data furnished by botany, especially those on the spontaneous existence of a species in a given country-. * Adolpho Pictet, Les Oriyines des Peuples Indo-Europden-<, 3 vols, in 8vo, Paris, 1878. METHODS FOE PROVING ORIGIN OF SPECIES. 27 These, if examined with care, may be very important. The assertions contained in the works of historians or even of naturalists respecting an epoch at which science Avas only beginning, have not the same value. Lastly, the common names are only an accessory means, especially in modern languages, and a means which, as we have seen, is not entirely trustworthy. So mnch may be said in a general way, but in each particular case one method or the other may be more or less important. Each can only lead to probabilities, since we are dealing with facts of ancient date which are beyond the reach of direct and actual observation. Fortunately, if the same probability is attained in three or four different ways, we approach very near to certainty. The same rule holds good for researches into the history of plants as for researches into the history of nations. A good author consults historians who have spoken of events, the archives in which unpublished documents are found, the inscriptions on ancient monuments, the news- papers, private letters, finally memoirs and even tradition. He gathers probabilities from every source, and then compares these probabilities, weighs and discusses them before deciding. It is a labour of the mind which requires intelligence and judgment. This labour differs widely from observation employed in natural history, and from pure reason which is proper to the exact sciences. Nevertheless, when, by several methods, we reach the same probability, I repeat that the latter is very nearly a certainty. We may even say that it is as much a certainty as historical science can pretend to attain. I have the proof of this when I compare my present work with that which I composed by the same methods in 1855. For the species which I then studied, I have now moi'e authorities and better authenticated facts, but my conclusions on the origin of each species have scarcely altered. As they were already based on a combination of methods, probabilities have usually become certainties, and I have not been led to conclusions absolutely contrary to those previously formed. Ai'chseological, philological, and botanical data become 28 ORIGIN OF CULTIVATED PLANTS. more and more numerous. By their means the history of cultivated plants is perfected, while the assertions of ancient authors lose instead of gaining in importance. From the discoveries of antiquaries and philologists, moderns are better acquainted than the Greeks with Chaldea and ancient Egypt. They can prove mistakes in Herodotus. Botanists on their side correct Theo- phrastus, Dioscorides, and Pliny from their knowledge of the flora of Greece and Italy, while the study of classical authors to which learned men have applied themselves for three centuries has already furnished all that it has to give. I cannot help smiling when, at the present day, savants repeat well-known Greek and Latin phrases, and draw from them what they call conclusions. It is trying to extract juice from a lemon which has already been repeatedly squeezed. We must say it frankly, the works which repeat and commentate on the ancient authors of Greece and Rome without gi'v'ing the first place to botanical and arch^ological facts, are no longer on a level with the science of the day. JNevertheless, I could name several German works which have attained to the honour of a third edition. It would have been better to reprint the earlier publications of Fraas and Lenz, of Targioni and Heldreich, which have always given more weight to the modern data of botany, than to the vague descriptions of classic authors; that is to say, to facts than to words and phrases. ■■■< i PART II. On the Study of Species, considered as to their Origin, their early Cultivation, and the Principal Facts of their Diffusion.^ CHAPTER I PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS, SUCH AS ROOTS, TUBERCLES, OR BULBS.^ Radish. — Raphanus sativus, Linnreus. The radish is cultivated for what is called the root, which is, properly speaking, the lower part of the stem with the tap root.^ Every one knows how the size, shape, and colour of those organs which become fleshy vary according to the soil or the variety. There is no doubt that the species is indigenous in the temperate regions of the old world ; but, as it has been cultivated in gardens from the earliest historic times, from Cliina and Japan to Europe, and as it sows ^ A certain number of species whose origin is well known, such as the carrot, sorrel, etc., are mentioned only in the summary at the begin- ning of the last part, with an indication of the principal facts concerning them. 2 Some species are cultivated sometimes for their roots and some- times for their leaves or seeds. In other chapters will be found species cultivated sometimes for their leaves (as fodder) or for their seeds, etc. I have classed them according to their commonest use. The alpha- betical index refers to the place assigned to each species. ^ Sec the young state of the plant when the part of the stem below the cotyledons is not yet swelled. Turpin gives a drawing of it in the Annales des Sciences Naturelles, series 1, vol. xxi. pi. 5. 30 ORIGIN OF CULTIVATED PLANTS. itself frequently round cultivated plots, it is difficult to fix upon its starting-point. Formerly Raphanus sativus was confounded with kindred species of the Mediterranean region, to which certain Greek names were attributed ; but Gay, the botanist, who has done a good deal towards eliminat- ing these analogous forms,-^ considered R. sativus as a native of the East, perhaps of China. Linnaeus also sup- posed this plant to be of Chinese origin, or at least that variety which is cultivated in China for the .--ake of ex- tractinof oil from the seeds.^ Several floras of the south of Europe mention the species as subspontaneous or escaped from cultivation, never as spontaneous. Lede- bour had seen a specimen found near Mount Ararat, had sown the seeds of it and verified the species.^ However, Boissier,^ in 18G7, in his Eastern Flora, says that it is only subspontaneous in the cultivated parts of Anatolia, near Mersivan (according to Wied), in Palestine (on his own authority), in Armenia (according to Ledebour), and probably elsewhere, which agrees with the assertions found in European floras.^ Buhse names a locality, the Ssahend mountains, to the south of the Caucasus, which appears to be far enough from cultivation. The recent Flora of British India^ and the earlier Flora of Cochin- China by Loureiro, mention the radish only as a culti- vated species. Maximowicz saw it in a garden in the north-east of China.'' Thunberg speaks of it as a plant of general cultivation in Japan, and growing also by the side of the roads,^ but the latter fact is not repeated by modern authors, who are probably better informed.^ Herodotus (Hist., 1. 2, c. 125) speaks of a radish which he calls simnaia, used by the builders of the pyramid of ^ In A. de Candolle, Geogr. Bot. Baisonnee, p. 82G. * Linnoeus, Spec. Plant, p. 935. 8 Ledebonr, Fl. Ross., i. p. 225. * Boissier, FL Orient, i. p. 400. 5 Buhse, Aufzclhlung Transcancasien , p. 30. 6 Hooker, Flora of British India, i. p. 16G. ' Maximowicz, Primitice Florae Amurensis, p. 47. 8 Thunberg, Fl. Jap., p. 263. ® Franchet and Savatier, Enum. Plant. Jap., i. p. 39. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 31 Cheops, according to an inscription upon the monument. Unger ^ copied from Lepsius' work two drawings from the temple of Karnak, of which the first, at any rate, appears to represent the radish. From all this we gather, first, that the species spreads easily from cultivation in the west of Asia and the south of Europe, while it does not appear with cer- tainty in the flora of Eastern Asia ; and secondly, that in the regions south of the Caucasus it is found without any sign of culture, so that we are led to suppose that the plant is wild there. From these two reasons it appears to have come originally from Western Asia between Palestine, Anatolia, and the Caucasus, perhaps also from Greece ; its cultivation spreading east and west from a very early period. The common names support these hypotheses. In Europe they offer little interest when they refer to the quality of the root (raclis), or to some comparison with the turnip (ravanello in Italian, rahica in Spanish, etc.), but the ancient Greeks coined the special name raphanos (easily reared). The Italian word ramoraccio is derived from the Greek armoracia, which was used for B. sativus or some allied species. Modern interpreters have erro- neously referred this name to Cochlearia Armoracia or horse-radish, which I shall come to presently. Semitic^ languages have quite different names (fugla in Hebrew, fuil,fidgel, figl, etc., in Arab.). In India, according to Roxburgh,^ the common name of a variety with an enormous root, as large sometimes as a man's leg, is "inoola or nioolee, in Sanskrit miooluka. Lastly, for (Jochin-China, China, and Japan, authors give various names which differ very much one from the other. From this diversity a cultivation which ranged from Greece to Japan must be very ancient, but nothing can thence be concluded as to its original home as a spontaneous plant. A totally different opinion exists on the latter point, * Ungcr, PJlanzen des Alien JEgypfens, p. 51, figs. 24 and 29. ^ In my manuscript dictionary of common names, drawn from the floras of thirty years ago. ^ Roxburgh, Fl. Ind., iii. p. 126. 32 ORIGIN OF CULTIVATED PLANTS. which we iimst also examine. Several botanists ^ suspect that Raphaniis sativus is simply a particular condition, with enlarged root and non-articulated fruit, of Rapha- nus raplianistruvi, a very common plant in the tem- perate cultivated districts of Europe and Asia, and which is also found in a wild state in sand and light soil near the sea — for instance, at St. Sebastian, in Dal- matia, and at Trebizond.^ Its usual haunts are in deserted fields; and many common names which signify wild radish, show the affinity of the two plants. I should not insist upon this point if their supposed identity were a mere presumption, but it rests upon experiments and observations which it is important to know. In R. raphanistru7)i the siliqua is articulated, that is to say, contracted at intervals, and the seeds placed each in a division. In R. sativus the siliqua is con- tinuous, and forms a single cavity. Some botanists had made this difference the basis of two distinct genera, Raphanistruni and Rap)haniis. But three accurate ob- servers, Webb, Gay, and Spach, have noticed among plants of Raphaniis sativus, raised from the same seed, both unilocular and articulated pods, some of them bilocular, others plurilocular. Webb^ arrived at the same results when he afterwards repeated these experi- ments, and he observed yet another fact of some import- ance : the radish which sows itself by chance, and is not cultivated, produced the siliquse of Raphanistvwni.^ Another difference between the two plants is in the root, fleshy in R. sativus, slender in R. raphanis- trum ; but this changes with cultivation, as appears from the experiments of Carriere, the head gardener of the nurseries of the Natural History Museum in Paris.^ It occurred to him to sow the seeds of the slender- * Webb, Phytogr. Canar., p. 83 ; Iter. Hisp., p. 71 ; Bentliam, Fl. Hong Kong, p. 17 ; Hooker, Fl. Brit. Ind., i. p. 166. ' Willkomm and Lange, Prod. Fl. Hisp., iii. p. 7-iS ; Viviani, Flor. Dahnat., iii. p. 104-; Boissier, Fl. Orient., i. p. 401. ' Webb, Phytographia Canariensis, i. p. 83. * Webb, Iter. Hispaniense, 1838, p. 72. ' Carriere, Origins des Plantes Domestiques denxontrie 'par la Culture du Radis Sauvage, in 8vo, E-l pp., 1869. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 33 rooted RaphanistruTn in both stiff and light soil, and in the fourth generation he obtained fleshy radishes, of varied colour and form like those of our gardens. He even gives the figures, which are really curious and con- clusive. The pungent taste of the radish was not wantinor. To obtain these chano-es, Carriere sowed in September, so as to make the plant almost biennial instead of annual. The thickening of the root was the natural result, since many biennial plants have fleshy roots. The inverse experiment remains to be tried — to sow cultivated radishes in a poor soil. Probably the roots would become poorer and poorer, while the siliqui© would become more and more articulated. From all the experiments I have mentioned, Ra- phanus sativus might well be a variety of B. ra- phanistruTYi, an unstable variety determined by the existence of several generations in a fertile soil. We cannot suppose that ancient uncivilized peoples made essays like those of Carriere, but they may have noticed plants of RaphanistruTii grown in richly manured soil, with more or less fleshy roots ; and this soon suggested the idea of cultivating them. I have, however, one objection to make, founded on geographical botany. Raphanus raphanistrum is a European plant which does not exist in Asia.^ It can- not, therefore, be this species that has furnished the in- habitants of India, China, and Japan with the radishes which they have cultivated for centuries. On the other hand, how could R. raphanistrwifn, which is supposed to have been modified in Europe, have been transmitted in ancient times across the whole of Asia ? The transport of cultivated plants has commonly proceeded from Asia into Europe. Chang-Kien certainly brought vegetables from Bactriana into China in the second century B.C., but the radish is not named amono; the number. Horse-radish — Coclilearia Arnioracia, Linnreus. This Crucifer, whose rather hard root has the taste of * Ledebour, Fl. Ross. ; Boissier, FL Orient. Works on the flora of the valley of the Amur. 84 ORIGIN OF CULTIVATED PLANTS. mustard, was sometimes called in French cran, or cranson de Bretagne. Tliis was an error caused by the old botanical name ArniiOTacia, Yvdiich was taken for a cor- ruption of ATinoTica (Brittany). ArmioracAa occurs in Pliny, and was applied to a crucifer of the Pontine province, which was perhaps Raphaniis sativus. After I had formerly ^ pointed out this confusion, I expressed myself as follows on the mistaken origin of the species : — Cochlearia Arinoracia is not wild in Brittany, a fact now established by the researches of botanists in the west of France. The Abbe Delalande mentions it in his little work, entitled Hoedic et Hoiiat,^ in which he gives so interesting an account of the customs and pro- ductions of these two little islands of Brittany. He quotes the opinion of M. le Gall, who, in an unpublished flora of Morbihan, declares the plant foreign to Brittany. This proof, however, is less strong than others, since the south coast of the peninsula of Brittany is not yet sufficiently known to botanists, and the ancient Armorica extended over a portion of Normandy where the wild horse-radish is now found.^ This leads me to speak of the original home of the species. English botanists mention it as wild in Great Britain, but are doubtful about its origin. Watson ^ considers it as introduced by cultivation. The difficulty of extirpating it, he says, from places where it is cultivated, is well known to gardeners. It is therefore not surprising that this plant should take possession of waste ground, and persist there so as to appear indigenous. Babington ^ mentions only one spot where the species appears to be really wild, namely, Swansea. We will try to solve the problem by further arguments. Cochlearia Arrtioracia is a plant belonging to the temperate, and especially to the eastern regions of Europe. It is diffused from Finland to Astrakhan, and to the * A. de Canclolle, Geographic Botaniqne Baisonnee, p. 654. ' Delalande, Hoedic et Hovat, 8vo pamphlet, Nantes, 1850, p. 109. * Hardouin, Eenou, and Leclerc, Catalogue du Calvadoa, p. 85 ; De Brebisson, Fl. de Normandie, p. 25. * Watson, Cyhele, i. p. 159. * Babington, Manual of Brit. Bot., 2nd edit., p. 28. PLANTS CULTIVATED FOR THEIE SUBTERRANEAN PARTS. 85 desert of Cunian.^ Grisebach mentions also several localities in Turkey in Europe, near Enos, for instance, where it abounds on the sea-shore.^ The further we advance towards the west of Europe, the less the authors of floras appear sure that the plant is indigenous, and the localities assigned to it are more scattered and doubtful. The species is rarer in Norway than in Sweden,^ in the British Isles than in Holland, where a foreign origin is not attributed to it.^ The specific names confirm the impression of its origin in the east rather than in the west of Europe ; thus the name chren^ in Russia recurs in all the Sclavonic languages, krenai in Lithuanian, chren in Illyrian,^ etc. It has introduced itself into a few German dialects, round Vienna,^ for instance, where it persists, in spite of the spread of the German tongue. We owe to it also the French names cran or cranson. The word used in Germany, Meerretig, and in Holland, meer-radys, whence the Italian Swiss dialect has taken the name iineridi, or 7}ieredi, means sea-radish, and is not primitive like the word chren. It comes probably from the fact that the plant grows well near the sea, a circumstance common to many of the Cruciferce, and which should be the case with this species, for it is wild in the east of Russia where there is a good deal of salt soil. The Swedish name ]pei')paT-rot^ suggests the idea that the species came into Sweden later than the introduction of pepper by commerce into the north of Europe. However, the name may have taken the place of an older one, which has remained unknown to us. The Eno^lish name of horse- radish is not of such an original nature as to lead to a belief in the existence of the species in the country before the Saxon conquest. It means a very strong * Ledebour, Fl. Ross., i. p. 159. ^ Grisebach, Spicilegium Fl. Eumel., i. p. 2G5. ^ Flies, Summa, p. 30. ^ Miquel, Disquisitio pi. regn. Bafav. ' Moritzi, Diet. Ined. des Noms Vulgaires. ^ Moritzi, ibid. ; Viviani, Fl. Dahnat., iii. p. 32^. 7 Neilreich, Fl. Wien, p. 502. ^ Liunaeus, FL Suecica, No, 540. 36 ORIGIN OF CULTIVATED PLANTS. radish. The Welsh name rhuddygl maurth ^ is only the translation of the English word, whence we may infer that the Kelts of Great Britain had no special name, and were not acquainted with the species. In the west of France, the name raifort, which is the commonest, merely means strong root. Formerly it bore in France the names of German, or Capuchin mustard, which shows a foreign and recent origin. On the contrary, the word chren is in all the Sclavonic languages, a wordv/hich has penetrated into some German arid French dialects under the forms of ki^een, cran, and cranson, and which is certainly of a primitive nature, and shows the antiquity of the species in temperate Eastern Europe. It is therefore most probable that cultivation has propagated and naturalized the plant westward from the east for about a thousand years. Turnips — Brassica species et varietates radice in- crassata. The innumerable varieties and sub varieties of the turnip known as swedes, Kohl-rabi, etc., may be all attri- buted to one of the four species of Linnaeus — Brassica napus, Br. oleracea, Br. rapa^ Br. campestris — of which the two last should, according to modern authors, be fused into one. Other varieties of the species are cultivated for the leaves (cabbages), for the inflorescence (cauliflowers), or for the oil which is extracted from the seed (colza, rape, etc.). When the root or the lower part of the stem ^ is fleshy, the seed is not abundant, nor worth the trouble of extracting the oil ; when those organs are slender, the production of the seed, on the contrary, becomes more important, and decides the economic use of the plant. In other words, the store of nutritious matter is placed sometimes in the lower, sometimes in the upper part of the plant, although the organization of the flower and fruit is similar, or nearly so. ^ H. Davies, Welsh Botanology, p. G3. * In turnips and swedes the swelled part is, as in the radish, the lower part of the stem, below the cotyledons, with a more or less per- EiNtent part of the root. (See Turpin, Ann. Sc. Natur., ser. 1, vol. xxi.) In the Kohl-rabi (Brassica oleracea caiilo-rapa) it is the stem. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 37 Touching the question of origin, we need not occupy ourselves with the botanical limits of the species, and with the classification of the races, varieties, and sub- varieties,^ since all the Brassicce are of European and Siberian origin, and are still to be seen in these regions wild, or half wild, in some form or other. Plants so commonly cultivated and whose germina- tion is so easy often spread round cultivated places ; hence some uncertainty regarding the really wild nature of the plants found in the open country. Nevertheless, Linngeus mentions that Brassica napus grows in the sand on the sea-coast in Sweden (Gothland), Holland, and Eng- land, which is confirmed, as far as Sweden is concerned, by Fries,^ who, with his usual attention to questions of this nature, mentions Br. Campestris, L. (type of the Rapa with slender roots), as really wild in the whole Scandinavian peninsula, in Finland and Denmark. Ledebour ^ indicates it in the whole of Russia, Siberia, and the Caspian Sea. The floras of temperate and southern Asia mention rapes and turnips as cultivated plants, never as escaped from cultivation.* This is already an indication of foreign origin. The evidence of philology is no less significant. There is no Sanskrit name for these plants, but only modern Hindu and Bengalee names, and those only for Brassica rapa and B. oleracea} Ksempfer^ gives Japanese names for the turnip — husei, or more commonly aona — but there is nothing to show that these names are ancient. Bretschneider, who has made a careful study of Chinese authors, mentions no Brassica. Apparently they do not occur in any of the ancient works on botany and agricul- ture, although several varieties are now cultivated in China. It is just the reverse in Europe. The old languages * This classification has been the subject of a pnper by Angustin Pyramus de Candolle, Transactions of the Horticultural Society, vol. v. ^ Fries, Summa Veget. Scand., i. p. 29. ' Ledebour, Fl. Ross., i. p. 216. * BoissieTj Flora Orient ali s ; Sir J. Hooker, Flora of British India; Thnnberg, Flora Japonicas Franchet and Savatier, Enumeratio Plan- tarum Japonicarum. ' Piddington, Index, • Kajmpfer, Amcen., p. 822. 38 ORIGIN OF CULTIVATED PLANTS. have a number of names which seem to be oric;inal. Brassica rapa is called oneipen or erfinen'^ in Wales; repa and rippa in several Slav tongues,^ which answers to the Latin rapa, and is allied to the iieipa of the Anglo- Saxons. The Brassica^ napus is in AVelsh bresych yr yd ; in Erse hraisscagh hvAgh, according to Threlkeld,^ who sees in hraisscagh the root of the Latin Brassica. A Polish name, karpjiele, a Lithuanian, jellazoji,^ are also given, without speaking of a host of other names, transferred sometimes in popular speech from one species to another. I shall speak of the names of Brassica oleracea when I come to vegetables. The Hebrews had no names for cabbages, rapes, and turnips/ but there are Arab names : selgam for the Br. na.pus, and suhjum or suhjuvii for Br. rapa; words which recur in Persian and even in Bengali, transferred perhaps from one species to another. The cultivation of these plants has therefore been diffused in the south-west of Asia since Hebrew antiquity. Finally, every method, whether botanical, historical, or philological, leads us to the following conclusions : — Firstly, the Brassicoi with fleshy roots were originally natives of temperate Europe. Secondly, their cultivation was diffused in Europe before, and in Asia after, the Aryan invasion. Thirdly, the primitive slender-rooted form of Bras- sica napus, called Br. caoiipestris, had probably from the beginninor a more extended ranoje, from the Scan- dinavian peninsula towards Siberia and the Caucasus. Its cultivation w^as perhaps introduced into China and Japan, through Siberia, at an epoch which appears not to be much earlier than Greco-Roman civilization. Fourthly, the cultivation of the various forms or species of Brassica was diflused throughout the south-west of Asia at an epoch later than that of the ancient Hebrews. * Davies, Welsh Botanology, p. 65. " Moritzi, Diet. MS., compiled from published flcras. * Threlkeld, Synopsis Stirpium Ilihernicarum, 1 vol. in 8vo, 1727. < Moritzi, Diet. MS. ' Koaenmiiller, Biblische Naturgeschichte, vol. i., gives none. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 39 Skirret — Slum Sisarum, Linnaeus. This vivacious Umbellifer, furnished with several diverging roots in the form of a carrot, is believed to come from Eastern Asia. Linnaeus indicates China, doubtfully ; and Loureiro,^ China and Cochin-China, where he says it is cultivated. Others have mentioned Japan and the Corea, but in these countries there are species which it is easy to confound with the one in question, particularly Sium Ninsi and Panax Ginseng. Maximowicz,^ who has seen these plants in China and in Japan, and who has studied the herbariums of St. Petersburgh, recognizes only the Altaic region of Siberia and the North of Persia as the home of the wild Siunn Sisarum. I am very doubtful whether it is to be found in the Himalayas or in China, since modern works on the region of the river Amoor and on British India make no mention of it. It is doubtful whether the ancient Greeks and Romans knew this plant. The names Sisaron of Dioscorides, Siser of Columella and of Pliny,^ are attributed to it. Certainly the modern Italian name sisaro or sisero seems to confirm this idea; but how could these authors have failed to notice that several roots descend from the base of the stem, whereas all the other umbels cultivated in Europe have but a single tap-root? It is just possible that the siser of Columella, a cultivated plant, may have been the parsnip ; but what Pliny says of the siser does not apply to it. According to him it was a medicinal plant, inter medica dicenduiyi^ He says that Tiberius caused a quantity to be brought every year from Germany, which proves, he adds, that it thrives in cold countries. If the Greeks had received the plant direct from Persia, Theophrastus would probably have known it. It came perhaps from Siberia into Russia, and thence into Germany, in which case the anecdote about Tiberius might well apply to the skirret. I cannot find any * Linnaeus, Species, p. 361; Loureiro, Fl. Cochin cldnensis, p. 225. * Maximowicz, Diagnoses Plantarum JaponiccB et Manshurice, in Melanges Biologiques du Bulletin de VAcad., St. JPetersburg, decad 13, p. 18. 3 Dioscorides, Mat. Med., 1. 2, c. 139 j Columella, 1. 11, c. 3, 18, 35 j Lenz, Bat. der Alien, p. 560. * Pliny, Hist. Plant., 1. 19, c. 5. 40 ORIGIN OF CULTIVATED PLANTS. Russian name, certainly, but the Germans have original names, Krizel or Grizel, Gorlein or Gierlein, which indicate an ancient cultivation, more than the ordinary name Zuckemvurzel, or sugar-root.^ The Danish name has the same meaning — soJcerot, whence the English sJdrret The name sisaron is not known in modern Greece ; nor was it known there even in the Middle Ages, and the plant is not now cultivated in that country.^ There are reasons for doubt as to the true sense of the words sisaron and siser. Some botanists of the sixteenth century thought that sisaron was perhaps the loarsnip proper, and Sprengel ^ supports this idea. The French names chervis and girole * would perhaps teach us something if we knew their origin. Littre derives chervis from the Spanisli chirivia, but the latter is more likely derived from the French. Bauhin ^ mentions the low Latin names servilhmi, cliervilluin, or servillam, words which are not in Ducange's dictionary. This may well be the origin of chervis, but whence came servillam or chervillurn ? Arracacha or Arracacia — Arracacha esculenta, de Can- dolle. An umbel generally cultivated in Venezuela, New Granada, and Ecuador as a nutritious plant. In the tem- perate regions of those countries it bears comparison with the potato, and even yields, we are assured, a lighter and more agreeable fecida. The lower part of the stem is swelled into a bulb, on which, when the plant thrives well, tubercles, or lateral bulbs, form themselves, and persist for several months, which are more prized than the central bulb, and serve for future planting.^ The species is probably indigenous in the region where ' Nemnich, Polygl. Lexicon, ii. p. 1313. * Lenz, I?of. der Alten, p. 560; Heldreich, Nutzpfianzen Griechenlands ; Lan.i^kavel, Bot. der Spdteren Griechen. ' Sprengel, Dioscoridis, etc., ii. p. 4G2. ^ Olivier de Serres, Theatre de V Agriculture, p. 471. * Bauhin, Hist. PI., iii. p. 154. ^ The best information about the cultivation of this plant was priven by- Bancroft to Sir W. Hooker, and may be found in the Botanical Magazine, pi. 3092. A. P. de Candolie published, in La 5' Notice sur les Plantes Raves des Jardin Bot. de Geneve, an illustration showing the principal bulb. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN TARTS. 41 it is cultivated, but I do not find in any author a positive assertion of the fact. The existing descriptions are drawn from cultivated stocks. Grisebach indeed says that he has seen (presumably in the herbarium at Kew) specimens gathered in New Granada, in Peru, and in Trinidad,^ but he does not say whether they were wild. The other species of the same genus, to the number of a dozen, grow in the same districts of America, which renders the above- mentioned origin more probable. The introduction of the arracacha into Europe has been attempted several times without success. The damp climate of England accounts for the failure of Sir William Hooker's attempts ; but ours, made at two different times, under very different conditions, have met with no better success. The lateral bulbs did not form, and the central bulb died in the house where it was placed for the winter. The bulbs presented to different botanical gardens in France and Italy and elsewhere shared the same fate. lb is clear that if the plant is in America really equal to the potato in productiveness and taste, this will never be the case in Europe. Its cultivation does not in America spread as far as Chili and Mexico, like that of the potato and sweet potato, which confirms the difficulty of pro- pagation observed elsewhere. Madder — Ruhia tinctorum, Linnaeus. The madder is certainly wild in Italy, Greece, the Crimea, Asia Minor, Syria, Persia, Ai^menia, and near Lenkoran.^ As we advance westward in the south of Europe, the wild, indigenous nature of the plant becomes more and more doubtful. There is uncertainty even in France. In the north and east the plant appears to be " naturalized in hedges and on walls," ^ or " subspon- taneous," escaped from former cultivation.* In Provence and Languedoc it is more spontaneous or wild, but here also it may have spread from a somewhat extensive * Grisebach, Flora of British West. India Islands. * Bertoloni, Flora Italica, ii. p. 146 ; Decaisne, Recherches sxtr la Garance, p. 68 ; Boissier, Flo7-a Orientalis, iii. p. 17 ; Ledebour, Flora Bossica, ii. p. 405. ' Cosson and Germain, Flore des Environs de Paris, ii. p. 3G5. * Kirschleger, Flore d' Alsace, i. p. 359. 42 ORIGIN OF CULTIVATED PLANTS. cultivation. In the Iberian peninsula it is mentioned as "subspontaneous."^ It is the same in the north of Africa.^ Evidently the natural, ancient, and undoubted habitation is western temperate Asia and the south-east of Europe. It does not appear that the plant has been found beyond the Caspian Sea in the land formerly occupied by the Indo-Europeans, but this region is still little known. The species only exists in India as a cultivated j^lant, and has no Sanskrit name.^ Neither is there any known Hebrew name, while the Greeks, Romans, Slavs, Germans, and Kelts had various names, which a philologist could perhaps trace to one or two roots, but which nevertheless indicate by their numerous modifications an ancient date. Probably the wild roots were gathered in the fields before the idea of cultivating the species was suggested. Pliny, however, says ^ that it was cultivated in Italy in his time, and it is possible that the custom v/as of older date in Greece and Asia Minor. The cultivation of madder is often mentioned in French records of the Middle Acjes.^ It was afterwards neglected or abandoned, until Althen reintroduced it into the neifdibourhood of Avisfnon in the middle of the eighteenth century. It flourished formerly in Alsace, Germany, Holland, and especially in Greece, Asia Minor, and Syria, whence the exportation was considerable ; but the discovery of dyes extracted from inorganic substances has suppressed this cultivation, to the great detriment of the provinces which drew large profits from it. Jerusalem Artichoke — HelianfJais tuherosus, Linn?eus. It was in the year 161 G that European botanists first mentioned this Composite, with a large root better adapted for the food of animals than of man. Columna^ had seen it in the garden of Cardinal Farnese, and called it Aster" peruanus tuherosus. Other authors of the same * WlllkomTn and Lange, Prodromns Florce Hispaniccp, ii. p. 307- * Ball, Spicilegium Florae Maroccance, p. 483; Munby, Cat at. Plant. Alger., edit. 2, p. 17. 3 riddiugton, Index. * Pliniu=!, lib. 19, cap. 3. ^ De Gasparin, Trait4 d* Agriculture, iv. p. 253. * Columna, Ecphrasisj ii. p. 11. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 43 century gave it epithets showing that it was believed to come irom Brazil, or from Canada, or from the Indies, that is to say, America. Linnseus^ adopted, on Parkinson's authority, the opinion of a Canadian origin, of which, however, he had no proof I pointed out formerly ^ that there are no species of the genus Helianthus in Brazil, and that they are, on the contrary, numerous in North America. Schlechtendal,^ after having proved that the Jeru- salem artichoke can resist the severe winters of the centre of Europe, observes that this fact is in favour of the idea of a Canadian origin, and contrary to the belief of its cominof from some southern reojion. Decaisne^ has eliminated from the synonymy of H. tuberosus several quotations which had occasioned the belief in a South American or Mexican origin. Like the American botanists, he recalls what ancient travellers had narrated of certain customs of the aborigines of the Northern States and of Canada. Thus ('hamplain, in 1603, had seen, "in their hands, roots which they cul- tivate, and which taste like an artichoke." Lescarbot ^ speaks of these roots with the artichoke flavour, which multiply freely, and which he had brought back to France, where they began to be sold under the name of topinanihaiix. The savages, he says, call them chiquehi. Decaisne also quotes two French horticulturists of the seventeenth century, Colin and Sagard, who evidently speak of the Jerusalem artichoke, and say it came from Canada. It is to be noted that the name Canada had at that time a vague meaning, and compre- hended some parts of the modern United States. Gookin, an American writer on the customs of the aborigines, says that they put pieces of the Jerusalem artichoke into their soups.^ ' Linnasas, Hortus Cliffortianus, p. 420. ' A. de Candolle, Oeogr. Bot. Raisonnee, p. 824. ' Schlechtendal, Bot. Zeit. 1858, p. 113. * Decaisne, Recherch.es sur VOrigine de quelques-nnes de nos Plantes Alimentaires, in Flore des Serres et Jardins, vol. 23, 1881, p. 113. * Lescarbot, Histoire de la NoicveUe France, edit. 3, 1618, t. vi. p. 931. ^ Pickering, Chron. Arrang.y pp. 749, 972. 44 ORIGIN OF CULTIVATED PLANTS. Botanical analogies and the testimony of con- temporaries agree, as we have seen, in considering tiiis plant to be a native of the north-east of America. Dr. Asa Gray, seeing that it is not found v/ild, had formerly supposed it to be a variety of ^. dor onicoides of Jj2im.a,rck, but he has since abandoned this idea {A')nerican Journal of Science, 1883, p. 224). An author gives it as wild in the State of Indiana.^ The French name topinainhovbr comes apparently from some real or supposed Indian name. The Eno-lish name Jerusalem artichoke is a cor- ruption of the Italian girasole, sunflower, combined with an allusion to the artichoke flavour of the root. Salsify — Tragopogon ijorrifolhiin, LinnoBus. The salsify was more cultivated a century or two ago than it is now. It is a biennial composite, found wild in Greece, Dalmatia, Italy, and even in Algeria.^ It frequently escapes from gardens in the w^est of Europe, and becomes half-naturalized.^ Commentators* give the name Tragopogon (goat's beard) of Theophrastus sometimes to the modern species, sometimes to Tragopogon croci folium, which also grows in Greece. It is difficult to know if the ancients culti- vated the salsify or gathered it wild in the country. In the sixteenth century Olivier de Serres says it was a new culture in his country, the south of France. Our word Salsijis comes from the Italian Sassefrica, that which rubs stones, a senseless term. Scorzonera — Scovzonera Idspanica, Linn?eus. This plant is sometimes called the Spanish salsify, from its resemblance to Tragopogon porrifolium ; but its root has a brown skin, whence its botanical name, and the popular name ecorce noire in some French provinces. It is wild in Europe, from Spain, where it abounds, the * Catalngne of Indiana Plants, 18S1, p. 15. ' Boissier, Ft. Orient., iii. p. 7-15; Viv-iani, Fl. Dalmat., ii. p. 108; Bertoloni, Fl. Ital., viii. p. 348 ; Gussone, Symipsis FL Siculce, ii. p. 384; Munby, Catal. Alger., edit. 2, p. 22. ' A. do Candolle, Geogr. Bot. Raifonnee, p. 671. * Fraas, Synopsis Fl. C/a-ss., p. 190 j Leuz, Bot. der Alten, p. 485. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 45 south of France, and Germany, to the region of Cau- casus, and perhaps even as far as Siberia, but it is wanting in Sicily and Greece.^ In several parts of Germany the species is probably naturalized from cultivation. It seems that this plant has only been cultivated within the last hundred or hundred and fifty years. The botanists of the sixteenth century speak of it as a wild species introduced occasionally into botanical gardens. Olivier de Serres does not mention it. It was formerly supposed to be an antidote against the bite of adders, and was sometimes called the viper's plant. As to the etymology of the name Scorzonera, it is so evident, that it is difficult to understand how early writers, even Tournefort,^ have declared the origin of the word to be escorso, viper in Spanish or Catalan. Viper is in Spanish more commonly vibora. There exists in Sicily a Scorzonera deliciosa, Gussone, whose very sugary root is used in the confection of bonbons and sherbets, at Palermo.^ How is it that its cultivation has not been tried ? It is true that I tasted at Naples Scorzonera ices, and found them detestable, but they were perhaps made of the common species {Scorzo- nera hispanica). Potato — Solanum tuherosum, Linnseus. In 1855 I stated and discussed what was then known about the origin of the potato, and about its introduction into Europe.^ I will now add the result of the researches of the last quarter of a century. It will be seen that the data formerly acquired have become more certain, and that several somewdiat doubtful accessory questions have remained uncertain, though the probabilities in favour of what formerly seemed the truth have grown stronger. It is proved beyond a doubt that at the time of the discovery of America the cultivation of the potato was ^ Willkomm and Lange, Frodromus Florce Hispanicce, ii. p. 223; De Candolle, Flore Frangaise, iv. p. 59 ; Koch, Synopsis Fl. Germ., edit. 2, p. 488; Ledebour, Fl. B.oss., ii. p. 794; Boissier, Fl. Orientalis, iii. p. 767 ; Bertoloni, Fl.^Ifal., viii. p. 365. * Tournefort, Elements de Botanique, p. 379. ' Gussone, Synopsis Flora) SicuUr;. * A. de Candolle, G4ogr. Bot. liaisonncey pp. 810. 816. 46 OKIGIN OF CULTIVATED PLANTS. practised, v>^itli every appearance of ancient usage, in the temperate regions extending from Chili to New Granada, at altitudes varying with the latitude. This appears from the testimony of all the early travellers, amonp- whom I shall name Acosta for Peru,^ and Pedro Gieca, quoted by de I'Ecluse," for Quito. In the eastern temperate region of South America, on the heights of Guiana and Brazil, for instance, the potato was not kno^sra to the aborigines, or if they were acquainted with a similar plant, it was Solanum CGininersoniiy v/hich has also a tuberous root, and is found v/ild in Montevideo and in the south of Brazil. The true potato is certainly now cultivated in the latter country, but it is of such recent introduction that it has received the name of the English Batata.^ According to Humboldt it was unknown in Mexico,* a fact confirmed by the silence of subsequent authors, but to a certain degree contradicted by another historical fact. It is said that Sir Walter Raleigh, or rather Thomas Herriott, his companion in several voyages, brought back to Ireland, in 1585 or 1586, some tubers of the Virginian potato.'^ Its name in its own country was openaivh. From Herriott's description of the plant, quoted by Sir Joseph Banks,^ there is no doubt that it was the potato, and not the batata, which at that period was sometimes con- founded with it. Besides, Gerard'' tells us that he received from Virginia the potato which he cultivated in his garden, and of which he gives an illustration w^hich agrees in all points with Solanum tuherosurti. He w^as so proud of it that he is represented, in his portrait at the beginning of the work, holding in his hand a flowering branch of this plant. * Acosta, p. 163, verso. 2 De I'Ecluse (or Clusius), Rariariim Plantarum Historioe, 1601, lib. 4, p. Ixxix., with illustration. ' De Martius, Flora Brasil., vol. x. p. 12. * Von Humboldt, Nourelle Espagne, edit. 2, vol. ii. p. 451 ; Essai sur la GSographie des Plantes, p. 29. * At that epoch Virginia was not distinguished from Carolina. ^ Banks, Travs. Hort. Soc, 1805, vol. i. p. 8. ' Gerard, Herbal, 1597, p. 781, with illustration. PLANTS CULTIVATED FOR THEIH SUBTERRANEAN PARTS. 47 The species could scarcely have been introduced into Virginia or Carolina in Raleigh's time (1585), unless the ancient Mexicans had possessed it, and its cultivation had been diffused among the aborigines to the north of Mexico. Dr. Roulin, who has carefully studied the works on North America, has assured me that he has found no signs of the potato in the United States before the arrival of the Europeans. Dr. Asa Gray also told me so, adding that Mr. Harris, one of the men most intimately acquainted with the language and customs of North American tribes, was of the same opinion. I have read nothing to the contrary in recent publications, and we must not forget that a plant so easy of cultivation would have spread itself even among nomadic tribes, had they possessed it. It seems to me most likely that some inhabitants of Virginia — perhaps English colonists — received tubers from Spanish or other travellers, traders or adventurers, during the ninety years which had elapsed since the discovery of America. Evidently, dating from the conquest of Peru and Chili, in 1535 to 1585, many vessels could have carried tubers of the potato as pro- visions, and Sir Walter Raleigh, making war on the Spaniards as a privateer, may have pillaged some vessel which contained them. This is the less improbable, since the Spaniards had introduced the plant into Europe before 1585. Sir Joseph Banks ^ and Dunal ^ were right to insist upon the fact that the potato was first introduced by the Spaniard, since for a long time the credit was generally given to Sir Walter Raleigh, who was the second intro- ducer, and even to other Englishmen, who had introduced, not the potato but the batata (sweet potato), which is more or less confounded with it.^ A celebrated botanist, de I'Ecluse,* had nevertheless defined the facts in a ^ Banks, Trans. Hort. Soc, 1805, vol. i. p. 8. * Dunal, Hist. Nat. des Solanum, in 4to. ' The plant imported by Sir John Hawkins and Sir Francis Drake was clearly the sweet potato, Sir J. Banks sa3's ; whence it results that the questions discussed by Humboldt touching the localities visited by these travellers do not apply to the potato. * De I'Ecluse, Rariariun Plantarum Histuria, 1601, lib. 4, p. Ixxviii. 48 ORIGIN OF CULTIVATED PLANTS. remarkable manner. It is he who published the first good description and illustration of the potato, under the significant name of Papas Peruanorum. From Avhat he says, the species has little changed under the culture of nearly three centuries, for it yielded in the beginning as many as fifty tubers of unequal size, from one to two inches long, irregularly ovoid, reddish, ripening in November (at Vienna). The flower was more or less pink externally, and reddish within, with five longi- tudinal stripes of green, as is often seen now. No doubt numerous varieties have been obtained, but the original form has not been lost. De I'Ecluse compares the scent of the flower with that of the lime, the only difl'erence from our modern plant. He sowed seeds which produced a white-flowered variety, such as we sometimes see now. The plants described by de I'Ecluse were sent to him in 1588, by Philippe de Sivry, Seigneur of Waldheim and Governor of Mons, who had received them from some one in attendance on the papal legate in Belgium. De I'Ecluse adds that the species had been introduced into Italy from Spain or America (certitm est vet ex Hispania, vet ex America hahuisse), and he wonders that, although the plant had become so common in Italy that it was eaten like a turnip and given to the pigs, the learned men of the University of Padua only became acquainted with it by means of the tuber which he sent them from Germany. Targioni ^ has not been able to discover any proof that the potato was as widely cultivated in Italy at the end of the sixteenth century as de I'Ecluse asserts, but he quotes Father Magazzini of Vallombrosa, Avhose posthumous work, published in 1G23, mentions the species as one previously brought, without naming the date, from Spain or Portugal by barefooted friars. It was, therefore, towards the end of the sixteenth or at the beginning of the seventeenth century that the cultivation of the potato became known in Tuscany. Independently of what de I'Ecluse and the agriculturist of Vallombrosa * Targioni-Tozzetti, Leszioni, ii. p. 10 ; Cenni Storici sulV hitroduzione di Varie Piante neW Ajricoltiira di Toscana, 1 vol. in 8yo, Floreuce, 1853, p. 37. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 49 say of its introduction from the Iberian peninsula, it is not at all likely that the Italians had any dealings with Raleigh's companions. No one can doubt that the potato is of American origin ; but in order to know from what part of that vast continent it was brought, it is necessary to know if the plant is found wild there, and in what localities. To answer this question clearly, we must first remove two causes of error : the confusion of allied species of the genus Solanum with the potato ; and the other, the mistakes made by travellers as to the wild character of the plant. The allied species are Solanum Commersonii of Dunal, of which I have already spoken; S. miaglia of Molina, a Chili species; S. iramite of Dunal, a native of Peru; and >S'. verrncosum'^ of Schlechtendal, which grows in Mexico. These three kinds of Solanum have smaller tubers than S. tuberosum, and differ also in other characteristics indicated in special works on botany. Theoretically, it may be believed that all these, and other forms growing in America, are derived from a single earlier species, but in our geological epoch they present themselves with differences which seem to me to justify specific distinctions, and no experiments have proved that by crossing one with another a product would be obtained of which the seed (not the tubers) would propagate the race. Leaving these more or less doubtful questions of species, let us try to ascertain whether the common form oi Solanum tuberosum has been found wild, and merely remark that the abundance of tuberous solanums growing in the temperate regions of America, from Chili or Buenos Ay res as far as Mexico, con- firms the fact of an American orio;in. If we knew nothinix more, this would be a strong presumption in favour of this country being the original home of the potato. The second cause of error is very clearly explained * Solanum verrucosum, whose iatrorluction into the neighbourhood of Gex, near Geneva, I mentioned in 1855, has since been abandoned becanse its tubers are too small, and because it does not, as it was hoped, withstand the 'potato -fungus. 50 ORIGIN OF CULTIVATED PLANTS. by the botanist Weddell/ who has carefully explored Bolivia and the neighbouring countries. " When we reflect," he says, " that on the arid Cordillera the Indians often establish their little plots of cultivation on points which would appear almost inaccessible to the gi-eat majority of our European farmers, we understand that when a traveller chances to visit one of these cultivated plots, long since abandoned, and finds there a plant of Solanum tuherosum which has accidentally persisted, he gathers it in the belief that it is really wild ; but of this there is no proof." We come now to facts. These abound concerning the wild character of the plant in Chili. In 1822, Alexander Caldcleugh,^ English consul, sent to the London Horticultural Society some tubers of the potato which he had found in the ravines round Valparaiso. He says that these tubers are small, some- times red, sometimes yellowish, and rather bitter in taste.^ " I believe," he adds, " that this plant exists over a great extent of the littoral, for it is found in the south of Chili, where the aborigines call it maglia." This is probably a confusion with S. maglia of botanists; but the tubers of Valparaiso, planted in London, produced the true potato, as we see from a glance at Sabine's coloured hgure in the Transactions of the Horticultural Society. The cultivation of this plant was continued for some time, and Lindley certified anew, in 1847, its identity with the common potato.^ Here is the account of the Valparaiso plant, given by a traveller to Sir William Hooker.^ "I noticed the potato on the shore as far as fifteen leagues to the north of this town, and to the south, but I do not know how far it extends. It ' Chloris Andina, in 4to, p. 103. 2 Sabine, Trans. Hort. Soc, vol. v. p. 249. ' No importance should be attached to this flavour, nor to the watery quality of some of the tubers, since in hot countries, even in the south of Europe, the potato is often poor. The tubers, which are subter- ranean ramifications of the stem, are turned green by exposure to the light, and are rendered bitter. * Journal Hort. Soc, vol. iii. p. 66. * Hooker, Botanical Miscellanies, 1831, vol. ii. p. 203. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 51 groves on cliffs and hills near the sea, and I do not remember to have seen it more than two or three leagues from the coast. Although it is found in mountainous places, far from cultivation, it does not exist in the immediate neighbourhood of the fields and gardens where it is planted, excepting when a stream crosses these en- closures and carries the tubers into uncultiA'ated places." The potato described by these two travellers had white flowers, as is seen in some cultivated European varieties, and like the plant formerly reared by de I'Ecluse. We may assume that this is the natural colour of the species, or at least one of the most common in its wild state. Darwin, in his voyage in the Beagle, found the potato growing wild in great abundance on the sand of the sea-shore, in the archipelago of Southern Chili, and growing with a remarkable vigour, which may be attri- buted to the damp climate. The tallest plants attained to the height of four feet. The tubers were small as a rule, though one of them was two inches in diameter. They were watery, insipid, but with no bad taste when cooked. "The plant is undoubtedly wild," says the author,^ "and its specific identity has been confirmed first by Henslow, and afterwards by Sir Joseph Hooker in his Flora Antarctica:^ A specimen in the herbarium collected by Claude Gay, considered by Dunal to be Sclanum tuberosum, bears this inscription : " From the centre of the Cordilleras of Talcagouay, and of Cauquenes, in places visited only by botanists and geologists." The same author, Gay, in his Flora Chilena,^ insists upon the abundance of the wild potato in Chili, even among the Araucanians in the mountains of Malvarco, where, he says, the soldiers of Pincheira used to go and seek it for food. This evidence sufficiently proves its wild state in Chili, so that I may omit other less convincing testimony — for instance, that of Molina and Meyen, whose specimens from Chili have not been examined. The climate of the coast of Chili is continued upon * Journal of the Voyage, etc., edit. 1852, p. 285. * Vol. i. part 2, p. 329. ^ VoL v. p. 74. 52 ORIGIN OF CULTIVATED PLANTS. the heights as we follow the chain of the Ancles, and the cultivation of the potato is of ancient date in the tem- perate regions of Peru, but the wild character of the species there is not so entirely proved as in the case of (jhili.^ Pavon declared he found it on the coast at Chancay, and near Lima. The heat of these districts seems very great for a species which requires a temperate or even a rather cold climate. Moreover, the specimen in Boissier's herbarium, gathered by Pavon, belongs, ac- cording to Dunal,^ to another species, to which he has given the name of S. inimite. I have seen the authentic specimen, and have no doubt that it belongs to a species distinct from the >S^. tiiberosum. Sir W. Hooker ^ speaks of McLean's specimen, gathered in the hills round Lima, without any information as to whether it was found wild. The specimens (more or less wild) which Matthews sent from Peru to Sir W. Hooker belong, according to Sir Joseph,^ to varieties which differ a little from the true potato. Mr. Hemsley,^ who has seen them recently in the herbarium at Kew, believes them to be "distinct forms, not more distinct, however, than certain varieties of the species." Weddell,^ whose caution in this matter we already know, expresses himself as follows: — "I have never found Solanurti tiiberosum in Peru under such circum- stances as left no doubt that it was indigenous; and I even declare that I do not attach more belief to the wild nature of other plants found scattered on the Andes outside Chili, hitherto considered as indiirenous." On the other hand, M. Ed. Andrd '^ collected with great care, in two elevated and wild districts of Columbia, and in another near Lima, specimens which he believed he might attribute to 8. tuhtrosnm. M. Andre has been kind enough to lend them to me. I have compared them attentively with the types of Dunal's species in ' Buiz and Pavon, Flora Peruviana, ii. p. 38. 2 Dunal, Prodromus, xiii., sect. i. p. 22. * Hooker, Bot. Miscell., ii. * Hooker, Fl. Antarctica, * Journal Hort. Soc, new series, vol. v. 0 Weddell, Chloris Andina, p. 103. ^ Andre, in Illustration Horticole, 1877, p. 114. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 53 my herbarium and in that of M. Boissier. None of these Solanaceee belong, in my opinion, to S. tuberosum, although that of La Union, near the river Cauca, comes nearer than the rest. None — and this is yet more certain — answers to S. mionite of Dunal. They are nearer to S. coluvihianuvi of the same author than to S. tuberosum or >S^. immite. The specimen from Mount Quindio presents a singular characteristic — it has pointed ovoid berries.^ In Mexico the tuberous Solanums attributed to >S^. tuberosum, or, according to Hemsley,^ to allied forms, do not appear to be identical with the cultivated plant. They belong to aSi. Fendleri, which Dr. Asa Gray con- sidered at first as a separate species, and afterwards^ as a variety of S. tuber osuin or of B. verrucosum. We may sum up as follows : — 1. The potato is wild in Chili, in a form which is still seen in our cultivated plants. 2. It is very doubtful whether its natural home extends to Peru and New Granada. 3. Its cultivation was diffused before the discovery of America from Chili to New Granada. 4. It was introduced, probably in the latter half of the sixteenth century, into that part of the United States now known as Virginia and North Carolina. 5. It was imported into Europe between 1580 and 1585, first by the Spaniards, and afterwards by the English, at the time of Raleigh's voyages to Virginia.^ Batata, or Sweet Potato — Convolvulus batatas, Lin- naeus ; Batatas edulis, Choisy. The roots of this plant, swelled into tubers, resemble potatoes, whence it arose that sixteenth-century navi- gators applied the same name to these two very different species. The sweet potato belongs to the Convolvulus family, the potato to the Solanum family ; the fleshy ^ The form of the berries in S. columhianum and S. immite is not yet known. * Hemsley, Journal Horf. Soc, new series, vol. v. ' Asa Gray, Synoptical Flora of North America, ii. p. 227. * See, for the successive introduction into the different parts of Europe, Clos, Quelques Documents sur I'Histoire de la Pomme de Terre, in 8vo, 1874, in Journal d'Agric. Pratiq. du Midi de la France. 54 ORIGIN OF CULTIVATED PLANTS. parts of the former are roots, those of the latter subter- ranean branches.^ The sweet potato is sugary as well as farinaceous. It is cultivated in all countries within or near the tropics, and perhaps more in the new than in the old world.^ Its origin is, according to a great number of authors, doubtful. Humboldt,^ Meyen,* and Boissier ^ hold to its American, Boyer,^ Choisy,*^ etc., to its Asiatic origin. The same diversity is observed in earlier works. The question is the more difficult since the Convolvulacese is one of the most widely diffused families, either from a very early ej)och or in consequence of modern transportation. There are powerful arguments in favour of an American origin. The fifteen known species of the genus Batatas are all found in America ; eleven in that continent alone, four both in America and the old world, with possibility or probability of transportation. The cultivation of the common sweet potato is widely diffused in America. It dates from a very early epoch. Marcgraff® mentions it in Brazil under the name of jetica. Humboldt says that the name camote comes from a Mexican Avord. The word Batatas (w^hence comes by a mistaken transfer the word potato) is given as American. Sloane and Hughes^ speak of the sweet potato as of a plant much cultivated, and having several varieties in the West Indies. They do not appear to suspect that it had a foreign origin. Clusius, who was one of the first to mention the sweet potato, says he had eaten some in the south of Spain, wdiere it was supposed to have come from the new world.^^ He quotes the * Turpin gives figures which clearlj show these facts. Mem. dit Museum, vol. xix. plates 1, 2, 5. 2 Dr. Sagot gives interesting details on the method of cultivation, the product, etc., in the Journal Soc. d'Hortic. de France^ second series, vol. V. pp. 450-458. ' Humboldt, Nouvelle Espagne, edit. 2, vol. ii. p. 4/0. * Meyen, Gru7idrisse Pflanz. Geogr., p. 373. * Boissier, Voyage Botanique eii Espagne. ^ Bojer, Hort. Maurit., p. 225. ■* Choisy, in Prodromus, p. 33S. * Marcgraff, Bres., p. 16, with illustration. » Sloane, Hist. Jam., i. p. 150; Hughes, Barh., p. 22S. " Clusius, Rist., ii. p. 77. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 55 names Batatas, camotes, amote.% ajes} which were foreign to the languages of the old world. The date of his book is 1601. Humboldt^ says that, according to Gomara, Christopher Columbus, when he appeared for the first time before Queen Isabella, offered her various productions from the new world, sweet potatoes among others. Thus, he adds, the cultivation of this plant was already common in Spain from the beginning of the six- teenth century. Oviedo,^ writing in 1526, had seen the sweet potato freely cultivated by the natives of St. Domingo, and had introduced it himself at Avila, in Spain. Rumphius * says positively that, according to the general opinion, sweet potatoes were brought by the Spanish Americans to Manilla and the Moluccas, whence the Portuguese diffused it throughout the Malay Archipelago. He quotes the popular names, which are not Malay, and which indicate an introduction by the Castillians. Lastly, it is certain that the sweet potato was unknown to the Greeks, Romans, and Arabs ; that it was not cultivated in Egypt even eighty years ago,^ a fact which it would be hard to explain if we supposed its origin to be in the old world. On the other hand, there are arguments in favour of an Asiatic origin. The Chinese Encydojxedia of Agricul- ture speaks of the sweet potato, and mentions different varieties;^ but Bretschneider ''^ has proved that the species is described for the first time in a book of the second or third century of our era. According to Thunberg,^ the sweet potato was brought to Japan by the Portuguese. Lastly, the plant cultivated at Tahiti, in the neighbouring islands, and in New Zealand, under the names umara, gumarra, and gumalla, described by Forster ^ under the name of Convolvulus chrysorhizus, is, * Jjes was a name for the yam (Humboldt, Nouvelle Espagne). ' Humboldt, ibid. 3 Oviedo, Eamusio's translation, vol. iii. pt. 3. * Rumphius, Amhoin., v. p. 368. 5 Forskal, p. 54 ; Delile, III. ® D'Hex'vej Saint-Denys, Rech. sur VAgric. des Chin., 1850, p. 109. ' Study and Value of Chinese Botanical Works, p. 13. ' Thunberg, Flora Japon., p. 84. • Forster, Plantce Escul., p. 56. 56 ORIGIN OF CULTIVATED PLANTS. according to Sir Joseph Hooker, the sweet potato.^ Seemann^ remarks that these names resemble the Quichuen name of the sweet potato in America, which is, he says, ciimar. The cultivation of the sweet potato be- came general in Hindustan in the eighteenth century.^ Several popular names are attributed to it, and even, according to Piddington,^ a Sanskrit name, ruJdalu, which has no analogy with any name known to me, and is not in Wilson's Sanskrit Dictionary. According to a note given me by Adolphe Pictet, riildalu seems a Bengalee name composed from the Sanskrit alu {Rulcta plus dlu, the name of Arum camjyanulatuvi). This name in modern dialects designates the yam and the potato. However, Wallich^ gives several names omitted by Piddington. Roxburgh^ mentions no Sanskrit name. Rheede'^ says the plant was cultivated in Malabar, and mentions common Indian names. The aricuments in favour of an American oriojin seem to me much stronger. If the sweet potato had been known in Hindustan at the epoch of the Sanskrit lan^ua^fe it would have become dili'used in the old world, since its propagation is easy and its utility evident. It seems, on the contrary, that this cultivation remained long unknown in the Sunda Isles, Egypt, etc. Perhaps an attentive examination mi^rht lead us to share the opinion of Meyer,^ who distinguished the Asiatic plant from the American species. However, this author has not been generally followed, and I suspect that if there is a different Asiatic species it is not, as Meyer believed, the sweet potato described by Rumphius, which the latter says was brought from America, but the Indian plant of Roxburgh. Sweet potatoes are grown in Africa ; but either the cultivation is rare, or the species arc different. Robert Brown ^ says that the traveller Lockhardt had not seen * Hooker, HandhooTc of New Zealand Flora, p. 194. ' Seemann, Journal of Bot., 1866, p, 328. ' Roxburgh, edit. Wall., ii. p. 09. * Piddington, Index. ' Wallich, Flora hid. « Roxburgh, edit. 1832, vol. i. p. 483. ^ Rheede, Mul., vii. p. 95. ' Meyer, Prindtioe Fl. Esseq., p. 103. ' R. Brown, Bot. Congo, p. 55. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 57 the sweet potato of wliose cultivation the Portuguese missionaries make mention. Thonning ^ does not name it. Vogel brought back a species cultivated on the western coa.st, which is certainly, according to the authors of the Flora Nigritiana, Batatas panictdata of Choisy. It was, therefore, a plant cultivated for ornament or for medicinal purposes, for its root is purgative.^ It might be supposed that in certain countries in the nld or new world Ipomcea tuherosa, L,, had been confounded with the sweet potato; but Sloane^ tells us that its enormous roots are not eatable.* Ijyomoia raanfirriosa, Choisy {Convolvulus mammosus, Loureiro ; Batata mammosa, Rumphius), is a Convol- vulaceous plant with an edible root, which may well be confounded with the sweet potato, but whose botanical character is nevertheless distinct. This species grows wild near Amboyna (Rumphius), where it is also culti- vated. It is prized in Cochin-China. As for the sweet potato (BataJas edulis), no botanist, as far as I know, has asserted that he found it wild him- self, either in India or America.^ Clusius ^ affirms upon hearsay that it grows wild in the new world and in the neighbouring islands. In spite of the probability of an American origin, there remains, as we have seen, much that is unknown or uncertain touching the original home and the trans- port of this species, which is a valuable one in hot coun- tries. Whether it was a native of the new or of the old world, it is difficult to explain its transportation from America to China at the beginning of our era, and * Schumaclier and Thonning', Beslc. Guin. 2 Wallich, in Roxburgh, Fl. Ind., ii. p. 63. ' Sloane, Jam., i. p. 152. * Several Convolvulaceae have large roots, or more properly root- stocks, but in this case it is the base of the stem with a part of the root which is swelled, and this root-stock is always purgative, as in the Jalap and Turbith, while in the sweet potato it is the lateral roots, a different organ, which swell. * No. 701 of Schomburgh, coll. 1, is wild in Gniana. According to Choisy, it is a variety of the Batatas edulis ; according to Benthara (Hook, Jour. Bot., v. p, 352), of the Batatas panicrdata. My specimen, which is rather imperfect, seems to me to be different from both. * Clusius, Hist., ii. p. 77. 58 ORIGIN OF CULTIVATED PLANTS. to the South Sea Islands at an early epoch, or from Asia and from Australia to America at a time sufficiently remote for its cultivation to have been early diffused from the Southern States to Brazil and Chili, We must assume a prehistoric communication between Asia and America, or adopt another hypothesis, which is not in- applicable to the present case. The order Gonvolvulacece is one of those rare families of dicotyledons in which certain species have a widely extended area, extending even to distant continents.^ A species which can at the present day endure the different climates of Virginia and Japan may well have existed further north before the epoch of the great extension of glaciers in our hemisphere, and prehistoric men may have transported it southward when the climatic conditions altered. According to this hypothesis, cultivation alone preserved the species, unless it is at last discovered in soine spot in its ancient habitation — in Mexico or Columbia, for instance.^ Beetroot — Beta vidgaris and B. Tiiaritinia, Linnaeus ; Beta vulgaris, Moquin. This plant is cultivated sometimes for its fleshy root (red beet), sometimes for its leaves, which are used as a vegetable (white beet), but botanists are generally agreed in not dividing the species. It is known from other examples that plants slender rooted by nature easily become fleshy rooted from the effects of soil or cultivation. The slender-rooted variety grows wild in sandy soil, and especially near the sea in the Canary Isles, and all along the coasts of the Mediterranean Sea, and as far as the Caspian Sea, Persia, and Babylon,^ perhaps even as " A. de Candolle, Geogr. Bot. Raisonnd, pp. 1041-1043, and pp. 516-518. 2 Dr. Bretschneider, after haviiisf read the above, MTote to me from Pekin that the cultivated sweet potato is of origin foreign to China, according to Chinese authors. The handbook of agriculture of Nung. chang-tsuan-shu, whose author died in 1633, asserts this fact. He speaks of a sweet potato wild in China, called chu, the cultivated species being kan-chu. The Min-shu, piiblished in the sixteenth century, says that the introduction took place between 1573 and 1620. The American origin thus receives a further proof. * Jkloquin-Tandon, in Prodromus, vol. xiii. pt. 2, p. 55 j Boissier, Flora Onentalis, iv. p. 898; Ledebour, Fl. Rossica, iii. p. 602, PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 59 far as the west of India, whence a specimen was brought by Jaqiiemont, although it is not certain that it was growing wild. Roxburgh's Indian flora, and Aitchison's more recent flora of the Punjab and of the Sindh, only mention the plant as a cultivated species. It has no Sanskrit name,^ whence it may be inferred that the Aryans had not brought it from western tem- perate Asia, where it exists. The nations of Aryan race who had previously migrated into Europe probably did not cultivate it, for I find no name common to the Indo- European languages. The ancient Greeks, who used the leaves and roots, called the species teidlion;^ the Romans, beta. Heldreich^ gives also the ancient Greek name sevlde, or sfekelie, which resembles the Arab name selg, silq,^ among the Nabatheans. The Arab name has passed into the Portuguese selga. No Hebrew name is known. Everything shows that its cultivation does not date from more than three or four centuries before the Christian era. The red and white roots were known to the ancients, but the number of varieties has greatly increased in modern times, especially since the beetroot has been cultivated on a large scale for the food of cattle and for the production of sugar. It is one of the plants most easily improved by selection, as the experiments of Vilmorin have proved.^ Manioc — Manihot utilissima, Pohl; Jatropha ma- nihot, Linnaeus. The manioc is a shrub belonging to the Euphorbia family, of which several roots swell in their first year ; they take the form of an irregular ellipse, and contain a fecula (tapioca) with a more or less poisonous juice. It is commonly cultivated in the equatorial or tropical regions, especially in America from Brazil to the West Indies. In Africa the cultivation is less general, and seems to be more recent. In certain Asiatic colonies it is ^ Roxburgh, Flora Indica, ii. p. 59 ; Piddington, Index. * Theophrastus and Dioscorides, quoted by Lenz, Botanik der Grie- clien und Earner, p. 446 ; Fraas, Synopsis Fl. Class., p. 233. ' Heldreich, Die Nutzpjlanzen Griechenlands, p. 22. * Alawam, Agriculture nahatheenne, from E. Meyer, Geschichte der BotaniJc, iii. p. 75. * Notice sur I' Amdlioration des Plantes par le Semis, p. 15. 4 60 ORIGIN OF CULTIVATED PLANTS. decidedly of modem introduction. It is propagated by budding. Botanists are divided in opinion whether the innu- merable varieties of manioc should be reo^arded as form- ing one, two, or several different species. Pohl ^ admitted several besides his Manihot utilissima, and Dr. Mliller,^ in his monograph on the Euphorbiaceae, places the variety aijn in an allied species, M. 'p<^ihnata, a plant cultivated with the others in Brazil, and of which the root is not poisonous. This last character is not so distinct as might be believed from certain books and even from the asser- tions of the natives. Dr. Sagot,^ who has compared a dozen varieties of manioc cultivated at Cayenne, says expressly, " There are maniocs more poisonous than others, but I doubt whether any are entirely free from noxious principles." It is possible to account for these singular differences of properties in very similar plants by the example of the potato. The Manihot and Solanum titherosiini both belong to suspected families {Eupliorhiacece and Solanaceoi). Several of their species are poisonous in some of their organs ; but the fecula, wherever it is found, is never harmful, and the same holds good of the cellular tissue, freed from all deposit; that is to say, reduced to cellulose. In the preparation of cassava, or manioc flour, great care is taken to scrape the outer skin of the root, then to pound or crush the fleshy part so as to express the more or less poisonous juice, and finally the paste is submitted to a baking which expels the volatile parts.* Tapioca is the pure fecula without the mixture of the tissues which still exist in the cassava. In the potato the outer pellicle contracts noxious quali- ties when it is allowed to become green by exposure to the light, and it is well known that unripe or diseased tubers, containing too small a propertion of fecula with ^ Pohl, Plantanim Brasilice Icones et Descriptioyies, in fol., vol. i. * J. Miiller, in Prodromus, xv., sect. 2, pp. 1062-1064. ' Sagot, Bull, de la Soc. Bot. de France, Dec. 8, 1871. * I give the essentials of the preparation ; the details vary accorcling to the country. See on this head : Aublet, Guyane, ii. p. 67 j De- courtilz, Flora des Antilles, iii. p. 113 ; Sagot, etc. PLANTS CULTIVATED FOPw THEIR SUBTERRANEAN PARTS. Gl much sap, are not good to eat, and would cause positive harm to persons who consumed any quantity of them. All potatoes, and probably all maniocs, contain something harmful, which is observed even in the products of dis- tillation, and which varies with several causes ; but only matter foreign to the fecula should be mistrusted. The doubts about the number of species into which the cultivated manihots should be divided are no source of difficulty regarding the question of geographic origin. On the contrary, we shall see that they are an important means of proving an American origin. The Abbe Raynal had formerly spread the erroneous opinion that the manioc was imported into America from Africa. Robert Brown ^ denied this in 1818, but without giving reasons in support of his opinion ; and Humboldt,^ Sloreau de Jonnes,^ and Saint Hilaire * insisted upon its American origin. It can hardly be doubted for the following reasons : — 1. Maniocs were cultivated by the natives of Brazil, Guiana, and the warm region of Mexico before the arrival of the Europeans, as all early travellers testify. In the West Indies this cultivation was, according to Acosta,^ common enough in the sixteenth century to inspire the belief that it was also there of a certain antiquity. 2. It is less widely diffused in Africa, especially in regions at a distance from the west coast. It is known that manioc was introduced into the Isle of Bourbon by the Governour Labourdonnais.^ In Asiatic countries, where a plant so easy to cultivate would probably have spread had it been long known on the African continent, it is mentioned here and there as an object of curiosity of foreign origin J * E. Brown, Botany of the Congo, p. 50. ' Humboldt, Nouvelle Espagne, edit. 2, vol, ii. p. 398, ' Hist, de I' Acad, des Science.?, 1824. * Guillemin, Archives de Botanique, i. p. 239. ' Acosta, Hi.'it. Nat. des hides, French trans., 1598, p. 163. * Thomas, Statistique de Bourhon, ii. p. 18. ^ The catalogue of the botanical gardens of Buitenzorg, 1806, p. 222, says expressly that the Manihot utilissima comes from Bourbon and Amei'ica. 62 ORIGIN OF CULTIVATED PLANTS. 3. The natives of America had several ancient names for the varieties of manioc, especially in Brazil/ which does not appear to have been the case in Africa, even on the coast of Guinea.^ 4. The varieties cultivated in Brazil, in Guiana, and in the West Indies are very numerous, whence we may presume a very ancient cultivation. This is not the case in Africa. 5. The forty-two known species of the genus Manihot, without counting M. idilissima, are all wild in America ; most of them in Brazil, some in Guiana, Peru, and Mexico; not one in the old world.^ It is very unlikely that a single species, and that the cultivated one, was a native both of the old and of the new world, and all the more so since in the family Eujyhorhiaceai the area of the woody species is usually restricted, and since phanerogamous plants are very rarely common to Africa and America. The American origin of the manioc being thus established, it may be asked how the species has been introduced into Guinea and Congo. It was probably the result of the frequent communications established in the sixteenth century by Portuguese merchants and slave-traders. The Manihot ut'iUsshna and the allied species or variety called aipi, which is also cultivated, have not been found in an undoubtedly wild state. Humboldt and Bonpland, indeed, found upon the banks of the Magdalena a plant of Manihot utilissima which they called almost wild,"* but Dr. Sagot assures me that it has not been found in Guiana, and that botanists who have explored the hot region in Brazil have not been more fortunate. We gather as much from the expressions of Pohl, who has carefully studied these plants, and who was acquainted with the collections of Martins, and had * Aypi, viandioca, manihot, manioch, yxica, etc., in Pohl, Icones and Besc, i. pp. 30, 33. Martius, Beitrdje z. Ethnographie, etc., Braziliens, ii. p. 122, gives a number of names. ' Thonuing (in Schumacher, BesJc. Guin.), who is accustomed to quote the common names, gives none for the manioc. » J. Miiller, in Profh-nmus, xv., Beet. 1, p. 1057. ♦ Kuuth, in Humboldt and B., Nova Genera, ii. p. 108. PLANTS CULTIVATED FOK THEIR SUBTERRANEAN PARTS. 63 no doubt of their American oris^in. If he had observed a wild variety identical with those which are cultivated, he would not have suggested the hypothesis that the manioc is obtained from his Manihot pusilla^ of the province of Goyaz, a plant of small size, and considered as a true species or as a variety of Manihot palmata.^ Martins declared in 1867, that is after havincr received a quantity of information of a later date than his journey, that the plant was not known in a wild state.^ An early traveller, usually accurate, Piso,^ speaks of a wild mandi- hoca, of which the Tapuyeris, the natives of the coast to the north of Rio Janeiro, ate the roots. "It is," he says, " very like the cultivated plant ; " but the illustra- tion he gives of it appears unsatisfactory to authors who have studied the maniocs. Pohl attributes it to his 31. aijn, and Dr. Miiller passes it over in silence. For my part, I am disposed to believe what Piso says, and his figure does not seem to me entirely unsatisfactory. It is better than that by Vellozo, of a wild manioc which is doubtfully attributed to M, aipi} If we do not accept the origin in eastern tropical Brazil, we must have recourse to two hypotheses : either the cultivated maniocs are obtained from one of the wild species modified by cultivation, or they are varieties which exist only by the agency of man after the disappearance of their fellows from modern wild vesretation. Garlic — Allium sativum^ Linnaeus. Linn?eus, in his Species Plantaruni, indicates Sicily as the home of the common garlic; but in his Hortiis Clifortianus, v;here he is usually more accurate, he does not give its origin. The fact is that, according to aU the most recent and complete floras of Sicily, Italy, Greece, France, Spain, and Algeria, garlic is not considered to be indigenous, although specimens have been gathered here and there which had more or less the appearance of * Pohl, Icones et Descr., i. p. 36, pi. 2(5. ^ Miiller, in Prodromus. ' De Martius, Beitrdge zur Ethnographie, etc., i. pp. 19, 136. * Piso, Historia Nahiralis Brazilice, in folio, 1658, p. 55, cum icone. ^ Jairopia Sylvestris Veil. Ft. Flum., 16, t. 83. See Miiller, in D. C. Prodromus, xv. p. 1063. 64 ORIGIN OF CULTIVATED PLANTS. being so. A plant so constantly cultivated and so easily propagated may spread from gardens and persist for a considerable time without being wild by nature. I do not know on what authority Kunth ^ mentions that the species is found in Egypt. According to authors who are more accurate^ in their accounts of the plants of that country, it is only found there under cultivation. Boissier, whose herbarium is so rich in Eastern plants, possesses no wild specimens of it. The only country where garlic has been found in a wild state, with the certainty of its reall}^ being so, is the desert of the Kirghis of Sungari ; bulbs were brought thence and cultivated at Dorpat,^ and specimens were afterwards seen by Regel.^ The latter author also says that he saw a specimen which Wallich had gathered as wild in British India ; but Baker,^ who had access to the rich herbarium at Kew, does not speak of it in his review of the "AU'lidus of India, China, and Japan." Let us see whether historical and philological records confirm the fact of an origin in the south-west of Siberia alone. Garlic has been long cultivated in China under the name of suan. It is written in Chinese by a single sign, which usually indicates a long known and even a wild species.^ The floras of Japan ' do not mention it, whence 1 gather that the species was not wild in Eastern Siberia and Dahuria, but that the Mongols brought it into China. According to Herodotus, the ancient Egyptians made great use of it. Archaeologists have not found the proof of this in the monuments, but this may be because the plant was considered unclean by the priests.^ ' KuTilh, Eninn., W. p. 381. 2 Sclnveinfurth and Ascherson, Aufzdhlunrj, p. 294. ' Ledebour, Flora Altaica, ii. p. 4 ; Flora Bossica, iv. p. 1G2. * Kegel, Allior. Monogr., p. 44, * Baker, in Journal of BoL, 1874, p. 295. * Bretschneider, Study and Value, etc., pp. 15, 4, and 7. ^ Thunbcrg, Fl. Jap. ; Francliet and Savatier, Enumeratio, lb7G, vol. ii. * Ungcr, PJlanzen dcs Alien Mgyptens, p. 43. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 65 There is a Sanskrit name, maJioushouda,^ become loshoun in Bengali, and to which appears to be related the Hebrew name schoum or schumin,^ which has pro- duced the Arab thoum or toum. The Basque name hara- tchouria is thought by de Charencey ^ to be allied with Aryan names. In support of his hypothesis I may add that the Berber name, tiskert, is quite different, and that consequently the Iberians seem to have received the plant and its name rather from the Aryans than from their probable ancestors of Northern Africa. The Lettons call it /vi2^Zo/J>;s,theEsthonians krunslauk,\YhenQe probably the German Knoblauch. The ancient Greek name appears to have been scorodon, in modern Greek scordon. The names given by the Slavs of lUyria are hill and cesan. The Bretons say quinen,'^ the Welsh craf, cenhinneii, or garlleg, whence the English garlic. The Latin cdlmm has passed into the languages of Latin origin.^ This great diversity of names intimates a long acquaintance with the plant, and even an ancient cultivation in Western Asia and in Europe. On the other hand, if the species has existed only in the land of the Kirghis, where it is now found, the Aryans might have cultivated it and carried it into India and Europe ; but this does not explain the existence of so many Keltic, Slav, Greek, and Latin names which differ from the Sanskrit. To explain this diversity, we must suppose that its original abode extended farther to the west than that known at the present day, an extension anterior to the migrations of the Aryans. If the genus Allium were once made, as a whole, the object of such a serious study as that of Gay on some * Piddington, Index. * Hiller, Hierophyton; Rosenmiiner, Bihl. Alierthuw, vol. iv", 3 De Charencey, Actes de la Soc. Phil., 1st March, lbG9. * Davies, Welsh Botanology . * All these common names are found in my dictionary compiled by .'\r()ritzi from floras. I could have quoted a larger number, and men- tioned the probable etymologies, as given by philologists — Hehn, for instance, in his Kulturpflanzen aus Aaien, p. 171 and following; but this is not necessary to show its origin and early cultivation in several different countries. 66 ORIGIN OF CULTIVATED PLANTS. of its species,^ perhaps it might be found that certain wild European forms, included by authors under A. arenariura, L,, A. arenariioii, Sm., oy A. scorodoprasiim, L., are only varieties of A. sativum. In that case every- thing would agree to show that the earliest peoples of Europe and Western Asia cultivated such form of the species just as they found it from Tartary to Spain, mvino' it names more or less different. Onion — Allinin Cepa, Linnseus. I will state first Avhat was known in 1855;^ I will then add the recent botanical observations which confirm the inferences from philological data. The onion is one of the earliest of cultivated species. Its original country is, according to Kunth, unknown.^ Let us see if it is possible to discover it. The modern Greeks call Allium Cepa, which they cultivate in abundance, krommunda.^ This is a good reason for be- lieving that the kromonuon of Theophrastus ^ is the same species, as sixteenth-century writers already supposed.^ Pliny "^ translated the word by ccepa. The ancient Greeks and Romans knew several varieties, which they distin guished by the names of countries : Cypriiim, Cretense, Samothraciae, etc. One variety cultivated in Egypt ^ was held to be so excellent that it received divine honours, to the grreat amusement of the Romans.^ Modern Egyptians designate A. Cepa by the name of hasal ^^ or hussul,^^ whence it is probable that the hezalim of the Hebrews is the same species, as commentators have said.-^"^ There are several distinct names — 2''^'-^^'^^^^^''^ latarka, sa- handaka^^ and a number of modern Indian names. The species is commonly cultivated in India, Cochin-China, * Annales des Sc. Nat., Srd series, vol. viii. ' A. de Candolle, Geogr. Bot.Paiisonnee, ii. p. 823. ' Kunth, Enumer., iv. p. 394. * Fraas, Syn. Fl. Class., p. 291. * Theophrastus, Hist., 1. 7, c. 4. « J. Bauhin, Hist., ii. p. 548. ' Pliny, Hist., 1. 19, c 6. * Ihid. ^ Juvenalis, Sat. 15. *" Foi'skal, p. 65. " Aiuslie's Mat. Med. Ind., i. p. 269. '^ Hiller, Hieroph., ii. p. 36 j Rosenmiiller, Handhh. Bill. Alterlc, iv. p. 96. '^ Piddington, Index ; Ainslie's Mat. Med. Ind. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 67 China,' and even in Japan.^ It was largely consumed by the ancient Egyptians. The drawings on their monuments often represent this species.^ Thus its cultivation in Southern Asia and the eastern region of the Mediterranean dates from a very early epoch. More- over, the Chinese, Sanskrit, Hebrew, Greek, and Latin names have no apparent connection. From this last fact we may deduce the hypothesis that its cultivation was begun after the separation of the Indo-European nations, the species being found ready to hand in different countries at once. This, however, is not the present state of things, for we hardly find even vague indications of the wild state of A. Cepa. 1 have not discovered it in European or Caucasian floras ; but Hasselquist * says, " It grows in the plains near the sea in the environs of Jericho." Dr. Wallich mentioned in his list of Indian plants, No. 5072, specimens which he saw in districts of Bengal, without mentioning whether they were cultivated. This indication, however insuflicient, together with the antiquity of the Sanskrit and Hebrew names, and the communication which is known to have existed between the peoples of India and of Egypt, lead me to suppose that this plant occupied a vast area in Western Asia, extending perhaps from Palestine to India. Allied species, sometimes mistaken for A. Cejm, exist in Siberia.^ The specimens collected by Anglo-Indian botanists, of which Wallich gave the first idea, are now better known. Stokes discovered Allium Cejoa wild in Beluchistan. He says, " wild on the Chehil Tun." Griffith brought it from Afghanistan and Thomson from Lahore, to say nothing of other collectors, who are not explicit as to the wild or cultivated nature of their specimens.^ Boissier possesses a wild specimen found in the mountainous regions of the Khorassan. The umbels are smaller than in the * Roxburgh, Fl. Ind., ii. ; Loureiro, Fl. Cochin , p. 249. * Thunberg, Fl. Jap., p. 132. ' Unger, Pfianzen d. Alt. JEgypt., p. 42, figs. 22, 23, 21. * Hasselquist, Voy. and Trav., p. 279. ' Ledebour, Fl. Rossica, iv. p. 169. * Aitchison, A Catalogue of the Plants of the Punjab and the Sindh, in 8vo, 1869, p. 19; Baker, in Journal ofBot., 1874, p. 295. G8 ORIGIN OF CULTIVATED PLANTS. cultivated plant, but there is no other difference. Dr. Kegel, jun., found it to the south of Kuldscha, in Western Siberia.-^ Thus my former conjectures are completely justified ; and it is not unlikely that its habitation extends even as far as Palestine, as Hasselquist said. The onion is designated in China by a single sign (pronounced tsimfj), which may suggest a long existence there as an indigenous plant.^ I very much doubt, how- ever, that tlie area extends so far to the east. Humboldt^ says that the Americans have always been acquainted with onions, in Mexican xonacatl. " Cortes, " he says, " speaking of the comestibles sold at the market of the ancient Tenochtillan, mentions onions, leeks, and garlic." I cannot believe, howev^er, that these names applied to the species cultivated in Europe. Sloane, in the seventeenth centurv, had onlv seen one Allmm cultivated in Jamaica (^i. Cepa), and tliat was in a garden with other European vegetables.^ The word xonacatl is not in Hernandez, and Acosta^ says distinctly that the onions and garlics of Peru are of European origin. The species of the genus Allium are rare in America. Spring, or Welsh Onion — Allhun fistudosum, Linnreus. This species was for a long time mentioned in floras and works on horticulture as of unknown origin ; but Russian botanists have found it wild in Siberia towards tlie Altai mountains, on the Lake Baikal in the land of the Kirghis.^ The ancients did not know the plant.' It must have come into Europe through Pussia in the Middle Ages, or a little later. Dodoens,^ an author of the sixteenth century, has given a figure of it, hardly recognizable, under the name of Cepa oblonga. Shallot — Allium ascalonicum, Linnaeus. It was believed, according to Pliny ,^ that this plant » ni. Hortic, 1877, p. 167. ' Bretscbneider, Study and Vahie, etc., pp. 47 and 7. ' Nouvelle Espagne, 2nd edit., ii. p. 476. * Sloane, Jam., i. p. 75. * Acosta, Hist. Nat. des Indes, French trans., p. 1G5. * Ledebour, ^lora Rossica, iv. p. 169. * Lenz, Botanik. der Alten Griechen vnd Bomer, p. 205. " Dodoens, Bemptadcs, p. 687. ' Plinj, Uist., 1. 19, o. 6. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. G9 took its name from Ascalon, in Judaea; but Dr. Fournier^ thinks that the Latin author mistook the meaning of the word Askalonion of Theophrastus. However this may be, the word has been retained in modern languages under the form o^echalote in French, chalote in Spanish, scalogno in Italian, Aschaluch or Eschlaucli in German. In 1855 I had spoken of the species as follows : ^ — "According to Roxburgh,^ Alliutn asccdonicuin is much cultivated in India. The Sanskrit name pidanclw is attributed to it, a word nearly identical with palandu, attributed to A. Cepa.^ Evidently the distinction be- tween the two species is not clear in Indian or Anglo- Indian works. " Loureiro says he saw Allium asccdonicum cul- tivated in Cochin-China,^ but he does not mention China, and Thunberg does not indicate this species in Japan. Its cultivation, therefore, is not universal in the east of Asia. This fact, and the doubt about the Sanskrit name, lead me to think that it is not ancient in Southern Asia. Neither, in spite of the name of the species, am I convinced that it existed in Western Asia. Rauwolf, Forskal, and Delile do not mention it in Siberia, in Arabia, or in Egypt. Linnaeus ^ mentions Hasselquist as having found the species in Palestine. Unfortunately, he gives no details about the localit}'', nor about its wild condition. In the Travels of Hasselquist ^ I find a Ce2xo "inontana mentioned as growing on Mount Tabor and on a neighbour- ing mountain, but there is nothing to prove that it was this species. In his article on the onions and garlics of the Hebrews he mentions only AUiaon Cepa, then A. porrum and A. sativuon. Sibthorp did not find it in Greece,^ and Fraas ^ does not mention it as now cultivated * He will treat of this in a publication entitled Ciharia, whicb will shortl}'^ appear. 2 Geog. Bof. Raisonnee, p. 829. 3 Roxburgh, FL hid.) edit. 1832, vol. ii. p. 142. ■* Piddington, Index. ^ Loureiro, Fl. Cochin., p. 251. * Lmnceus, Species, p. 429. ^ Hasselquisr, Voy. and Trav., 17^6, pp. 281, 282. 8 Sibthorp, Frodr, " Fraas, Sjn. Fl. Class., p. 291. 70 ORIGIN OF CULTIVATED PLANTS. in tliat country. According to Kocli,^ it is naturalized among the vines near Fiume. However, Viviani ^ only speaks of it as a cultivated plant in Dalmatia. "From all these facts I am led to believe that Allium ascaloniciim is not a species. It is enough to render its primitive existence doubtful, to remark : (1) that Theophrastus and ancient Avriters in general have spoken of it as a form of the Alliiini Cejxc, having the same importance as the varieties cultivated in Greece, Thrace, and elsewhere ; (2) that its existence in a wild state cannot be proved ; (3) that it is little cultivated, or not all, in the countries Avhere it is supposed to have had its origin, as in Syria, Egypt, and Greece ; (4-) that it is commonly without flowers, whence the name of Cejxc sterilis given by Bauhin, and the number of its bulbs is an allied fact; (5) when it docs flower, the organs of the flower are similar to those of A. Cejxi, or at least no difference has been hitherto discovered, and according to Koch ^ the only diflerence in the whole plant is that the stalk and leaves are less swelled, although fistulous." Such was formerly my opinion.^ The facts published since 1855 do not destroy my doubts, but, on the contrary, justify them. Kegel, in 1875, in his monograph of the genus Allium, declares he has only seen the shallot as a cultivated species. Aucher Eloy has distributed a plant from Asia Minor under the name of A. ascalonicum, but judging from my specimen this is certainly not the species. Boissier tells me that he has never seen A. ascalonicum in the East, and it is not in his herbarium. The plant from the Morea which bears this name in the flora of Bory and Chaubard is quite a different species, which he has named A. gomphrcnoides. Baker,^ in his review of the Alliums of India, China, and Japan, mentions A. ascalonicum in districts of Benofal and of the Punjab, from specimens of Griffith and Aitchison ; but he adds, "They are probably cultivated plants." * Koch, Svn. FJ. Germ., 2nd edit., p. 833. ' Viviani, Fl. Palmot., p. 138. ^ Koch, Ffyn. FJ, Germ. * A. dc Candollc, Gvogr. Bof. Raisnynu'c, p. 82*J. * Baker, in Joarn. of Bot., 1874, p. 200. PLANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 71 He attributes to A. ascalonicum Allium sulvia, Ham., of Nepal, a plant little known, and whose wild character is uncertain. The shallot produces many bulbs, which may be propagated or preserved in the neighbourhood of cultivation, and thus cause mistakes as to its origin. Finally, in spite of the progress of botanical investiga- tions in the East and in India, this form of Allium has not been found wild with certainty. It appears to me, therefore, more probable than ever that it is a modifica- tion of A. Cejpa, dating from about the beginning of the Christian era — a modification less considerable than many of those observed in other cultivated plants, as, for instance, in the cabbage. Rocambole — Allium scorodoprasum, Linnseus. If we cast a glance at the descriptions and names of A. scorodoiorasum in w^orks on botany since the time of Linngeus, we shall see that the only point on which authors are asrreed is the common name of rocam- hole. As to the distinctive characters, they sometimes approximate the plant to Alliiom sativum, sometimes recrard it as altoojether distinct. With such different definitions, it is difficult to know in what country the plant, well known in its cultivated state as the rocambole, is found wild. According to Cosson and Germain,^ it grows in the environs of Paris. According to Grenier and Godron,^ the same form grows in the east of France. Burnat says he found the species undoubtedly wild in the Alpes-Maritimes, and he gave specimens of it to Boissier. Willkomm and Lange do not consider it to be wild in Spain,^ though one of the French names of the cultivated plant is ail or eschcdote d'Espagne. Many other European localities seem to me doubtful, since the specific characters are so uncertain. I mention, however, that, according to Ledebour,* the plant which he calls A. scorodoprasum is very common in Russia from Fin- land to the Crimea. Boissier received a specimen of it ^ Cosson and Germain, Flore, ii. p. 553. 2 Grenier and Godron, Flore de France, iii. p. iri7. 8 Willkomm and Lange, Prodr. Fl. Hisp., i. p. SS5. * Ledebour, Flora Rossica, iv. p. 163. 72 ORIGIN OF CULTIVATED PLANTS. from Dobrutscha, sent by the botanist Sintcnis. The natural habitat of the species borders, therefore, on that of Allium sativum., or else an attentive study of all these forms will show that a single species, comprising several varieties, extends over a great part of Europe and the bordering countries of Asia. The cultivation of this species of onion docs not appear to be of ancient date. It is not mentioned by Greek and Roman authors, nor in the list of plants recommended by Charlemagne to the intendants of his gardens.^ Neither does Olivier de Serres speak of it. We can only give a small number of original common names among ancient peoples. The most distinctive are in the North. Shoiiug in Denmark, keipe and rachenholl in Sweden.^ liockoiholle, whence comes the French name, is German. It has not the meaninir oiven by Littre. Its etymology is Bolle, onion, growing among the rocks, Rochen.^ Chives — Allium schccnoprasum, Linnneus. This species occu])ies an extensive area in the northern hemisphere. It is found all over Europe, from Corsica and Greece to the south of Sweden, in Siberia as far as Kamtschatka, and also in Korth America, but only near the Lakes Huron and Superior and further north ^ — a remarkable circumstance, considering its Euro- pean habitat. Tlie variety found in the Alps is the nearest to the cultivated form.^ The ancient Greeks and Romans must certainly have known the species, since it is wild in Italy and Greece. Targioni believes it to be the Scorodon schiston of Theophrastus ; but we are dealing with words without descriptions, and authors whose specialty is the inter- pretation of Greek text, like Fraas and Lenz, are prudent enouoh to affirm nothin^^ If the ancient names are doubtful, the fact of the cultivation of the plant at this epoch is yet more so. It is possible that the custom of gathering it in the fields existed. * Le Grand d'Aussy, Histoire de la Tie des Franrais, vol. i. p. 122. ' Nemnich, PolygJott. Lexicon, p. 187. ' Ibid. * Asx Gray, Botany of the Northern Siafcf^, edit. 5, p. 534. ' De CaudoUe, Flo7-e Fravgaise, iv. p. 227. PLANTS CULTIVATED FOE THEIR SUBTERRANEAN PARTS. 73 Colocasia — Arum esculentum, Linnaeus; Colocasia antiquoi'iim, Schott.-'- This species is cultivated in the damp districts of the tropics, for the swelled lower portion of the stem, which forms an edible rhizome similar to the subterraneous part of the iris. The petioles and the young leaves are also utilized as a vegetable. Since the different forms of the species have been properly classed, and since we have possessed more certain information about the floras of the south of Asia, we cannot doubt that this plant is wild in India, as Roxburgh ^ formei'ly, and Wight ^ and others have more recently asserted ; likewise in Ceylon,^ Sumatra,^ and several islands of the Malay Archipelago.'^ Chinese books make no mention of it before a work of the year 100 B.c!^ The first European navigators saw it cultivated in Japan and as far as the north of New Zealand,^ in consequence probably of an early introduc- tion, and without the certain co-existence of wild stocks. When portions of the stem or of the tuber are thrown away by the side of streams, they naturalize themselves easily. This was perhaps the case in Japan and the Fiji Islands,^ judging from the localities indicated. The colocasia is cultivated here and there in the West Indies, and elsewhere in tropical America, but much less than in Asia or Africa, and without the least indication of an American orioin. In the countries where the species is wild there are common names, sometimes very ancient, totally difl'erent from each other, which confirms their local origin. Thus the Sanskrit name is kuchoo, which persists in modern * Aru7n Egijptium, Columma, Ecphrasis, u. p. 1, tab. 1; Rura- pbius, A^nhoin, vol. v. tab. 109. Arum colocasia and A. esculentum, Linnaeus; Colocasia antiquorum, Scbott, Melet., i. 18; Eiigler, in D. C. Moiiorj. Phaner., ii. p. 491. 2 Roxburgb, Fl. Ind., iii. p. 495. ' Wigbt, Icones, t. 786. * Thwaites, Enum. Plant. Zeylan., p. 335. * Miquel, Sumatra, p. 258. * Rumpbius, Amboin, vob v. p. 318. ^ Bretscbneider, On the Study and Value, etc., p. 12. * Forster, De Plantis EscuL, p. 58. * Franc-bet and Savatier, Enum,, p. 8; Seeuiann, Flora VitiensiSf p. 284. 74 OHIGIN OF CULTIVATED PLANTS. Hindu languages — in Bengali, for instance.^ In Ceylon the wild plant is styled gahala, the cultivated plant kandalla.^ The Malay names are kelady'^ tallits, tallas, tales, or taloes,^ from which perhaps comes the well- known name of the Otahitans and New Zealanders — tallo or tarro^ clalo^ in the Fiji Islands. The Japanese have a totallv distinct name, hno^ which shows an existence of loner duration either indif^enous or cultivated. European botanists first knew the colocasia in Egypt, where it has perhaps not been very long cultivated. The monuments of ancient Egypt furnish no indication of it, but Pliny ^ spoke of it as the Arum ^jijptiiun. Prosper Alpin saw it in the sixteenth century, and speaks of it at length.^ He says that its name in its country is culcas, which Delile^^ writes qolkas, and kovblkas. It is clear that this Arab name of the Egyptian arum has some analogy with the Sanskrit kiichoo, which is a confirmation of the hyjwthesis, sufficiently j)robable, of an introduction from India or Ceylon. De I'Eclusc^Miad seen the plant cultivated in Portui»:al, as introduced from Africa, under the name alcoleaz, evidently of Arab origin. In some parts of the south of Italy, where the plant has become naturalized, it is, according to Parlatore, called aro di Egitto}^ The name colocasia, given by the Greeks to a plant of which the root was used by the Egyptians, may evidently come from colcas, but it has been transferred to a plant differing from the true colcas. Indeed, Dioscorides applies it to the Egyptian bean, or neluinho}^ which has a large root, or rather rhizome, rather stringy » Roxbiargli, Fl. Lid. * Thwaites, Enum. Plant. Zeylan. ' Rnmpliius, Amhoin. * IMiquel, Sumatra, p. 258; Hasskarl, Cat. Hnrti. Bogor. A'ter., p. 55. * Forster, De Plantis Escul., p. 58. ® Seemann, Flora Vitiensis. ' Frauchet and Savatier, Enum. * Pliny, Hist., 1. 19, c. 5. ^ Alpiuns, Hist. Mgypt. Naturalis, edit. 2, vol. i. p. 166; ii. p. 192. " Delile, Fl. JEgypt. III., p. 28 ; De la Colocase des Anciens, in Svo, 1846. " Clusins, His^oria, ii. p. 75. " Parlatore, Fl. IfaL, ii. p. 255. *' Prosper Alpinus, ifi'si. ^gypt. Naturalis; Columna ; De\i\e, Ann. du Mus., i. p. 375; De la Colocase des Ancie)is } Keynier, Economie des Egyptiena, p. 321. PJ.ANTS CULTIVATED FOR THEIR SUBTERRANEAN PARTS. 75 and not good to eat. The two plants are very different, especially in the flower. The one belongs to the Aracece, the other to the Nymphceaceoe ; the one belongs to the class of Monocotyledons, the other to that of the Dico- tyledons. The nelumbo of Indian origin has ceased to grow in Egypt, while the colocasia of modern botanists has persisted there. If there is any confusion, as seems probable in the Greek authors, it must be explained by the fact that the colcas rarely flowers, at least in Egypt. From the point of view of botanical nomenclature, it matters little that mistakes were formerly made about the plants to which the name colocasia should be applied. Fortunately, modern scientific names are not based upon the doubtful definitions of the ancient Greeks and Romans, and it is sufiicient to say now, if the etymology is insisted upon, that colocasia comes from colcas in consequence of an error. Ape, or Large-rooted Alocasia — Alocasia macrorrhiza, Schott ; Arum niaGrorrJdzuni, Linmieus. This araceous plant, which Schott places now in the genus Colocasia, now in the Alocasia, and whose names are far more complicated than might be supposed from those indicated above,^ is less frequently cultivated than the common colocasia, but in the same manner and nearly in the same countries. Its rhizomes attain the length of a man's arm. They have a distinctly bitter taste, which it is indispensable to remove by cooking. The aborigines of Otahiti call it ape, and those of the Friendly Isles happe.^ In Ceylon, the common name is hahara, according to Thwaites.^ It has other names in the Malay Archipelago, which argues an existence prior to that of the more recent peoples of these regions. The plant appears to be wild, especially in Otahiti.* It is also wild in Ceylon, according to Thwaites, who has studied botany for a long time in that island. It is * See Engler, in D. C. Monographioe FTianerogariim, n. p. 502. * Eorstei', De Plantis Esculentis Insularum Oceani Ausiralis, p. 58. ' Thwaites, Enum. PI. Zeyl., p. 336. * Nadeaud, Enum. des Flantcs Indigenes^ p. 40. 7G ORIGIN OF CULTIVATED PLANTS. mentioned also in India ^ and xn Australia,^ but its wild condition is not affirmed — a fact always difficult to establish in the case of a species cultivated on the banks of streams, and which is propagated by bulbs. More- over, it is sometimes confounded with the Colocasia indica of Kunth, which grows in the same manner, and is found here and there in cultivated ground ; and this species grows wild, or is naturalized in the ditches and streams of Southern Asia, although its history is not yet well known. Konjak — AmorphophaUus Konjak, Koch ; Amor- lohopliallvis Eirieri, du Rieu, var. Konjak, Engler.^ The konjak is a tuberous plant of the family Aracece, extensively cultivated by the Japanese, a culture of which Vidal has given full details in the Bulletin de la Societe d'Acdlmatatioii of July, 1877. It is consi- dered by Engler as a variety of Amorphophallas lltvieri, of Cochin-China, of which horticultural periodicals have given several illustrations in the last few years.* It can be cultivated in the south of Europe, like the dahlia, as a curiosity ; but to estimate the value of the bulbs as food, they should be prepared with lime-water, in Japanese fashion, so as to ascertain the amount of fecula which a given area will produce. Dr. Vidal gives no proof that the Japanese plant is wild in that country. He supposes it to be so from the meaning of the common name, which is, he says, honni- yakou, or yamagonniyakou, yama meaning " mountain." Franchet and Savatier^ have only seen the plant in gardens. The Cochin-China variety, believed to belong to the same species, grows in gardens, and there is no proof of its being wild in the country. Yams — Di06Corea sativa, D. batatas, D. japonica, and D. alata. The yams, monocotyledonous plants, belonging to 1 Engler, in D. C. Monog. Phaner. 2 Bentham, Flora Austr., viii. p. 155. ' En?/es * in Albanian, a language derived from the Pelasgic ; without mentioning names drawn from the similarity of taste with that of the water- cress {Nastiirtiiim officinale). There are very distinct names in Hindustani and Bencrali, but none are known in Sanskrit.^ At the present day the plant is cultivated in Europe, in the north of Africa, in Eastern Asia, India, and else- where, but its origin is somewhat obscure. I possess several specimens gathered in India, wdiere Sir Joseph Hooker^ does not consider the species indigenous. Kotschy brought it back from Karrak, or Karek Island, in the Persian Gulf. The label does not say that it was a cultivated plant. Boissier '^ mentions it without com- ment, and he afterwards speaks of specimens from Ispahan and Egypt gathered in cultivated ground. Olivier is quoted as having found the cress in Persia, but it is not said whether it was growing w^ild.^ It has been asserted that Sib thorp found it in Cyprus, but reference to his work shows it was in the fields.^ Poech does not mention 1 See Fraas, Syn. FI. Class., pp. 120, 124 ; Lenz, Bot. der Alien, p. G17. 2 Sibthorp, Prodr, Ft. Groec.y ii. p. 6 ; Heldreich, Nutzpjl. GriechenL, p. 47. 2 Ainslie, Mat. Med. Ind., i. p. 95. * Heldreich, Nutz. Gr. ^ Piddiugton, Index ; Ainslie, Mat. Med. Ind., i. p. 95. '^ Hooker, Fl. Brit. Ind., i. p. 160. ^ Boissier, Fl. Orient. j vol. i. 8 De Candolle, Syst., ii. p. 533. ® Sibthorp and Smith, Prodr. FL Grcccae, ii. p. 6. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 87 it in Cyprus.^ Unger and Kotschy ^ do not consider it to be wild in that island. According to Ledeboiir,^ Koch found it round the convent on Mount Ararat; Pallas near Sarepta; Falk on the banks of the Oka, a tributary of the Volga ; lastly, H. Martins mentions it in his flora of Moscow ; but there is no proof that it was wild in these various localities. Lindeniann,* in 18G0, did not reckon the species among those of Russia, and he only indicates it as cultivated in the Crimea.^ According to Nyman,^ the botanist Schur found it wild in Transylvania, while the Austro-Hungarian floras either do not mention the species, or give it as cultivated, or growing in cultivated ground. I am led to believe, by this assemblage of more or less doubtful facts, that the plant is of Persian origin, whence it may have spread, after the Sanskrit epoch, into the gardens of India, Syria, Greece, and Egypt, and even as far as Abyssinia.'^ Purslane — Portulaca oleracea, Linnaeus. Purslane is one of the kitchen garden plants most widely diffused throughout the old world from the earliest times. It has been transported into America,^ where it spreads itself, as in Europe, in gardens, among rubbish, by the wayside, etc. It is more or less used as a vege- table, a medicinal plant, and is excellent food for pigs. A Sanskrit name for it is known, lonica or lounia, which recurs in the modern languages of India.^ The * Poech, Enum. PI. Cypri, 1842. ' Unj^er and Kotschy, Inseln Cypern., p. 331. » Ledebour, Fl. Eoss., i. p. 203. * LindemsLTin, Index Plant. inRoss. , Bull. Soc.Nat.Mosc.18GO, vol. xxxiii. * Lindemann, Prodr. Fl. Cherson, p. 21. « Nyman, Conspectus Fl. Eur op., 1878, p. 65. » Schweinfurth, Beitr. Fl. Mh., p. 270. ' In the United States purslane was believed to be of foreign origin (Asa Gray, Fl. ofNorthern States, ed. 5; Bot. of California, i. p. 79), but in a recent publication, Asa Gray and Trumbull give reasons for believing that it is indigenous in America as in the old world. Columbus had noticed it at San Salvador and at Cuba ; Oviedo mentions it in St. Domingo and De Lery in Brazil. This is not the testimony of botanists, but Nuttall and others found it wild in the upper valley of the Missouri, in Colorado, and Texas, where, however, from the date, it might have been introduced. — Author's Note, 1884. * Piddington, Index to Indian Plants. 88 ORIGIN OF CULTIVATED PLANTS. Greek name andracJtne and the Latin portidaca are very different, as also the group of names, cholza in Per- sian, kliursa or koursa in Hindustani, kourfa kara-or in Arab and Tartar, which seem to be the orig^in of Jcurza noka in Polish, kiirj-noha in Bohemian, Kreusel in Ger- man, without speaking of the Russian name schrucJia, and some others of Eastern Asia.^ One need not be a philologist to see certain derivations in these names show- ing that the Asiatic peoples in their migrations trans- ported with them their names for the plant, but this does not prove that they transported the plant itself They may have found it in the countries to which they came. On the other hand, the existence of three or four different roots shows that European peoples anterior to the Asiatic migrations had already names for the species, which is consequently very ancient in Europe as well as in Asia. It is very difficult to discover in the case of a plant so widely diffused, and which propagates itself so easily by means of its enormous number of little seeds, whether a specimen is cultivated, naturalized by spreading from cultivation, or really wild. It does not appear to be so ancient in the east as in the Avest of the Asiatic continent, and authors never say that it is a wild plant.^ In India the case is veiy different. Sir Joseph Hooker says^ that it gi'ows in India to the height of five thousand feet in the Himalayas. He also mentions having: found in the north-west of India the variety with upright stem, which is cultivated together with the common species in Europe. I find nothing positive about the localities in Persia, but so many are mentioned, and in countries so little cultivated, on the shores of the Caspian Sea, in the neighbourhood of the Caucasus, and even in the south of Russia,^ that it is difficult not to admit that the plant is indigenous in that central region whence the Asiatic peoples overran * Neranicb, Polyglot. Lex. Nahirgesch., ii. p. 1047. ^ Loureiro, FL Cochin., i. p. 359 ; Franchet and Savatier, Enum. PI. Japon., i. p. 53 ; Bentham, Fl. Hongkong, p. 127. 3 Hooker, Fl. Brit. Ind., i. p. 240. * Ledebour, Fl. Ross., ii. p. 145; Lindemann, in Prodr. Fl. Chers., p. 74, says, " In desertis et arenosis inter Chersou et Berislaw, circa Odessam." PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 80 Europe. In Greece the plant is wild as well as culti- vated.^ Further to the west, in Italy, etc., we begin to find it indicated in Horas, but only growing in fields, gardens, rubbish-heaps, and other suspicious localities.^ Thus the evidence of philology and botany alike show that the species is indigenous in the whole of the region which extends from the western Himalayas to the south of Russia and Greece. New Zealand Spinach — Tetragonia expansa, Murray. This plant was brought from New Zealand at the time of Cook's famous vo3^age, and cultivated by Sir Joseph Banks, and hence its name. It is a singular plant from a double point of view. In the first place, it is the only cultivated species which comes from New Zealand ; and secondly, it belongs to an order of usually fleshy plants, the Ficoidece, of which no other species is used. Hor- ticulturists ^ recommend it as an annual vesfetable, of which the taste resembles that of spinach, but which bears drought better, and is therefore a resource in seasons when spinach fails. Since Cook's voyage it has been found wild chiefly on the sea coast, not only in New Zealand but also in Tas- mania, in the south and west of Australia, in Japan, and in South America.^ It remains to be discovered whether in the latter places it is not naturalized, for it is found in the neighbourhood of towns in Japan and Chili.^ Garden Celery — Ajnum graveolens, Linnseus. Like many Umbellifers which grow in damp places, wild celery has a wide range. It extends from Sweden to Algeria, Egypt, Abyssinia, and in Asia from the Caucasus to Beluchistan, and the mountains of British India.^ * Lenz, Bot. der Alten, p. 632 ; Heldreich, Fl. Attisch. Ehene., p. 483. * Bertoloni, Fl. It., vol. v. ; Gussone, Fl. Sic, vol. i. ; Moris, Fi. Sard., vol. ii. ; Willkomm and Lange, Prodr. Fl. Hisp., vol. iii. ' Botanical Magazine, t. 23G2 ; Bon Jardinier, 1880, p. 567- ■* Sir J. Hooker, Hayidhnok of New Zealand Flora, p. 8-1 ; Benthani, Flora Ausiraliensis, iii. p. 327 ; Franchet and Savatier, Enum. Plant. Japonice, i. p. 177. * CL Gay, Flora Chilena, ii. p. 468. ' Fries, Summa Veget. Scand. ; Munby, Catal. Alger.,-p. 11; Boissier, Fl. Orient., vol. ii. p. 856 ; Schweinfurth and Asclierson, Aitfzdhlung, p. 272; Hooker, Fl. Brit. Ind., ii. p. 679. 90 ORIGIN OF CULTIVATED PLANTS. It is spoken of in the Odyssey tinder the name of selinon, and in Theophrastiis ; but later, Dioscorides and Pliny ^ distinguish between the wild and cultivated celery. In the latter the leaves are blanched, which greatly diminishes their bitterness. The long course of cultivation explains the numerous garden varieties. The one Avliich diti'ers more widely from the wild plant is that of which the fleshy root is eaten cooked. Chervil — Scandix cerefoliam, Linnaeus; Anthriscus cerefolium, Hoffmann. Not Ions: aofo the orij^in of this little Umbellifer, so com- mon in our gardens, was unknown. Like many annuals, it sprang up on rubbish-heaps, in hedges, in waste places, and it was doubted whether it should be con- sidered wild. In the west and south of Europe it seems to have been introduced, and more or less naturalized ; but in the south-east of Russia and in western tempei*ate Asia it a})pears to be indigenous. Steven'^ tells us that it is found " here and there in tlie woods of the Crimea." Boissier ^ received several specimens from the provinces to the south of the Caucasus, from Turcomania and the mountains of the north of Persia, localities of which tlie species is probably a native. It is wanting in the floras of India and the east of Asia. Greek authors do not mention it. The first mention of the plant by ancient writers occurs in Columella and Pliny,* tliat is, at the beginning of the Christian era. It was then cultivated. Pliny calls it cerefolium. The species was probably introduced into the Greco-Roman world after the time of Thcophrastus, that is in the course of the three centuries which preceded our era. Parsley — Pefrosdinuvi sativum, Moench. This biennialUmbellifer is wild in the south of Europe, from Spain to Turkey. It has also been found at Tlemcen in Alireria, and in Lebanon.^ ' Dioscorides, Mat Med., \. 3, c. 67, G8 ; Pliny, Hist., 1. 19, c. 7, 8 ; Lenz, Bot. der Alien Griechen xind PCower, p. 557. 2 Steven, Verzeichniss Taurischen Halhinseln, p. 183. » Boissier, Fl. Orient, ii. p. 913. * Lenz, Bot. d. Alt. Gr. und R., p. 572. • Muuby, Catal. Alger., edit. 2, p. 22 j Boissier, Fl. Orient., ii. p. 857. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 91 Dioscoricles and Pliny speak of it under the names of Petroselinon and Petroselinum} but only as a wild medicinal plant. Nothing proves that it was cultivated in their time. In the Middle Ages Charlemagne counted it among the plants which he ordered to be cultivated in his gardens.^ Olivier de Serres in the sixteenth century cultivated parsley. English gardeners received it in 1548.^ Although this cultivation is neither ancient nor important, it has already developed two varieties, which would be called species if they were found wild ; the parsley with crinkled leaves, and that of which the fleshy root is edible. Smyrnium, or Alexanders — Svxyrniurn ohis-atruniy Linn?eus. Of all the Umbellifers used as vegetables, this was one of the commonest in gardens for nearly fifteen centuries, and it is now abandoned. We can trace its bes^inuincr and end. Theophrastus spoke of it as a medicinal plant under the name of Ij^poselinon, but three centuries later Dioscorides * says that either the root or the leaves might be eaten, which implies cultivation. The Latins called it olus-atrmn, Charlemagne olisatuni, and com- manded it to be sown in his farms.^ The Italians made great use of it under the name onacerone.^ At the end of the eighteenth century the tradition existed in Eng- land that this plant had been formerly cultivated ; later Enojlish and French horticulturists do not mention it."^ The Smyrnium olns-atruni is wild throughout Southern Europe, in Algeria, Syria, and Asia Minor.^ Corn Salad, or Lamb's Lettuce — Valerianella olitoria, Linnaeus. > Dioscorides, 3Tat. Med., \. 3, c. 70 ; Pliny, Hht., 1. 20, ch. 12. ^ The list of these plants may be found in Meyer, Gesch. der Bot., iii. p. 401. ^ Phillips, Companion to the Kitchen Garden, ii. p. 35. * Theophrastus, Hist., 1. 1, 9 ; 1. 2, 2 ; 1. 7, 6 ; Dioscorides, Mat. Med., 1. 3, c. 71. * E. Meyer, Gesch. der Bot., iii. p. 401. ^ Targioni, Cenni Storici, p. 58. ' English Botany, t. 230; Phillips, Companion to the Kitchen Garden; Le Bon Jardinier. * Boissier, Fl. Orient., ii. p. 927. 02 ORIGIN OF CULTIVATED PLANTS. Frequently cultivated as a salad, this annual, of the Valerian family, is found wild throughout temperate Europe to about the sixtieth degree of latitude, in Southern Europe, in the Canary Isles, Madeira, and the Azores, in the north of Africa, Asia Minor, and the Caucasus.^ It often grows in cultivated ground, near villages, etc., Avhicli renders it somewhat ditticult to know where it grew before cultivation. It is mentioned, iKnvever, in Sardinia and Sicily, in the meadows and mountain pastures.^ I suspect that it is indigenous only in these islands, and that everywhere else it is introduced or naturalized. The grounds for this opinion are the fact that no name which it seems possible to assign to this l)lant has been found in Greek or Latin authors. We cannot even name any botanist of the Middle Ages or of the sixteenth century who has spoken of it. Neither is it mentioned anionic the veci'etables used in France in o o the seventeenth century, either by the Jardinier Fraugais of 1G'31, or by Laurenberg's \vovk, Hurt icidt lira (Frankfurt, 1G32). The cultivation and even the use of this sahid a2^2:)ear to be modern, a fact which has not been noticed. Cardoon — Cynara carduncidus, Linniiius. Artichoke — Cynara scolymua, Linnaeus; C. cardun- cidus, var. sativa, Moris. For a long time botanists have held the opinion that the artichoke is probablv a form obtained bv cultivation from the wild cardoon.^ Careful observations have lately proved this hypothesis. Moris,"* for instance, having cul- tivated, in the garden at Turin, the wild Sardinian plant side by side with the artichoke, affirmed that true characteristic distinctions no Ioniser existed. Willkomm and Lange,^ who have carefvdly observed the plant in Spain, both wild and cultivated, share the * Krolc, Monographie des Valei-ianella, Stockholm, 186 i, p. 88; Boissier, FJ. Orieyit., iii. p. 101. * Bertoloui, Fl. ItaL, i. p. 185; Moris, Fl. Sard., ii. p. 31 i; Gussone, Synopsis Fl. SicuJoe, edit. 2, vol. i. p. 30. ' Dodoens, Hist. Pla^it., p. 724; Linnaeus, Species, p. 11^9; De Can* dolle, Prndr., vi. p. 620. * Moris, Fh^ra Sardoa, ii. p. 01. * Willkomm and Lange, Prodr. Fl. Hisp., ii. p. 180. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 93 same opinion. Moreover, tlie artichoke has not been found out of gardens ; and since the Mediterranean region, the home of all the Cynarce, has been thoroughly explored, it may safely be asserted that it exists nowhere wild. The cardoon, in which we must also include C. horrida of Sibthorp, is indigenous in Madeira and in the Canary Isles, in the mountains of Marocco near Mogador, in the south and east of the Iberian peninsula, the south of France, of Italy, of Greece, and in the islands of the Mediterranean Sea as far as Cyprus.^ Munby ^ does not allow G. cardunculus to be wild in Algeria, but he does admit Cynara hitmilis of Linnaeus, which is considered by a few authors as a variety. The cultivated cardoon varies a good deal with regard to the division of the leaves, the number of spines, and the size — diversities which indicate long cultivation. The Romans eat the receptacle which bears the flowers, and the Italians also eat it, under the name of girello. Modern nations cultivate the cardoon for the fleshy part of the leaves, a custom which is not yet introduced into Greece.^ The artichoke offers fewer varieties, which bears out the opinion that it is a form derived from the cardoon. Targioni,^ in an excellent article upon this plant, relates that the artichoke was brought from Naples to Florence in 1466, and he proves that ancient writers, even Athen?eus, were not acquainted with the artichoke, but only with the w^ild and cultivated cardoons. I must mention, however, as a sign of its antiquity in the north of Africa, that the Berbers have two entirely distinct names for the two plants : addad for the cardoon, taga for the artichoke.^ * Webb, Phyt. Canar., iii. sect. 2, p. 384 ; Ball, Spicilegium Fl. Maroc, p. 52i ; Willkomm and Lange, Pr. Fl. Hisp. ; Bertoloni, Fl. Ital., ix. p, 86 ; Boissier, Fl. Orient., iii. p. 357 ; Unger and Kotschy, Inseln Cypern. V. 246. 2 Munby, Cotal., edit. 2. ' Heldreich, Nxdzpflanzen Grieclienlands, p. 27. * Targioni, Cenni Storici, p. 52. ^ Diciionnaire Frangaia-Berhere, published by the Government, 1 vol. in 8vo. 94j origin of cultivated plants. It is believed that the hactos, Iclnara, and scolimos of the Greeks, and the cardicus of Roman horticulturists, were Cynara cardunculics,^ although the most detailed description, that of Theophrastus, is suffici(uitly confused. "The plant," he said, "grows in Sicily " — as it does to this day — "and," he added, "not in Greece." It is, therefore, possible that the plants observed in our day in that country may have been naturalized from cultivation. According to Athengeus,^ the Egyptian king Ptolemy Energetes, of the second century before Christ, had found in Libya a great quantity of wild kuiara, by which his soldiers had profited. Although the indigenous species was to be found at such a little distance, I am very doubtful whether the ancient Egyptians cultivated the cardoon or the artichoke. Pickering and Unger ^ believed they recognized it in some of the drawings on the monuments ; but the two figures which Uniier considers the most admis-^ible seem to me extremely doubtful. Moreover, no Hebrew name is known, and the Jews would probably have spoken of this vege- table had they seen it in Egypt. The diffusion of the species in Asia must have taken place somewhat late. There is an Arab name, Idrschuff or kersckouf, and a Persian name, kungJdr,'^ but no Sanskrit name, and the Hindus have taken the Persian word kiinjir,^ which shows that it was introduced at a late epoch. Chinese authors do not mention any Cynara.^ The cultivation of the artichoke was only introduced into England in 1548.'^ One of the most curious facts in the history of Cynara cardiuiculus is its naturalization in the present century over a vast extent of the Pampas of Buenos Ay res, where its abundance is a hindrance to travellers.^ * Theophrastns, Hist., 1. 6, c. 4; Pliny, Hist., \. 19, c. 8; Lenz, Bot. der Alten Griechen and Romer, p. 4S0. * Athenaeus, Deipn., ii. 84. ' Pickering, Chron. Arrangement, p. 71 ; Uugor, PJlanzen der Alten .^(jyptens, p. 40, figs. 27 and 28. * Ainslie, Mat. Med. Ind., i. p. 22. * Pirldington, Index. ^ BretscUneidcr, Study, etc, and Letters of 1881. ^ Phillips, Companion to the Kitchen Garden, p. 22. * Aug.de Saint Hilary, Planfes Jlemarkables du Bresil, liitrod., p. 58; Darwin, Aiiimals and Plants under Domestication, ii. p. 34. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 95 It is becoming equally troublesome in Chili.^ It is not asserted that the artichoke has anywhere been naturalized in this manner, and this is another sign of its artificial origin. Lettuce — Latuca Scariola, var. sativa. Botanists are agreed in considering the cultivated lettuce as a modification of the wild species called Latuca Scariola.^ The latter grows in temperate and southern Europe, in the Canary Isles, Madeira,^ Algeria,^ Abys- sinia/ and in the temperate regions of Eastern Asia. Boissier speaks of specimens from Arabia Petrea to Mesopotamia and the Caucasus.^ He mentions a variety with crinkled leaves, similar therefore to some of our garden lettuces, which the traveller Hausknecht brought with him from the mountains of Kurdistan. I have a specimen from Siberia, found near the river Irtysch, and it is now known with certainty that the species grows in the north of India, in Kashmir, and in Nepal.'^ In all these countries it is often near cultivated ground or among rubbish, but often also in rocky ground, clearings, or meadows, as a really wild plant. The cultivated lettuce often spreads from gardens, and sows itself in the open country. No one, as far as I know, has observed it in such a case for several genera- tions, or has tried to cultivate the wild L. Scariola, to see whether the transition is easy from the one form to the otlier. It is possible that the original habitat of the species has been enlarged by the difi'usion of cultivated lettuces reverting to the wild form. It is knoAvn that there has been a great increase in the number of culti- vated varieties in the course of the last two thousand * Cl. Gay, Flora Chilena, iv. p. 317. ' The author who has gone into this question most carefull}' is BischofF, in his Beitrdge zur Flora Deutschlands unci der Schweitz, p. ISl. See also Moris, Flo7-a Sardoa, ii, p. 530. ' AVebb, Phytogr. Canariensis, in. p. 422 ; Lowe, Flora of Madeira, p. 514. * Munby, Catal., edit. 2, p. 22, under the name of L. sylvestris. ' Schweiufurth and Ascherson, Aufzahlung, p. 285. « Boissier, Fl. Orient., iii. p. 809. ' Clarke, Comx>os. Indicoe, p. 203. 96 ORIGIN OF CULTIVATED PLANTS. years. Tlieoplirastus indicated three ; ^ le Bon Jardlnier of 1880 gives forty varieties existing in France. The ancient Greeks and Romans cultivated the lettuce, especially as a salad. In the East its cultivation possibly dates from an earlier epoch. Nevertheless it does not appear, from the original common names both in Asia and Europe, that this plant was generally or very anciently cultivated. There is no Sanskrit nor Hebrew name known, nor any in the reconstructed Aryan tongue. A Greek name exists, tridax ; Latin, latiica ; Persian and Hindu, /i:a/m; and the analogous Arabic form chiiss or chass. The Latin form exists also, slightly modified, in the Slav and Germanic languages,^ which may indicate either that the Western Aryans diffused the jilant, or that its culti- vation spread with its name at a later date from the south to the north of Europe. Dr. Bretschneider has confirmed my supposition ^ that the lettuce is not very ancient in China, and that it was introduced there from the West. He says that the first work in which it is mentioned dates from A.D. 600 to A.D. 900> Wild Chicory — Cichoriuvi Intyhus, Linnaeus. The wild perennial chicory, which is cultivated as a salad, as a vegetable, as fodder, and for its roots, which are used to mix with cofiee, grows throughout Europe, except in Lapland, in Marocco, and Algeria,^ from Eastern Europe to Afghanistan and Beluchistan,^ in the Punjab and Kashmir,'^ and from Russia to Lake Baikal in Siberia.^ The plant is certainly wild in most of these countries ; but as it often groAvs by the side of roads and fields, it is probable that it has been transported by man from its original home. This must be the case in India, for there is no known Sanskrit name. The Greeks and Romans employed this species wild ^ Theophrastns, 1. 7, c. 4. ^ Netnnich, Polygl. Lexicon, * A. de Caudolle, Gcogr. Bot. Uaisonnee, p. 843. * Bretschneider, Study and Value of Chinese Botanical WorTi?m?., vol. ii., and by Heer, p. 22. ^ Ball, Spirilegium Fl. Marocc, p. 380. 7 Munby, Catal., edit. 2, p. 7. 8 Rohlf, according to Cosson, Bulle. Soc. Bot. de Fr., 1875, p. 46. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 123 as far as England/ the Alps, and the Balkan Mountains ; and lastly, in Asia from the south of the Caucasus^ to Lebanon and Palestine.^ I do not find it mentioned in the Crimea, nor beyond the Caspian Sea. Let us now turn to the cultivation of flax, destined in most instances to furnish a textile substance, often also to yield oil, and cultivated among certain peoples for tlie nutritious properties of the seed. I first studied the question of its origin in 1855/ and with the following result : — It was abundantly shown that the ancient Egyptians and the Hebrews made use of linen stuffs. Herodotus afiirms this. Moreover, the plant may be seen figured in the ancient Egyptian drawings, and the microscope indubitably shows that the bandages which bind the mummies are of linen.^ The culture of flax is of ancient date in Europe ; it was known to the Kelts, and in India according to history. Lastly, the widely different com- mon names indicate likewise an ancient cultivation or long use in different countries. The Keltic name lin, and Greco-Latin linon or linum, has no analogy with the Hebrew pischta,^ nor with the Sanskrit names oo'ma, atasi, utasi. "^ A few botanists mention the flax as " nearly wild " in the south-east of Russia, to the south of the Caucasus and to the east of Siberia, but it was not known to be truly wild, I then summed up the probabilities, saying, " The varying etymology of the names, the antiquity of cultivation in Egypt, in Europe, and in the north of India, the circumstance that in the latter district flax is cultivated for the yield of oil alone, * PlaTichon, in Hooker's Journal of Botany, vol. 7; Beutham, Handhk. of Brit. Flora, edit. 4, p. 89. * Planchon, ibid. ^ Boissier, Fl. Or., i. p. 861. * A. de.CandoUe, Geogr. Bot. Rais., p. 833. * Thomson, Annals of Philosophy, June, 1834; Dntrochet, Larrey, and Costaz, Comptes rendus de VAcad. des. Sc, Paris, 1837, sem. i. p. 13d; Unger, Bot. Streifziige, iv. p. 62. '^ Other Hebrew words are interpreted " flax," but this is the most certain. See Hamilton, La Botanique de la Bible, Nice, 1871, p. 58. ^ Piddington, Index hid. Plants; Roxburgh, Fl. Ltd., edit, 1832, ii. p. 110. The name matusi indicated by Piddington belongs to other plants, according to Ad. Pictet, Origineslndo-Euro., edit. 2, vol. i. p. 396. 124 ORIGIN OF CULTIVATED PLANTS. lead me to believe that two or tlire3 species of different origin, confounded by most authors under the name of Linwinusitatisshnvmi, were formerly cultivated in different countries, without imitation or communication the one with the other. ... I am very doubtful whether the species cultivated by the ancient Egyptians was the species indigenous in Russia and in Siberia." My conjectures were confirmed ten years later by a very curious discovery made by Oswald fleer. The lake- dwellers of Eastern Switzerland, at a time when they only used stone implements, and did not know the use of hemp, cultivated and wove a flax which is not our common annual flax, but the perennial flax called Limim angusti- foliion, which is wild south of the Alps. This is shown by the examination of the capsules, seeds, and especially of the lower part of a plant carefully extracted from the sediment at Robenhausen.^ The illustration published by Heer shows distinctly a root surmounted by from two to four stems after the manner of perennial plants. The stems had been cut, whereas our common flax is plucked U]) by the roots, another proof of the persistent nature of the plant. With the remains of the Robenhausen flax some grains of S'llene cretica were found, a species which is also foreign to Switzerland, and abundant in Italy in the fields of flax.^ Hence Heer concluded that the Swiss lake-dwellers imported the seeds of the Italian flax. This was apparently the case, unless we suppose that the climate of Switzerland at that time differed from that of our own epoch, for the perennial flax would not at the present day survive the winters of Eastern Switzerland.^ Heer's opinion is supported by the surprising fact that flax has not been found among the remains of the lake-dwellings of Lay bach and Mondsee * Heer, Die Pflanzen der Pfahlhauten, 8vo pamphlet, Zurich, LSGo, p. 35 ; Ueber den Flachs und die Flachskultur in Alterthum, pampUlet iu 8vo, Zurich, 1872. ' Bertoloni, FI. Ital, iV. p. G12. ' We have seen that flax is found towards the north-west of Europe, but not immediately north of the Alps. Perhaps the climate of Switzer- land was fomiei 1y more equable than it is now, with more snow to shelter perenn al plants. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 125 of the Austrian States, where bronze has been discovered.^ The late epoch of the introduction of flax into this region excludes the hypothesis that the inhabitants of Switzer- land received it from Eastern Europe, from which, more- over, they were separated by immense forests. Since the ingenious observations of the Zurich savant, a flax has been discovered which was employed by the prehistoric inhabitants of the peat-mosses of Lagozza, in Lombardy; and Sordelli has shown that it was the same as that of Robenhausen, L. angusti folium.^ This ancient people was ignorant of the use of hemp and of metals, but they possessed the same cereals as the Swiss lake-dwellers of the stone age, and ate like them the acorns of Quercits rohiir, var. sessiliflora. There was, therefore, a civilization which had reached a certain development on both sides of the Alps, before metals, even bronze, were in common use, and before hemp and the domestic fowl were known.^ It was probably before the arrival of the Aryans in Europe, or soon after that event.^ The common names of the flax in ancient European languages may throw some light on this question. The name lin, llin, linu, linon, linuvi, lein, Ian, exists in all the European languages of Aryan origin of the centre and south of Europe, Keltic, Slavonic, Greek, or Latin. This name is, however, not common to the Aryan languages of India; consequently, as Pictet^ justly says, the cultivation must have been begun by the * Mittheil. Anthropol. Gesellschoft, Wien, vol. vi. pp. 122, 161; Ahhandl., Wien Akacl., 84, p. 488. * Sordelli, Sulle %>ianie della torhtera e deJla stazione preistorica dell-a Lagozza, pp. 37, 51, printed at the conclusion of Castelfranco's Noiizie alia stazione lacustre della Lagozza, in 8vo, Atti della Soc. Ital. Sc. NaL, 1880. ' The fowl was introduced into Greece from Asia in the sixth century before Christ, according to Heer, Ueh. d. Flachs, p. 25. ^ These discoveries in the peat.mosses of Lagozza and elsewhere in Italy show how far Hehn was mistaken in supposing that {Kidturpfl., edit. 3, 1877, p. 524) the Swiss lake-dwellers were near the time of Caesar. The men of the same civilization as they to the south of the Alps were evidently more ancient than the Roman republic, perhaps than the Ligurians. * Ad. Pictet, Origines Indo-Europ., edit. 2, vol. i. p. 396. 126 ORIGIN OF CULTIVATED PLANTS. western Aryans, and before their arrival in Europe. Another idea occurred to me which led me into further researches, but they were unproductive. I thought that, since this flax was cultivated by the lake-dwellers of Switzerland and Italy before the arrival of the Aryan peoples, it was probably also grown by the Iberians, who then occupied Spain and Gaul ; and perhaps some special name for it has remained among the Basques, the sup- posed descendants of the Iberians. Now, according to several dictionaries of their language/ liho, lino, or li, according to the dialects, signifies flax, which agrees with the name diffused throughout Southern Europe. The Basques seem, therefore, to have received flax from peoples of Aryan origin, or perhaps they have lost the ancient name and substituted that of the Kelts and Romans. The name flachs or jiax of the Teutonic lan- guages comes from the Old German yZr(//.s. There are also special names in the north-west of Europe — pellawa, aiivina, in Finnish;^ hor, hdrr, hor, in Danish;^ hor and tone in ancient Gothic.'* Haar exists in tlie German of Salzburg.^ This word may be in the ordinary sense of the German for thread or hair, as the name li may be connected with the same root as ligare, to bind, and as hor, in the plural horvar, is connected by philologists ^ with harva, the German root for Flachs ; but it is, never- theless, a fact that in Scandinavian countries and in Finland terms have been used which differ from those employed throughout the south of Europe. This variety shoAVS the antiquity of the cultivation, and agrees Avith the fact that tlie lake-dwellers of Switzerland and Italy cultivated a species of flax before the flrst invasion of the Aryans. It is possible, I might even say probable, that * Van Eys, Diet. Basque-Franrais, 1876; Geze, Elements de Gram- maire Basque suivis d'un vocahidaire, Bayonne, 1873; Salaberry, ^[nts Basques Navarrais, Bayonne, 1856; I'Ecluse, Vocah. Frang. -Basque, 1826. ' Nemnich, Poly. Lex. d. Naturgesch., ii. p. 420 ; Kafn, Banmark Flora, ii. p. 390. 8 Nemnich, ibid. * Tbid. ' Ihid. ^ Fick, Vergl. Worterhuch. Ind. Germ., 2nd edih, i. p. 722. He also derives the name Lina from the Latin linum ; but this name is of earlier date, being common to several European Aryan languages. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 127 the latter imported the name li rather than the plant or its cultivation ; but as there is no wild flax in the north of Europe, an ancient people, the Finns, of Turanian origin, introduced the flax into the north before the Aryans. In this case they must have cultivated the annual flax, for the perennial variety will not bear the severity of the northern winters; while we know how favourable the climate of Riga is in summer to the culti- vation of the annual flax. Its first introduction into Gaul, Switzerland, and Italy may have been from the south, by the Iberians, and in Finland by the Finns ; and the Aryans may have afterwards diffused those names which were commonest among themselves — that of linum in the south, and of flalis in the north. Perhaps the Ar3''ans and Finns had brought the annual flax from Asia, which would soon have been substituted for the perennial variety, which is less productive and less adapted to cold countries. It is not known precisely at what epoch the cultivation of the annual flax in Italy took the place of that of the perennial linum angusti- folium, but it must have been before the Christian era ; for Latin authors speak of a well-established cultivation, and Pliny says that the flax was sown in spring and rooted up in the summer.^ Metal implements were not then wanting, and therefore the flax would have been cut if it had been perennial. Moreover, the latter, if sown in spring, would not have ripened till autumn. For the same reasons the flax cultivated by the ancient Egyptians must have been an annual. Hitherto neither entire plants nor a great number of capsules have been found in the catacombs of a nature to furnish direct and incontestable proof linger ^ alone was able to ex- amine a capsule taken from the bricks of a monument, which Leipsius attributes to the thirteenth or fourteenth century before Christ, and he found it more like those of L. usitatissimum than of L. angustifoliwni. Out oT three seeds which Braun^ saw in the Berlin Museum, ' Pliny, bk. xix. c. 1 : Vere satum cestafe velUtur. * Unger, Botanische Streifzuge, 1866, No. 7, p. 15. " A. Braun, Die Pflanzenreste des ^(jyptischen Museums in Berlin, in 8vo, 1877, p. 4. 128 ORIGIN OF CULTIVATED PLANTS. mixed with those of other cultivated plants, one appeared to him to belong to L. angustifoliuin, and tlie other to L. hitmile ; but it must be owned that a sino-le seed without plant or capsule is not sufficient proof. Ancient Eg3^ptian paintings show that flax was not reaped with a sickle like cereals, but uprooted.^ In Eg3^pt flax is cultivated in the winter, for the summer drought would no more allow of a perennial variety, than the cold of northern countries, where it is sown in spring, to be gathered in in summer. It may be added that the annual flax of the variety called hiiinile is the only one now grown in Abyssinia, and also tlie only one that modern collectors have seen in Egypt.^ Heer suggests that the ancient Egyptians may have cultivated L. angustifoliuin of the Mediterranean region, sowing it as an annual plant.^ I am more inclined to believe that they had previously imported or re- ceived their flax from Egypt, already in the form of the species L. humile. Their modes of cultivation, and the flnrures on the monuments, show that their knowledore of the plant dated from a remote antiquity. Now it is known that the Egyptians of the first dynasties before Cheops belonged to a proto-semitic race, which came into Egypt by the isthmus of Suez.* Flax has been found in a tomb of ancient Chaldea prior to the existence of Babylon,^ and its use in this region is lost in the remotest antiquity. Thus the first Egyptians of white race may have imported the cultivated flax, or their im- mediate successors may have received it from Asia before the epoch of the Phcenician colonies in Greece, and before direct communication was established between Greece and Egypt under the fourteenth dynasty.^ ' Rosellini, pis. 35 and 36, quoted by Unger, Bot. Streifziige, No. 4, p. 62. 2 W. Schimper, Ascherson, Boissier, Schweinfarth, quoted by Brann. 3 Heer, Ueb. d. Flach.t, p. 26. * Maspero, Histoire Ancienne des Peuples de V Orient., edit. 3, Paris, 1S78, p. 13. * Journal of the Royal Asiat. Sac, vol. xv. p. 271, quoted by Heer, Ueh. den Fl. * Maspero, p. 213. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 129 A very early introduction of the plant into Egypt from Asia does not prevent us from admitting that it was at different times taken from the East to the West at a later epoch than that of the first Egyptian dynasties. Thus the western Aryans and the Phoenicians may have introduced into Europe a flax more advantageous than L. angtistifolium during the period from 2500 to 1200 years before our era. The cultivation of the plant by the Ar3^ans must have extended further north than that by the Phoenicians. In Greece, at the time of the Trojan war, fine linen stuffs were still imported from Colchis; that is to say, from that reofion at the foot of the Caucasus where the com- mon annual flax has been found wild in modern times. It does not appear that the Greeks cultivated the plant at that epoch.^ The Aryans had perhaps already intro- duced its cultivation into the valley of the Danube. How- ever, I noticed just now that the lacustrine remains of Mondsee and Laybach show no trace of any flax. In the last centuries before the Christian era the Romans pro- cured very fine linen from Spain, although the names of the plant in that country do not tend to show that the Phoenicians introduced it. There is not any Oriental name existing in Europe belonging either to antiquity or to the Middle Ages. The Arabic name kattan, kettane, or kittane, of Persian origin,^ has spread westward only among the Kabyles of Algeria.^ The sum of facts and probabilities appear to me to lead to the following statements, which may be accepted until they are modified by further discoveries. 1. Linum angiistifolinim, usually perennial, rarely biennial or annual, which is found wild from the Canary Isles to Palestine and the Caucasus, was cultivated in Switzerland and the north of Italy by peoples more ancient than the conquerors of Aryan race. Its cultiva- tion was replaced by that of the annual flax. * The Greek texts are quoted in Lenz, Bot der Alt. Gr. und Rom., p. 672 ; and in Hehn, Culturpji. und Hausthiere, edit. 3, p. 144. * Ad, Pictet, Origines Indo-Europ. " Dictionnaire Franf.-Berhere, 1 vol. in 8vo, 1814. 130 ORIGIN OF CULTIVATED PLANTS. 2. The annual flax (Z. usitatisshnuiii), cultivated for at least four thousand or five thousand years in Mesopo- tamia, Assyria, and Egypt, was and still is wild in the districts included between the Persian Gulf, the Caspian Sea, and the Black Sea. 3. This annual flax appears to have been introduced into the north of Europe by the Finns (of Turanian race), afterwards into the rest of Europe by the western Aryans, and perhaps here and there by the Phcenicians; lastly into Hindustan by the eastern Aryans, after their sepa- ration from the European Aryans. 4. These two principal forms or conditions of flax exist in cultivation, and have probably been wild in their modern areas for the last five thousand years at least. Tt is not possible to guess at their previous condition. Their transitions and varieties are so numerous that they may be considered as one species comprising two or three hereditary varieties, which are each again divided into subvarieties. Jute — CorcJiorus capsularis and Corchorus olitorlus, Linnpeus. The fibres of the jute, imported in great quantities in the last few years, especially into England, are taken from the stem of these two species of Corchorus, annuals of the family of the Tiliacea?. The leaves are also used as a vegetable. C. capsularis has a nearly spherical fruit, flattened at the top, and surrounded by longitudinal ridges. There is a good coloured illustration of it in tlie work of the younger Jacquin, Eduga', pi. 119. C. olitorius, on the contrary, has a long fruit, like the pod of a Crucifer. It is figured in the Botanical Magazine, fig. 2810, and in Lamarck, fig. 478. The species of the genus are distributed nearly equally in the warm regions of Asia, Africa, and America ; con- sequently the origin of each cannot be guessed. It must be sought in floras and herbaria, with the help of his- torical and other data. Corchorus capsulirh is commonly cidtivated in the Sunda Islands, in Ceylon, in the peninsula of Hin- PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 131 diistan, in Bengal, in Southern China, in the Philippine Islands,^ generally in Southern Asia. Forster does not mention it in his work on the plants in use among the inhabitants of the Pacific, whence it may be inferred that at the time of Cook's voyages, a century ago, its cul- tivation had not spread in that direction. It may even be suspected from this fact that it does not date from a very remote epoch in the isles of the Indian Archipelago. Blume says that Corchorus capsidaris grows in the marshes of Java near Parang,^ and I have two speci- mens from Java which are not given as cultivated.^ Thwaites mentions it as " very common " in Ceylon.^ On the continent of Asia, authors speak more of it as a plant cultivated in Bengal and China. Wight, who gives a good illustration of the plant, does not mention its native place. Edgeworth,^ who has studied on the spot the flora of the district of Banda, says that it is found in " the fields." In the Flora of British India, Masters, who drew up the article on the Tiliacese from the herbarium at Kew, says " in the hottest regions of India, cultivated in most tropical countries." ^ I have a specimen from Bengal which is not given as cul- tivated. Loureiro says "wild, and cultivated in the province of Canton in ChinaJ which probably means wild in Cochin-China, and cultivated in Canton. In Japan the plant grows in cultivated soil.^ In conclusion, I am not convinced that the species exists in a truly wild state north of Calcutta, although it may perhaps have spread from cultivation and have sown itself here and there. G. capsidaris has been introduced into various parts of tropical Africa and even of America, but it is only cultivated on a large scale for the production of jute thread in Southern Asia, and especially in Bengal. * Rumphius, Amhoin, vol. v. p. 212; Roxburgh, Fl. Lid., ii. p. 5S1; Loureiro, Fl. Cochinchine, vi. p. 408. * Blume, Bijdmrjen, i. p. 110. ^ Zollinger, Xos. 1G98 and 27GI. * Thwaites, Eiium. PL Zeylan., p. 31. * Edgeworth, LinncBan Soc. Journ., ix. * Masters, in Hooker's Fl. Brit, hid., i, p. 307, ' Loureiro, Fl. Cochin., i. p. 408. 8 Franchet aud Savatier, I^nurti'^ i. P. 69. 7 132 ORIGIN OF CULTIVATED PLANTS. C. olitorius is more used as a vegetable than for its fibres. Out of Asia it is employed exclusively for the leaves. It is one of the commonest of culinary plants among the modern Egyptians and Syrians, who call it in Arabic mclol'ych, but it is not likely that they had any knowledge of it in ancient times, as we know of no Hebrew name.^ The present inhabitants of Crete cultivate it under the name of moucJdia'^ evidently derived from the Arabic, and the ancient Greeks were not acquainted with it. According to several authors^ this species of Corchorus is wild in several provinces of British India. Thwaites says it is common in the hot districts of Ceylon ; but in Java, Blume only mentions it as growing among rubbish {in rudcvatis). I cannot find it mentioned in Cochin-China or Japan. Boissier saw specimens from Mesopotamia, Afghanistan, Syria, and Anatolia, but gives as a general indication, "culta, et in iiideratis subspontanea." !No Sanskrit name for the two cultivated species of Corchorus is kno^vn.* Touching the indigenous character of the plant in Africa, Masters, in Oliver's Flora of Tropical Africa (i. p. 262), says, " wild, or cultivated as a vegetable througli- out tropical Africa." He attributes to the same species two plants from Guinea which G. Don had described as different, and as to whose Avild nature he probably knew nothing. I have a specimen from Kordofan gathered by Kotschy, No. 45, "on the borders of the fields of sorghum." Peters, as far as I know, is the only author who asserts that the plant is wild. He found C. olitomvs " in dry places, and also in the meadows in the neighbour- hood of Sena and Tette." Schweinfurth only gives it as a cultivated plant in the whole Nile Valley.^ This is also the case in the flora of Senegambia by Guillemin, Perrottet, and Richard. ^ Rosenmuller, Bihh Naturgesch. ' Von Heldreich, Die Kiitzpjl. Griechenl., p. 53. ' Masters, in Hooker's Fl. Brit. Ind., i. p. 3t)7 ; Aitchison, Catal. Punjah, p. 23 ; Eoxburgh, FL Ind., ii. p. 581. * Pidclin,2:ton, Index. * Snliweinfurth, Beitr. Z. Fl. A^lhinp., p. 2Gk PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 133 To sum up, C. olitorius seems to be wild in the mode- rately waim regions of Western India, of Kordofan, and probably of some intermediate countries. It must have spread from the coast of Timor, and as far as Northern Australia, into Africa and towards Anatolia, in the wake of a cultivation not perhaps ot earlier date than the Christian era, even at its origin. In spite of the assertions made in various works, the cultivation of this plant is rarely indicated in America. I note, however, on Grisebach's authority,^ that it has become naturalized in Jamaica from gardens, as often happens in the case of cultivated annuals. Sumach. — Rhus coriaria. This tree is cultivated in Spain and Italy ^ for the young shoots and leaves, which are dried and made into a powder for tanning. I recently saw a plantation in Sicily, of which the product was exported to America. As oak-bark becomes more rare and substances for tan- ning are more in demand, it is probable that this cultiva- tion will spread ; all the more that it is suitable to sandy, sterile regions. In Algeria, Australia, at the Cape, and in the Argentine Republic, it might be introduced with advantage.^ Ancient peoples used the slightly acid fruits as a seasoning, and the custom has lingered here and there ; but I find no proof that they cultivated the species. It grows wild in the Canaries and in Madeira, in the Mediterranean reo^ion and in the neicjhbourhood of the Black Sea, preferring dry and stony ground. In Asia its area extends as far as the south of the Cau- casus, the Caspian Sea, and Persia.* The species is so common that it may have been in use before it was cultivated. 1 Grisebach, Fl. of Brit. West Ind., p. 97. * Bosc, Diet. d'Agric, at the word " Sumac." ' The conditions and methods of the culture of the sumach are the subject of an important paper by Inzenga, translated in the Bull. Soc. d'Acclim., Feb. 1877. In the Trans. Bot. Soc. of Edinhurgh , ix. p. 311, maybe seen an extract from an earlier paper by the author on the t-jnmo subject. * Ledebour, Fl. Boss., i. p. 509 ; Boissier, Fl. Orient., ii. p. 4. 134} ORIGIN OF CULTIVATED PLANTS. Sumacli is the Persian and Tartar name ; ^ rous, rhus, the ancient name among the Greeks and Romans.^ A proof of the persistence of certain common names is found in the French " Currier's roux or roure.'' Khat, or Arab Tea — Catha edulis, Forskal; Celastrus edulis, Vahl. This shrub, belonging to the family of the Celastracece, is largely cultivated in Abyssinia, under the name of tchut or tchat, and in Arabia under that of cat or gat. Its leaves are chewed, when green, like those of the coca in America, and they have the same exciting and strength- ening properties. Those of uncultivated plants have a stronger taste, and are even intoxicating. Botta saw that in Yemen as much importance is attributed to the cultivation of the Cdtha as to tliat of coffee, and he mentions that a sheik, who is obliged to receive many visits of ceremony, bought as much as a hundi-ed francs' worth of leaves a day.^ In Abyssinia an infusion is also made from the leaves.* In spite of the eagerness with which .stimulants are sought, this species has not spread into the adjoining countries, such as Beluchistan, Soutliern India, etc., where it might succeed. The Catha is w^ild in Abyssinia,^ but has not yet been found wild in Arabia. It is true that the interior of the country is nearly unknown to botanists. It cannot be ascertained from Botta's account whether the wuld plants he mentions are wild and indigenous, or escaped from cultivation and more or less naturalized. Perhaps the Catha was introduced from Abyssinia with the coffee plant, which likewise has not been discovered wild in Arabia. Mate — Ilex paraguarieoisis, Saint-Hilaire. The inhabitants of Brazil and of Paraguay have em- * Nemnich, Polygl. Lexicon, ii. p. 1156 ; AiusHe, Mat. Med. Ind., i. p. 414. * Fraas, Sim. Fl. Class., p. 85. ' Forskal, Flora JEijypto-Arahica,p. 65 ; Ricliard, Tentamen Fl. Abyss., i. p. 134, pi. 30; Botta, Archives du Museum, ii. p. 73. * Hochstetter, Flora, 1841, p. G63. ^ Schweinfui-th and Asclierson, Aufzdhlung , p. 263; Oliver, FL Trap. Afr., i. p. 3G4. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 135 ployed from time immemorial the leaves of this shrub, as the Chinese have those of the tea plant. They gather them especially in the damp forests of the interior, between the degrees of 20 and 30 south latitude, and commerce trans- ports them dried to great distances throughout the greater part of South America. These leaves contain, with aroma and tannin, a principle analogous to that of tea and cotfee ; they are not, however, much liked in the countries where Chinese tea is known. The plantations of mate are not yet as important as the product of the wild shrub, but they may increase as the population increases. More- over, the preparation is simpler than that of tea, as the leaves are not rolled. Illustrations and descriptions of the species, with a number of details about its use and properties, may be found in the works of Saint-Hilaire, of Sir William Hooker, and of Martins.-^ Coca. — ErytliToxylon Coca, Lamarck. The natives of Peru and of the neighbouring pro- vinces, at least in the hot moist regions, cultivate this shrub, of which they chew the leaves, as the natives of India chew the leaves of the betel. It is a very ancient custom, which has spread even into elevated regions, where the species cannot live. Kow that it is known how to extract the essential part of the coca, and its virtues are recognized as a tonic, which gives strength to endure fatigue without having the drawbacks of alcoholic liquors, it is probable that an attempt will be made to extend its cultivation in America and elsewhere. In Guiana, for instance, the Malay Archipelago, or the valleys of Sikkim and Assam, or in Hindustan, since both moisture and heat are requisite. Frost is very injurious to the species. The best sites are the slopes of hills where water cannot lie. An attempt made in the neighbourhood of Lima failed, because of the infrequency of rain and perhaps because of insufficient heat.^ * Aug. de Saint Hilaire, Mem. du MusSiim, ix. p. 351 ; Ann. Sc. Naf., 3rd series, xiv. p. 52 ; Hooker, London Journal of Butany, i. p. 34 j Martius, Flora Brasiliensis, vol. ii. part 1, p. 119. ' Martinet, Bull. Soc. d'Acclim., 1874, p. 449. 136 ORIGIN OF CULTIVATED PLANTS. I shall not repeat here what may be found in several excellent treatises on the coca ; ^ I need only say that the original home of the species in America is not yet clearly ascertained. Gosse has shown that early authors, such as Joseph de Jussieu, Lamarck, and Cavanilies, had only seen cultivated specimens. Mathews gathered it in Peru, in the ravine (quehrada) of Chiiichao,^ which appears to be a place beyond the limits of cultivation. Some specimens from Cuchero, collected by Poeppig,^ are said to be wild ; but the traveller himself was not convinced of their wild nature.* D'Orbigny thinks he saw the wild coca on a hill in the eastern part of Bolivia.^ Lastly, M. Andre has had the courtesy to send me the specimens of Bry- thvoxylon in his herbarium, and I recognized the coca in several specimens from the valley of the river Cauca iu New Granada, with the note " in abundance, wild or half- wdld." Triana, however, does not admit that the species is wild in his countrv, New Granada.^ Its extreme im- portance in Peru at the time of the Incas, compared to the rarity of its use in New Granada, seems to show that it has escaped from cultivation in places where it occurs in the latter country, and that the species is in- digenous only in the east of Peru and Bolivia, according to the indications of the travellers m.entioned above. Dyer's Indig^o. — Tndigofera tindoria, Linnseus. The Sanskrit name is nili '^ The Latin name, indicitm, shov/s that the Romans knew that the indigo was a substance brouo^ht from India. As to the wild nature of the plant, Roxburgh says, "Native place un- known, for, though it is now common in a wild state in most of the provinces of India, it is seldom found far from the districts where it is now cultivated, or has been culti- vated formerly." Wight and Royle, who have published illustrations of the species, tell us nothing on this head, * Particularly in Gosse's Monographie de V Erythroxylon Coca, in 8vo, LSf3L ' Hooker, Comp. to the Bot. Mag., ii. p. 25. ' Poyritscli, in tlie Flora Brasil., fasc. 81, p. 156. ^ Hooker, Comp. to the Bot. Mag. ' Gc sse, Monogr., p. 12. • Triana and I'lanchon, Ann. Sciences Nat., 4th series, vol. 18, p. 338. ' Roxburgh, Fl. Ind., iii. p. 379. PLANTS CULTIVATED FOE THEIR STExMS OR LEAVES. 137 and more recent Indian floras mention the plant as cultivated.^ Several other indigoes are wild in India. This species has been found in the sands of Senegal,^ but it is not mentioned in other African localities, and as it is often cultivated in Senegal, it seems probable that it is naturalized. The existence of a Sanskrit name renders its Asiatic origin most probable. Silver Indigo — Tndigofera argentea. This species is certainly wild in Abyssinia, Nubia, Kordofan, and Senaar.^ It is cultivated in Egypt and Arabia. Hence we might suppose that it was from this species that the ancient Egyptians extracted a blue dye ; * but perhaps they imported their indigo from India, for its cultivation in Eg3^pt is probably not of earlier date than the Middle Ages.^ A slightly different form, which Koxburgh gives as a separate species {Indigofera cceridea), and which appears rather to be a variety, is wild in the plains of the peninsula of Hindustan and of Beluchistan. American Indigoes. There are probably one or two indigoes indigenous in America, but ill defined, and often intermixed in cultiva- tion with the species of the old world, and naturalized beyond the limits of cultivation. This interchange makes the matter too uncertain for me to venture upon any researches into their original habitat. Some authors have thought that /. Anil, Linnaeus, was one of these species. Linnaeus, however, says that his plant came from India (Mantissa, p. 273). The blue dye of the ancient Mexicans was extracted from a plant which, according to Hernandez' account,^ differs widely from the indigoes. * Wip^ht, IconeSy t. 3G5 ; Rojle, HI. Himal., t. 195 ; Baker, in Flora of Brit. Ind.y ii. p. 98 ; Brandis, Fo7-est Flora, p. 136. 2 Guillemin, Perrottet, and Richard, Florce Seneg. Tentamen, p. 178. ^ Richard, Tentamen Fl. Abyss., i. 184 ; Oliver, Fl. of Trop. Afr., ii. p. 97 ; Schweinfurth and Ascherson, Aufzdhlung, p. 256. ** Ungei', Pjianzen d. Alt. .^gyptens, p. 66; Pickeriug, Chronol. Arrang., p. 443. * Rcynier, Economte des Juifs, p. 439j des Egypticns, p. 35-1. ® Hernandez, Thes., p. 108. / 138 ORIGIN OF CULTIVATED PLANTS. Henna — Lcavsonia cdha, Lamarck {Laicsonia inermis and L. spinosa of different authors). The custom amonsr Eastern women of staminof their nails red with the juice of henna-leaves dates from a remote antiquity, as ancient Egyptian paintings and mummies show. It is difficult to know when and in what country this species was first cultivated to fulfil the requirements of a fashion as absurd as it is persistent, but it may be from a very early epoch, since the inhabitants of Babylon, Nineveh, and the towns of Egypt had gardens. It may be left to scholars to show whether the practice of stain- ing the nails began in Egypt under this or that d3masty, before or after certain relations were established with Eastern nations. It is enough for our purpose to know that Lciwsoiiia, a shrub belonging to the order of the Lythraceee, is more or less wild in the warm regions of Western Asia and of Africa to the north of the equator. I have in my possession specimens from India, Java, Timor, even from China ^ and Nubia, which are not said to be taken from cultivated plants, and others from Guiana and the West Indies, which are doubtless fur- nished by the imported species. Stocks found it indige- nous in Beluchistan.^ Roxburgh also considered it to be wild on the Coromandcl ^ coast, and Thwaites * mentions it in Ceylon in a manner which seems to show that it is wild there. Clarke ^ says, " very common, and cultivated in India, perhaps wild in the eastern part." It is pos- sible that it spread into India from its original home, as into Amboyna ^ in the seventeenth century, and perhaps more recently into the West Indies,''' in the wake of culti- vation ; for the plant is valued for the scent of its flowers, as well as for the dye, and is easily propagated by seed. ^ Fortune, No. 32. "^ Aitchiscn, CataL of PL of Punjab and Sindh, -p. GOj Boissier, Fl, Orient., ii. p. 744. 3 Roxburgh, Fl. Ind., ii. p. 258. * Thwaites, Emim. PLZeyL, p. 122. * Clarke, in Hooker's Fl. Brit. Ind., ii. p. 273. ' Ramphius, Amh., iv. p. 42. ' Grisebach, Fl. Brit. W. Ind., i. p. 271. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 139 There is the same doubt as to whether it is indigenous in Persia, Arabia, and Egypt (an essentially cultivated country), in Nubia, and even in Guinea, where specimens have been gathered.'^ It is even possible that the area of this shrub extends from India to Nubia. Such a wide geographical distribution is, however, always somewhat rare. The common names may furnish some indication. A Sanskrit name, sahachera,^ is attributed to the species, but as it has left no trace in the different modern languages of India, I am inclined to doubt its reality. The Persian name hanna is more widely diffused and retained than any other {hina of the Hindus, henneh and alhenna of the Arabs, Jcinna of the modern Greeks). That of cypros, used by the Syrians of the time of Dioscorides,^ has not found so much favour. This fact supports the opinion that the species grew originally on the borders of Persia, and that its use as well as its cultivation spread from the East to the West, from Asia into Africa. Tobacco — Nicotiana TahacuTn, Linnasus ; and other species of Nicotiana. At the time of the discovery of America, the custom of smoking, of snuff- taking, or of chewing tobacco was diffused over the greater part of this vast continent. The accounts of the earliest travellers, of which the famous anatomist Tiedemann ^ has made a very complete collection, show that the inhabitants of South America did not smoke, but chewed tobacco or took snuff, except in the district of La Plata, Uruguay, and Paraguay, where no form of tobacco was used. In North America, from the Isthmus of Panama and the West Indies as far as Canada and California, the custom of smoking was universal, and circumstances show that it was also very ancient. Pipes, in great numbers and of wonderful work- manship, have been discovered in the tombs of the Aztecs . » Oliver, Fl. of Trop. Afr., ii. p. 483. ^ Piddington, Index. ^ Dioscorides, 1, c. 124 ; Lenz, Bot. d. Alien, p. 177- * Tiedemann, Geschichte des Tabaks, in 8vo, 1854. For Brazil, see Martins, Beitrage ziir Ethnographie und Sprachkunde Amerikas, i. p. 719. 14)0 ORIGIN OF CULTIVATED PLANTS. in Mexico ^ and in the mounds of the United States ; some of them represent animals foreign to North America.^ As the tobacco plant is an annual which gives a great quantity of seeds, it was easy to sow and to cultivate or naturalize them more or less in the neighbourhood of dw^ellings, but it must be noted that different species of the genus Nicotiana w^ere employed in different parts of America, wdiich shows that they had not all the same origin. Kicotiana Tahacum, commonly cultivated, was the most widely diffused, and sometimes the only one in use in South America and the West Indies. The use of tobacco was introduced into La Plata, Paraguay,^ and Uruguay by the Spaniards, consequently we must look further to the north for the origin of the plant. De Martius does not think it was indigenous in Brazil,^ and he adds that the ancient Brazilians smoked the leaves of a species belonging to their country known to botanists as Nicotiana Langsdorjii. When I w^ent into the question in 1855,^ I had not been able to dis- cover any wild specimens of Kicotiana TahacvbVii except those sent by Blanchet from the province of Bahia, numbered 32:^3,0'. No author, either before or since that time, has been more fortunate, and I see that Messrs. Fliickiger and llanbury, in their excellent work on vegetable drugs,^ say positively, " The common tobacco is a native of the new w^orld, though not now known in a wild state." I venture to gainsay this assertion, although the w^ild nature of a plant may always be disputed in the case of a plant wdiich spreads so easily from cultivation. We find in herbaria a number of specimens gathered in Peru wdthout indication that they w^ere cultivated or that they grew near plantations. Boissier's herbarium contains * Tiedemann, p. 17, pi. 1. * The drawings on these pipes are reprodnced in Naidaillac*a recent work, Les Premiers Homines et les Temps Frehistoriques, vol. ii. pp. 45, 48. ' Tiedemann, pp. 38, 39. * Martius, Syst. Mat. Med. Bras., p. 120 ; Fl. Bras., vcl. x. p. 191. 5 A. de Candolle, Gcogr. Bot. Rai^otnu'e, p. 84i). ' Fliickiger and Haubury, Pharmacographia, p. 418. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 141 two specimens collected by Pavon, from different locali- ties.-^ Pavon sa^^s in his liora that the species grows in the moist warm forests of the Peruvian Andes, and that it is cultivated. But — and this is more significant — Edouard Andre gathered specimens in the republic of Ecquador at Saint Nicholas, on the v/estern slope of the volcano of Corazon in a virgin forest. These he was kind enough to send me. They are evidently the tall variety (four to six feet) of N. Tahacum, with the upper leaves narrow and acuminate, as they are represented in the plates of Hayne and Miller.^ The lower leaves are wanting. The flower, which gives the true characters of the species, is certainly that of N, Tahacum, and it is well known that the height of this plant and the breadth of the leaves vary in cultivation.^ It is very possible that its original country extended north as far as Mexico, as far south as Bolivia, and eastward to Venezuela. Nicotiana rustica, Linnreus, a species with yellow flowers, very different from Tahacum,^ and which yields a coarse kind of tobacco, was more often cultivated by the Mexicans and the native tribes north of Mexico. I have a specimen brought from California by Douglas in 1837, a time when colonists were still few; but American authorities do not admit that the plant is wild, and Dr. Asa Gray says that it sows itself in waste places.^ This v/as perhaps the case with the specimens in Boissier's herbarium, gathered in Peru by Pavon, and which he does not mention in the Peruvian flora. The species grows in abundance about Cordova in the Argentine Republic/ but from what epoch is unknown. From the * One of these is classed under the name Nicot. fruticosa^ which in my opinion is the same species, tall, but not woody, as the name would lead one to believe. N. auriculata, Bertero, is also Tahacum, according to my authentic specimens. ^ Hayne, Arzneikunde Gewachset vol. xii. t. 41; Miller, Figures of Plants, pi. 185, f. 1. ^ Tlie capsule is sometimes shorter and sonaetimos longer than the calix, on the same plant, in Andre's specimens. * See the figures of N. rustica in Plee, Types de Families Natiirelles de France, Solanees ; BuUiard, Herhier de France, t. 289. ^ Asa Gray, 8'jn. Flora of North Amer. (1878), p. 21-1. ^ Martin de Moussy, Descr. de la Repub. Argent. ^ i. p. 193. 142 ORIGIN OF CULTIVATED PLANTS. ancient use of the plant and the home of the most analo- gous species, the probabilities are in favour of a Mexican, Texan, or Californian origin. Several botanists, even Americans, have believed that the species came from the old world. This is certainly a mistake, although the plant has spread here and there even into our forests, and sometimes in abundance,^ having escaped from cultivation. Authors of the six- teenth century spoke of it as a foreign plant introduced into gardens and sometimes spreading from them.^ It occurs in some herbaria under the names of iV. tar- tarica, turcica^ or sihirica ; but these are garden-grown specimens, and no botanist has found the species in Asia, or on the borders of Asia, with any appearance of wildness. This leads me to refute a widespread and more per- sistent error, in spite of what I proved in 1855, namely, that of regarding some species ill described from culti- vated specimens as natives of the old world, of Asia in particular. The proofs of an American origin are so numerous and consistent that, without entering much into detail, I may sum them up as follows : — A. Out of fifty species of the genus Nicotiana found in a wild state, tv\^o only are foreign to America ; namely, N. suavolens of New Holland, with which is joined N". Totundifolia of the same country, and that which Ventinat had wrongly styled i\^. undulata ; and A^. /ra- gans, Hooker, of the Isle of Pines, near New Caledonia, which differs very little from the preceding. B. Though the Asiatic people are great lovers of tobacco, and have from a very early epoch sought the smoke of certain narcotic plants, none of them made use of tobacco before the discovery of America. Tiedemann has distinctly proved this fact by thorough researches into the writinsfs of travellers in the Middle As^es.^ He even quotes for a later epoch, not long after the dis- covery of America, between 1540 and KiOo, the fact that * Bulliard, Herhier de France. 2 CaDsalpinus, lib. viii. cap. 44; Banhin, Hist., iii. p. 630. ' Tiedemaun, Geschichte des TahaJm (1854), p. 208. Two years earlier, Volz, Beifrage zur CnUurgeschichte, had collected a number of facts relative to the introduction of tobacco into different countries. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 143 several travellers, some of whom were botanists, such as Belon and Rauwolf, who travelled through the Turkisli and Persian empires, observing their customs with much attention, have not once mentioned tobacco. It was evidently introduced into Turkey at the beginning of the seventeenth century, and the Persians soon received it from the Turks. The first European who mentions the smoking of tobacco in Persia is Thomas Herbert, in 1626. No later travellers have omitted to notice the use of the hookah as well established. Olearius describes this ap- paratus, which he saw in 1633. The first mention of tobacco in India is in 1605,^ and it is probable that it was of European introduction. It was first introduced at Arracan and Pegu, in 1619, according to the traveller Methold.^ There are doubts about Java, because Rum- phius, a very accurate observer, who wrote in the second half of the seventeenth century, says^ that, according to the tradition of some old people, tobacco had been employed as a medicine before the arrival of the Portu- guese in 1496, and that only the practice of smoking it had been communicated by the Europeans. Rumphius adds, it is true, that the name tahaco or tainhuco, which is in use in all these places, is of foreign origin. Sir Stamford Raffles,* in his numerous historical researches on Java, gives, on the other hand, the year KiOl as the date of the introduction of tobacco into Java. The Portuguese had certainly discovered the coasts of Brazil between 1500 and 1504, but Yasco di Gama and his successors went to Asia round the Cape, or through the Red Sea, so that they could hardly have established frequent or direct communications between America and Java. Nicot had seen the plant in Portugal in 1560, so that the Portuguese probably introduced it into Asia in the latter half of the sixteenth century. Thunberg affirms^ that the use of tobacco was introduced into * According to an anonymous Indian author quoted by Tiedemann, p. 229. * Tiedemann, p. 234. ' Rumphius, Herb. Amhoin v. p. 225. * RaflSes, Descr. of Java, p. 85. ® Thunberg, Flora Japonica, p. 91. 144 ORIGIN OF CULTIVATED PLANTS, Japan by the Portuguese, and according to early travellers quoted by Tiedemann, this was at the beginning of the seventeenth century. Lastly, the Chinese have no original and ancient sign for tobacco ; their paintings on china in the Dresden collection often present, from the year 1700 and never before that date, details relating to tobacco,^ and Chinese students are aofreed that Chinese works do not mention the plant before the end of the sixteenth century.^ If it be remembered with what rapidity the use of tobacco has spread wherever it has been intro- ^ duced, these data about Asia have an incontestable force. (.- G. The common names of tobacco confirm its American origin. If there had been any indigenous species in the old world there would be a great number of different names ; but, on the contrary, the Chinese, Japanese, Javanese, Indian, Persian, etc., names are derived from the American names, petiun, or tabak, to.hoJc, tamboc, slightly modified. It is true that Pid- dington gives Sanskrit names, dluiinrapatra and tani- rakouta^ but Adolphe Pictet informs me that the first of these names, which is not in Wilson's dictionary, means only leaf for smoking, and appears to be of modern com- position; while the second is probably no older, and seems to be a modern modification of the American names. The Arabic v^ord docchan simply means smoke.* Lastly, we must inquire into the two so-called Asiatic Kicotiance. The one, called by Lehmann N'tcotiana chinensis, came irom the Russian botanist Fischer, who said it was Chinese. Lehmann said he had seen it in a garden. Now, it is well known how often an erroneous origin is attributed to plants grown by horticulturists and besides, from the description, it seems that it was siniply N. Tabacum, of which the seeds had perhaps come from China.^ The second species is N. persica, * Klemm, quoted by Tiedemann, p. 256. ^ Stanislas Julien, in de CandoUe, Geogr. Bof. Rais., p, 851 ; BretscVineider, Study and Value, etc., p. 17. ^ Piddington, Inde.v. * Forskal, p. G3. ^ Lehmann, Historia Nicofinarum, p. 18. The epithet sufruticosa is an exatrgeration applied to the tobacc( s, Avhich are always annual. I have said already that N. suffruticosa of ditloreut auth )rs is N. Tabacum. PLANTS CULTIVATED FOR THEIK STEMS OR LEAVES. 145 Lindley, figured in the Botanical Register (pi. 1592), of wliich the seeds had been sent from Ispahan to tlie Horticultural Society of London, as those of the best tobacco cultivated in Persia, that of Schiraz. Lindley did not observe that it corresponded exactly to i\". alata, drawn three years before by Link and Otto ^ from a plant in the gardens at Berlin. The latter was grown from seed sent by Sello from Southern Brazil. It is certainly a Brazilian species, with a white elongated corolla, allied to N. suaveolens of New Holland. Thus the tobacco cultivated sometimes in Persia along with the common species, is of American origin, as I declared in my Geogvapltical Botany of 1855. I do not under- stand how this species was introduced into Persia. It must have been from seed taken from a garden, or brought by chance from America, and it is not likely that its cultivation is common in Persia, for Olivier and Bruguiere, and other naturalists who have observed the tobacco plantations in that country, make no mention of it. From all these reasons I conclude that no species of tobacco is a native of Asia. They are all American, except N. suaveolens of New Holland, and N. fragrans of the Isle of Pines to the south of New Caledonia. Several Nicotiance, besides H. Tahacum and N. rus- tica, have been cultivated here and there by savages, or as a curiosity by Europeans. It is strange that so little notice is taken of these attempts, by means of which very choice tobacco might be obtained. The species with white flowers would yield probably a light and perfumed tobacco, and as some smokers seek the strongest tobaccos and the most disagreeable to non- smokers, I would recommend to their notice N. angusti- folia of Chili, which the natives call tabaco del diablo.^ ^ Link and Otto, Icones Plant. Bar. Hort. Ber., in 4to, p. 63, t. 32. Sendtner, in Flora Brasil, vol. x. p. 167, describes the same plant as Sello, as it seems from the specimens collected by this traveller; and Grisebach, Symholce Fl. Argent., p. 2id, mentions N. alata in the pro- vince of Fntrerios of the Argentine republic. * Bertero, in De Cand., Prodr., xii., sect. 1, p. 568. 14G ORIGIN OF CULTIVATED PLANTS. Cinnamon — Cinyiamonuvi zeylanicum, Breyn. This little tree, belonging to the laurel tribe, of which the bark of the young branches forms the cinnamon of commerce, grows in great quantities in the forests of Ceylon. Certain varieties "which grow wild on the con- tinent of India were formerly considered to be so many distinct species, but Anglo-Indian botanists are agreed in connectinor them with that of Cevlon.^ The bark of C. zeylanicum, and that of several uncul- tivated species of Cinnamonvjn, which produce the cassia, or Chinese cassia, have been an important article of commerce from a very early period. Fluckiger and H anbury ^ have treated of this historical question with so much learning and thoroughness, that we need only refer to their work, entitled Fharinacographia, or His- tory of the Principal Drugs of Vegetable Origin. It is important from our point of view to note how modern the culture is of the cinnamon tree in comparison with the trade in its product. It Avas only between 17G5 and 1770 that a Ceylon colonist, named de Koke, aided by Falck, the governor of the island, made some planta- tions which were w^onderfully successful. They have diminished in Ceylon in the last few years, but others have been established in the tropical regions of the old and new worlds. The species becomes easily naturalized beyond the limits of cultivation,^ as birds are fond of the fruit, and drop the seeds in the forests. China Grass — Boehnieria nivea, Hooker and Arnott. The cultivation of this valuable Uiiicacea has been introduced into the south of France and of the United States for about thirty years, but commerce had pre- viously acquainted us w^ith the great value of its fibres, more tenacious than hemp and in some cases flexible as silk. Interestino^ details on the manner of cultivatinj]^ ' Thwaitcs, Enum. PI. Zelanice, p. 252 ; Branclis, Forest Flora of India, p. 375. * Fluckiger and Hanburr, Pharmacographia, p. 467 ; Porter, The Tropical Agriculturist p. 268. ^ Brandis, Forest Flora ; Grisebach, Flora of Brit. TV. India Is., p. 179. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 147 the plant and of extracting its fibres ^ may be found in several books ; I shall confine myself here to defining as clearly as I can its geographical origin. To attain this end we must not timst to the vague expressions of most authors, nor to the labels attached to the specimens in herbaria, since frequently no dis- tinction has been made betv/een cultivated, naturalized, or truly wild plants, and the two varieties of Boclimeria nivea (Urtica nivea, Linnaeus), and Boehineria tenacis- sima, Gaudichaud, or B. candicans, Hasskarl, have been confounded together ; forms which appear to be varieties of the same species, because transitions between them have been observed by botanists. There is also a sub- variety, with leaves green on both sides, cultivated by Americans and by M. de Malartic in the south of France. The variety earliest known (Urtica nivea, L.), with leaves white on the under side, is said to grow in China and some neighbouring countries. Linmeus says it is found on walls in China, which would imply a plant naturalized on rubbish-heaps from cultivation. But Loureiro ^ says, " habitat et ahitndanter colitiir in Cochin- Ghina et China,'' and according to Bentham,^ the collector Champion found it in abundance in the ravines of the island of Hono^konor. Accordinor to Franchet and Sava- tier,* it exists in Japan in clearings and hedges (in fruti- cetis umibrosis et sepibiis). Blanco ^ says it is common in the Philippine Isles. I find no proof that it is wild in Java, Sumatra, and other islands of the Malay Archi- pelago. Rumphius ^ knew it only as a cultivated plant. Roxburgh"^ beheved it to be a native of Sumatra, but Miquel ^ does not confirm this belief. The other varieties * De Malartic, Journ. d'Agric. Pratique, 1871, 1872, vol. ii. No. 31; de la Roque, ibid., No. 29, Bull. Soc. d'AccHm., 1872, p. 463; Vilmorin, Bon Jardinier, 1880, pt. 1, p. 700; Vetillart, Etudes sur les Fibres Vegetates Textiles, p. 99, pi. 2. ^ Loureiro, Fl. Cochin., ii. p. 683. ^ Bentham, Fl. Hongkong, p. 331. * Francliet and Savatier, Enum. Plant. Jap., i. p. 439, ' Blanco, Flora de Filip., edit. 2, p. 484. * Rumphius, Amboin, v. p. 214. ^ Roxburgh, Fl. Ind., iii. p. 590. * Miquel, Sumatra, Germ, edit., p. 170. 148 ORIGIN OF CULTIVATED PLA^'TS, have nowhere been found wild, which supports the theory that they are only the result of cultivation. Hemp — Cannabis sativa, Linnseus. Hemp is mentioned, in its two forms, male and female, in the most ancient Chinese works, particularly in the Shu-King, written 500 B.c.-^ It has Sanskrit names, hhanga and gangiJca.^ The root of these words, ang or an, recurs in all the Indo- European and modern Semitic languages : hang in Hindu and Persian, ganga in Bengali,^ haiif in German, hem]^ in English, chanvre in French, kanas in Keltic and modern Breton,^ cannabis in Greek and Latin, cannab in Arabic.^ According to Herodotus (born 484 B.C.), the Scythians used hemp, but in his time the Greeks w^ere scarcely acquainted with it.^ Hiero II., King of Syracuse, bought the hemp used for the cordage of his vessels in Gaul, and Lucilius is the earliest Roman writer who speaks of the plant (100 B.C.). Hebrew books do not mention hemp.*^ It was not used in the fabrics which enveloped the mummies of ancient Eg3^pt. Even at the end of the eighteenth century it was only cultivated in Egypt for the sake of an intoxicating liquid extracted from the plant.^ The compilation of Jewish laws known as the Talmud, made under the Roman dominion, speaks of its textile properties as of a little-known fact.^ It seems probable that the Scythians transported this plant from Central Asia and from Russia when they migrated westward about 1500 B.C., a little before the Trojan war. It may also have been introduced by the earlier incursions of the Aryans into Thrace and Western Europe ; yet in that case it would have been earlier known in Ital}^ Hemp has * Bretschneider, On the Study and Value, etc., pp. 5, 10, 48. 2 Piddington, Index ; Boxburgh, Fl. Ind., edit. 2, vol. iii. p. 772. ^ Roxburgh, ibid. * Reynier, Economie des Celtes, p. 448 ; Legnm'dec, Diet. Baa-Breton. ^ J. Humbert, formorlv professor of Arabic at Geneva, says the name is hannah, kon-nab, hon-nab, hen-nnb, kanedir, according to the locality. ^ Athenacns, quoted by Hehn, Culturpf.anzen, p. 1G8. ^ Rosenmiiller, Hand. Bibl. Alterth. ' Forskal, Flora ; Delile, Flore d'Egypte. * Reynier, Economie des Arabes, p. 434. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 149 not been found in the lake -dwellings of Switzerland ^ and Northern Italy .^ The observations on the habitat of Cannabis sativa agree perfectly with the data furnished by history and philology. I have treated specially of this subject in a monograph in Froclronius, ISGO.^ The species has been found wild, beyond a doubt, to the south of the Caspian Sea,^ in Siberia, near the Irtysch, in the desert of the Kirghiz, beyond Lake Baikal, in Dahuria (government of Irkutsh). Authors mention it also throughout Southern and Central Russia, and to the south of the Caucasus,^ but its wild nature is here less certain, seeing that these are populous countries, and that the seeds of the hemp are easily diffused from gardens. The antiquity of the cultivation of hemp in China leads me to believe that its area extends further to the east, although this has not yet been proved by botanists.^ Boissier mentions the species as " almost wild in Persia." I doubt whether it is indigenous there, since in that case the Greeks and Hebrews Avould have known of it at an earlier period. White Mulberry — Morns alba, Linnaeus. The mulberry tree, which is most commonly used in Europe for rearing silkworms, is Moms alba. Its very numerous varieties have been carefully described by Seringe,'' and more recently by Bureau.*^ That most widely cultivated in India, Morus indica, Linnreus (Morus alba, var. Indica, Bureau), is wild in the Punjab and in Sikkim, according to Brandis, inspector-general of forests in British India.^ Two other varieties, serrata and cuspidata, are also said to be wild in different pro- ' Heer, JJeher d. Flachs, p. 25. 2 Sordelli, Notizie suU. Staz. di Lagozza, ISSO. ' Vol. xvi. sect. 1, p. 30. * De Buuge, Bull. Soc. Bat. de Fr., 1860, p. 30. * Ledebour, Flora Rossica, iii. p. 634. ® Bunge found hemp in the north of China, but among rubbish (Enum. No. 338). ^ Sex'inge, Description et Culture des Muriers. * Bureau, in De Candolle, Prodromus, xvii. p. 238. ' Brandis, Forest Flora of North-West and Central India, 1874, p. 408. This variety has black fruit, like that of 3Iorus nijra. 150 ORIGIN OF CULTIVATED PLANTS. vinces of Northern India.^ The Abb^ David found a perfectly wild variety in Mongolia, described under the name of onongolica by Bureau; and Dr. Bretschneider ^ quotes a name yen, from ancient Chinese authors, for the wild mulberry. It is true he does not say whether this name applies to the white mulberry, pe-sa'iig, of the Chinese planta- tions.^ The antiquity of its culture in China,^ and in Japan, and the number of different varieties grown there, lead us to believe that its original area extended east- Avard as far as Japan; but the indigenous flora of Southern China is little known, and the most trustworthy authors do not affirm that the plant is indigenous in Japan. Franchet and Savatier ^ say that it is " cultivated from time immemorial, and become wild here and there." It is worthy of note also that the white mulberry appears to thrive especially in mountainous and temperate coun- tries, whence it may be argued that it was formerly introduced from the north of China into the plains of the south. It is known that birds are fond of the fruit, and bear the seeds to great distances and into unculti- vated ground, and this makes it difficult to discover its really original habitat. This facility of naturalization doubtless explains the presence in successive epochs of the white mulberry in Western Asia and the south of Europe. This must have occurred especially after the monks brought the silk- worm to Constantinople under Justinian in the sixth century, and as the culture of silkworms was gradually propagated westwards. However, Targioni has proved that only the black mulberry, M. nigra, was known in Sicily and Italy when the manufacture of silk was intro- duced into Sicily in 1148, and two centuries later into ^ Bureau, ihid., from the specimens of several travellers. ' Bretsebneider, Study and Value, etc., p. 12. ' This name occurs in the Fent-sao, according to Eitter, Erdhiinde, xvii. p. 489. * Piatt says (ZeitschH/t d. Ge.'^eUsch. Erdlwnde, 1871, p. 162) that its cultivation dates from 4000 years n.c. * Franchet and Savatier, Enum. Plant. Jap., i. p. 433. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 151 Tuscany.-^ According to the same author, the introduction of the white mulberry into Tuscany dates at the earliest from the year 1340. In like manner the manufacture of silk may have begun in China, because the silkworm is natural to that country ; but it is very probable that the tree grew also in the north of India, where so many travellers have found it wild. In Persia, Armenia, and Asia Minor, I am inclined to believe that it was natura- lized at a very early epoch, rather than to share Grise- bach's opinion that it is indigenous in the basin of the Caspian Sea. Boissier does not give it as wild in that region.^ Buhse^ found it in Persia, near Erivan and Bashnaruschin, and he adds, *' naturalized in abundance in Ghilan and Masenderan." Ledebour,^ in his Russian flora, mentions numerous localities round the Caucasus, but he does not specify whether the species is wild or naturalized. In the Crimea, Greece, and Italy, it exists only in a cultivated state.^ A variety, tatarica, often cultivated in the south of Russia, has become naturalized near the Volga.^ If the white mulberry did not originally exist in Persia and in the neighbourhood of the Caspian Sea, it must have penetrated there a long while ago. I may quote in proof of this the name tid, tutti, tuta, which is Persian, Arabic, Turkish, and Tartar. There is a Sanskrit name, tida,'^ which must be connected with the same root as the Persian name ; but no Hebrew name is known, which is a confirmation of the theory of a successive extension towards the west of Asia. I refer those of my readers who may desire more de- tailed information about the introduction of the mulberry and of silkworms to the able works of Tarofioni and * Ant. Targioni, Cenni Storici sidV Introduzione di Varie Piante nelV ArjricoUura Toscana, p, 188. 2 Boissier, Fl. Orient., iv. p. 1153. ' Buhse, Aufzahhmg der Transcaucasien und Persien PJlanzen, p. 203. * Ledebour, Fl. Ross., iii. p. 643. * Steven, Verseichniss d. Taurisch. Halhins, p. 313 ; Heldreich, PJlan- zen des Attischen Ebene, -p. 508; Bertoloni, Fl. Ital., x. p. 177 j Caruel, F . Toscana, p. 171. " Bureau, de Cand., Prodr., xvii. p. 238. ' Roxburgh, FL Ind. ; Piddington, Index. 152 ORIGIN OF CULTIVATED PLANTS. Ritter, to which I have already referred. Recent dis- coveries made by various botanists have permitted me to add more precise data than those of Hitter on the question of origin, and if there are some apparent contra- dictions in our opinions on other points, it is because the famous geographer has considered a number of varieties as so many different species, whereas botanists, after a careful examination, have classed them together. Black Mulberry — Morus nigra, Linnreus, This tree is more valued for its fruit than for its leaves, and on that account I should have included it in the list of fruit trees ; but its history can hardly be separated from that of the white mulberry. Moreover, its leaves are employed in many countries lor the feeding of silkworms, although the silk produced is of inferior quality. The black mulberry is distinguished from the Avliite by several characters independently of the black colour of the fruit, which occurs also in a few varieties of the M. alba} It has not a great number of varieties like the latter, which argues a less ancient and a less general cultivation and a narrower primitive area. Greek and Latin authors, even the poets, have men- tioned Morns nigra, which they compare to Ficiis syco- onorus, and which they even confounded originally with this Egyptian tree. Commentators for the last two centuries have quoted a number of passages which leave no doubt on this head, but which are devoid of interest in themselves.^ They furnish no proof touching the origin of the species, which is presumably Persian, unless Ave are to take seriously the fable of Pyramus and Thisbe, of which the scene was in Babylonia, according to Ovid. Botanists have not yet furnished any certain proof that this species is indigenous in Persia. Boissier, who is the most learned in the floras of the East, contents * Eeichenbach gives good figures of both species in his Icones Fl, Germ., G57, 658. ' Fraas, Syn. Fl. Class., p. 230 ; Lenz, Bof. der Alien Gr. iind Foin., ]). 419; Hitter, Erdkunde, xvii. p. 482 j Hebn, Culturpfianzen, edit. 3, p- 336. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 153 himself with quoting Hohenacker as the discoverer of 31. nigra in the forests of Lenkoran, on the south coast of the Caspian Sea, and he adds, " probably wild in the north of Persia near the Caspian Sea." ^ Ledebour, in his Kussian flora, had previously indicated, on the authority of different travellers, the Crimea and the provinces south of the Caucasus ; ^ but Steven denies the existence of the species in the Crimea except in a cultivated state.^ Tchi- hatcheff and Koch found the black mulberry in high wild districts of Armenia. It is very probable that in the reofion to the south of the Caucasus and of the Caspian Sea Alorus nigra is wild and indigenous rather than naturalized. What leads me to this belief is (1) that it is not known, even in a cultiva^ted state, in India, China, or Japan ; (2) that it has no Sanskrit name ; (3) that it was so early introduced into Greece, a country which had intercourse with Armenia at an early period.* Morns nigra spread so little to the south of Persia, that no certain Hebrew name is known for it, nor even a Persian name distinct from that of Morus alhoj. It was widely cultivated in Italy until the superiority of the white mulberry for the rearing of silkworms was recognized. In Greece the black mulberry is still the most cultivated.^ It has become naturalized here and there in these countries and in Spain.^ American Aloe — Agave Americana, Linn?eus. This ligneous plant, of the order of Amari/lliclacea3, has been cultivated from time immemorial in Mexico under the names maguey or Trietl, in order to extract from it, at the moment when the flower stem is developed, the wine known as pulque. Humboldt has given a full descrip- tion of this culture/ and he tells us elsewhere ® that the » Boissier, Fl. Orient., iv. p. 1153 (published 1879). « Ledebour, Fl. Ross., iii. p. 641. ' Steven, Verseichniss d. Taur. Halh. Pflan., p. 313. * TchihatchefP, trans, of Grisebach's Vegetation du GlohCf i. 424. ' Heldreicli, Nutzpflanzen Griechenlands, p. 19. « Bertoloui, Flora Hal., x. p. 179; Viviaui, Fl. Dalmat., i. p. 220; Willkomm and Lange, Prodr. Fl. Hisp., i. p. 250. ^ Humboldt, Nouvelle Espagne, ed. 2, p. 487. * Humboldt, in Kuntli, Nova Genera, i. p. 297. 154 ORIGIN OF CULTIVATED PLANTS. species grows in the whole of South America as far as five thousand feet of altitude. It is mentioned^ in Jamaica, Antigua, Dominica, and Cuba, but it must be observed that it multiplies easily by suckers, and that it is often planted far from dwellings to form fences or to extract from it the fibre known as pite, and this makes it difficult to ascertain its orio-inal habitat. Transported long since into the countries Avliich border the Mediterranean, it occurs there with every appearance of an indigenous species, although there is no doubt as to its origin.^ Probably, to judge from the various uses made of it in Mexico before the arrival of the Euro- peans, it came originally from thence. Sugar-Cane — Saccharum ojfficinarum, Linnaeus. The origin of the sugar-cane, of its cultivation, and of the manufacture of sugar, are the subject of a very remarkable work by the geographer, Karl Rittcr.^ I need not follow his purely agricultural and economical details ; but for that which interests us particularly, the primitive habitat of the species, he is the best guide, and the facts observed during the last forty years for the most part support or confirm his opinions. The sugar-cane is cultivated at the present day in all the warm regions of the j^lobe, but a number of historical facts testify that it was first grown in Southern Asia, whence it spread into Africa, and later into America. The question is, therefore, to discover in what districts of the continent, or in which of the southern islands of Asia, the plant exists, or existed at the time it was first employed. Ritter has followed the best methods of arrivmg at a solution. He notes first that all the species known in a ' Grisebach, FI. of Brit. W. Ind. Is., p. 582. ' Alph. de Caudolle, Geoijr. Bot. Raisonn4e, p. 739; H. Hoffmann, in Kegel's Gartcvfiora, 1875, p. 70. • K. Ritter, Ueher die Geographische Yerhreitung des Zxic'kerrohrs, in 4to, 108 pages (according to Pritzel, Thes. Lit. Bot.) ; Die Cultur des Zuckerrohrs, Saccharum, in Asien, Geogr. Verbreitungj etc., etc., in 8vo, 64 pages, Avitbout date. This monograph is full of learning and judgment, worthy of the best epoch of German science, when English or French authors were quoted by all authors with as much care as Germans. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 155 wild state, and undoubtedly belonging to the genus Bac- chariun, grow in India, except one in Egypt.^ Five species have since been described, growing in Java, New Guinea, Timor, and the Philippine Isles.'-^ The proba- bilities are all in favour of an Asiatic origin, to judge from the data furnished by geographical botany. Unfortunately no botanist had discovered at the time when Ritter wrote, or has since discovered, Saccharum oficinaruTii wild in India, in the adjacent countries or in the archipelago to the south of Asia. All Anglo- Indian authors, Roxburgh, Wallich, Royle, etc., and more recently Aitchison,^ only mention the plant as a culti- vated one. Roxburgh, who was so long a collector in India, says expressly, " where wild I do not know." The family of the Graminece has not yet appeared in Sir Joseph Hooker's flora. For the island of Ceylon, Thwaites does not even mention the cultivated plant.'* Rumphius, who has carefully described its cultivation in the Dutch colonies, says nothing about the home of the species. Miquel, Hasskarl, and Blanco mention no wild specimen in Sumatra, Java, or the Philippine Isles. Crawfurd tried to discover it, but failed to do so.^ At the time of Cook's voyage Forster found the sugar-cane only as a cultivated plant in the small islands of the Pacific.^ The natives of New Caledonia cultivate a number of varieties of the sugar-cane, and use it constantly, sucking the syrup from the cane ; but Vieillard "^ takes care to say, " From the fact that isolated plants of Saccharum offici- narum are often found in the middle of the bush and even on the mountains, it would be wrong to conclude that the plant is indigenous ; for these specimens, poor and weak, only mark the site of old plantations, or ^ Kuntli, Einim. Plant. (1838), vol. i. p. 474. There is no more recent descriptive work on the family of the Graminece, nor the geuu;^ Saccharum. * Miquel, Florce Indice Batavo?, 1855, vol. iii. p. 511. ' Aitchison, Catalogue of Punjab and Sindh Plants, 18G9, p. 173. * Thwaites, Ennm. PI. Zeylonics. - Crawfurd, Indian Archip.^ i. p. 475. " Forster, De Plantis Esctdentis. ' Vieillard, Annales des Sc. Nat , 4th series, vol. xyi. p. 32. 8 156 ORIGIN OF CULTIVATED PLANTS. are sprung from fragments of cane left by the natives, who seldom travel without a piece of cane in the hand." In 1861, Bentham, who had access to the rich herbarium of Kew, says, in his Flora of Hongkong, " We have no authentic and certain proof of a locality where the common sugar-cane is wild." I do not know, however, why Ritter and every one else has neglected an assertion of Loureiro, in his Flora of Cochin-China} " Habitat, et colitur abundantissime in omnibus provinciis regni Cochin-Chinensis : simul in aliquibus imperii sinensis, sed minori copia." The word habitat, separated by a comma from the rest, is a distinct assertion. Loureiro could not have been mistaken about the Saccharuiii officinarnni, which he saw cultivated all about him, and of which he enumerates the principal varieties. He must have seen plants wild, at least in appearance. They may liave spread from some neigh- bouring plantation, but I know nothing which makes it unlikely that the plant sliould be indigenous in this warm moist district of the continent of Asia. Forskal ^ mentions the species as wild in the moun- tains of Arabia, under a name which he believes to be Indian. If it came from Arabia, it would have spread into Egypt long ago, and the Hebrews would have known it. Roxburoh had received in the botanical crardons of Calcutta in 1796, and had introduced into the planta- tions in Bengal, a Saccharum to which he gave the name of S. sinense, and of which he published an illustration in his great work Planta3 Coromandeliance, vol. iii. pi. 232. It is perhaps only a form of S. officinavuin, and moreover, as it is only known in a cultivated state, it tells nothing about the primitive country either of this or of any other variety. A few botanists have asserted that the suo-ar-cane iloAvers more often in Asia than in America or Africa, and even that it produces seed ^ on the banks of the * Loureiro, Cochin-Ch., edit. 2, vol. i. p. 66. ' Forskal, Fl. JEgypto-Arahica, p. 103. ' ]\In(.-fa(!yen, On the Botanical Characters of the S\igar-Cane, in Hooker's Bot. Miscell., i. p. 101 ; Maycock, Fl. Barbad., p. 50, PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 157 Ganges, which they regard as a proof that it is indigenous. Macfadj^en says so without giving any proof. It was an assertion made to him in Jamaica by some traveller ; but Sir W. Hooker adds in a note, " Dr. Eoxburgh, in spite of his long residence on the banks of the Ganges, has never seen the seeds of the sugar-cane." It rarely Howers, and still more rarely bears fruit, as is commonly the case with plants propagated by buds or suckers, and if any variety of sugar-cane Ave re disposed to seed, it would probably be less productive of sugar and would soon be abandoned. RumphiuS; a better observer than many modern botanists, has given a good description of the cultivated cane in the Dutch colonies, and makes an interesting remark.^ " It never produces flowers or fruit unless it has remained several years in a stony place." Neither he, nor any one else to my knowledge, has de- scribed or drawn the seed. The flower, on the contrary, has often been figured, and I have a fine specimen from Martinique.^ Schacht is the only person who has given a good analysis of the flower, including the pistil ; he had not seen the seed ripe.^ De Tussac,^ who gives a poor analysis, speaks of the seed, but he only saw it ' young in the ovary. In default of precise information as to the native country of the species, accessory means, linguistic and historical, of proving an Asiatic origin, are of some interest. Ritter gives them carefully ; I will content myself with an epitome. The Sanskrit name of the sugar- cane was iksku, ikshiira, or ikshava, but the sugar was called sarhara, or sakkara, and all its names in our Euro- pean languages of Aryan origin, beginning with the ancient ones — Greek, for example — are clearly derived from this. This is an indication of Asiatic orio-in, and that the produce of the cane -was of ancient use in the southern regions of Asia with which the ancient Sanskrit-speak- ing nation may have had commercial dealings. The two Sanskrit words have remained in Bengali under the * Ramphius, Amhoin, voL v. p. 186. * Hehn, No. 480. * Schacht, Madeira und Teneriffe, tab. i. * Tussac, Flore des Antilles^ i. p. 153, pi. 23. 158 ORIGIN OF CULTIVATED PLANTS. forms ifc and akh} But in other languages beyond the Indus, we find a singular variety of names, at least when they are not akin to that of the Aryans ; for instance : pancKadara in Telinga, kyam in Burmese, mia in the dialect of Cochin-China, lean and tche, or tsche, in Chinese ; and further south, among the Malays, titbit or tabu for the plant, and gula for the product. This diversity proves the great antiquity of its cultivation in those regions of Asia in which botanical indications point out the origin of the species. The epoch of its introduction into different countries aijrees with the idea that its ori^^^in was in India, Cochin- China, or the Malay Archipelago. The Chinese were not acquainted with the sugar-cane at a very remote period, and they received it from the West. Ritter contradicts those authors who speak of a very ancient cultivation, and I find most positive con- firmation of his opinion in Dr. Bretschneider's pamphlet, drawn up at Pekin with the aid of all the resources of Chinese literature.^ " I have not been able to discover," he says, "any allusion to the sugar-cane in the most ancient Chinese books (the five classics)." It appears to have been mentioned for the first time by the authors of the second century before Christ. The first description of it appears in the Nan-fang-tsao-mu-chuang, in the fourth century : " The cite che, han-cJie (Jean, sweet, chc\ bamboo) grows," it says, " in Cochin-China. It is several inches in circumference, and resembles the bamboo. The stem, broken into pieces, is eatable and very sweet. The sap which is drawn from it is dried in the sun. After a few days it becomes sugar (here a compound Chinese character), which melts in the mouth. ... In the year 286 (of our era) the kingdom of Funan (in India, beyond the Ganges) sent sugar as a tribute." According to the Pent- Sao, an emperor who reigned from 627 to 650 A.D., sent a man into the Indian province of Behar to learn how to manufacture sugar. There is nothing said in these works of the plant * Piddington, Index. * Bretschn eider, On the Study and Value, etc., pp. 45-47. PLANTS CULTIVATED FOR THEIR STEMS OR LEAVES. 159 growing wild in China ; on the contrary, the origin in Cochin-China, indicated by Loureiro, finds an unexpected confirmation. It seems to me most probable that its primitive range extended from Bengal to Cochin-China. It may have included the Sunda Isles and the Moluccas, whose climate is very similar ; but there are quite as many reasons for believing that it was early introduced into these from Cochin-China or the Malay peninsula. The propagation of the sugar-cane from India west- ward is well known. The Greco-Roman world had a vague idea of the reed (calamus) which the Indians delighted to chew, and from which they obtained sugar.^ On the other hand, the Hebrew writings do not mention sugar ; ^ whence we may infer that the cultivation of the suofar-cPcue did not exist west of the Indus at the time of the Jewish captivity at Babylon. The Arabs in the Middle Ages introduced it into Egypt, Sicily, and the south of Spain,^ where it flourished until the abundance of suiiar in the colonies caused it to be abandoned. Don Henriquez transported the sugar-cane from Sicily to Madeira, whence it was taken to the Canaries in 1503.* Hence it was introduced into Brazil in tlie beginning of the sixteenth century.^ It was taken to St. Domingo about 1520, and shortly afterwards to Mexico;^ to Guadeloupe in 1644, to Martinique about 1650, to Bour- bon when the colony was founded.'^ The variety known as Otahiti, which is not, however, wild in that island, and which is also called Bourhorif was introduced into the French and English colonies at the end of the last and the beginning of the present century.^ * See the quotations from Strabo, Dioscorides, PIitjj, etc., in Lenz, Botanik der Alien Griechen und Romery 1859, p. 2G7 ; Fingerhut, in Flora, 1839, vol. ii. p. 529 ; and many other authors. ' Rosenmiiller, Handhuch der Bihl. Alterth. ' Calendrier Rural de Uarihy written in the tenth century for S[>ain, translated by Dureau de la Malle in his Cliniatologie de Vltalie et de VA7idalousie, p. 71. * Von Buch, Canar. Ins. * Piso, Bresil, p. 4.9. * Humboldt, Nouv. E^jparjne, ed. 2, vol. iii. p. 34. ' Not. Stat, sur les Col. Franc, i. pp. 207, 29, 83. "* Macfadyen, in Hooker, Bot. Miscell., i. p. 101 j May cock, Fl. Barhad., p. 50. IGO ORIGIN OF CULTIVATED PLANTS. The processes of cultivation and preparation of the sugar are described in a number of works, among which the following may be recommended : de Tussac, Flore des Antilles, 3 vols., Paris; vol. i. pp. 151-182; and Macfadyen, in Hooker's Boianical Miscellany, 1830, vol. i. pp. 103-llG. en AFTER HI. PLANTS CULTIVATED FOR THEIR FLOWERS, OR FOR THE ORGANS WHICH ENVELOP THEM. Clove — Caryophyllus aromaticus, Linngsus. The clove used for domestic purposes is the calix and flower-bud of a plant belonging to the order of Myr- tacese. Although the plant has been often described and very well drawn from cultivated specimens, some doubt remains as to its nature when wild. I spoke of it in my Geogixtphical Botany in 1855, but it does not appear that the question has made any further progress since then, which induces me to repeat here what I said then. " The clove must have come originally from the Moluc- cas," as Rumphius asserts,^ for its cultivation was limited two centuries ago to a few little islands in this archipelago. I cannot, however, find any proof that the true clove tree, with peduncles and aromatic buds, has been found in a wild state. Rumphius ^ considers that a plant of which he gives a description, and a drawing under the name Caryophyllumi sylvestre, belongs to the same species, and this plant is wild throughout the Moluccas. A native told him that the cultivated clove trees degenerate into this form, and Rumphius himself found a plant of G. sylvestre in a deserted plantation of cultivated cloves. Nevertheless plate 3 differs from plate 1 of the cultivated clove in the shape of the leaves and of the teeth of the calix. I do not speak of plfite 2, which appears to be an * ii. p. 3. ' iu tab. 3. 1G2 ORIGIN OF CULTIVATED PLANTS. abnormal form of the cultivated clove. Kumphius says that C. sylvestre has no aromatic properties ; now, as a rule, the aromatic properties are more developed in the wild plants of a species than in the cultivated plants. Sonnerat -^ also publishes figures of the true clove and of a spurious clove found in a small island near the country of the Papuans. It is easy to see that his false clove differs completely by its blunt leaves from the true clove, and also from the two species of Rumphius. I cannot make up my mind to class all these different plants, wild and cultivated, together, as all authors have done.^ It is especially necessary to exclude plate 120 of Sonnerat, which is admitted in the Botanical Magazine. An historical account of the cultivation of the clove, and of its introduction into different countries, will be found in the last-named work, in the Dictionnaire d' Agriculture, and in the dictionaries of natural history. If it be true, as Roxburgh says,^ that the Sanskrit language had a name, luvunga, for the clove, the trade in this spice must date from a very early epoch, even supposing the name to be more modern than the true Sanskrit. But I doubt its genuine character, for the Romans would have known of a substance so easily trans- ported, and it does not appear that it was introduced into Europe before the discovery of the Moluccas by the Portuguese. Hop — Huvvulus Liipnlus, Linnaeus. The hop is wild in Europe from England and Sweden as far south as the mountains of the Mediterranean basin, and in Asia as far as Damascus, as the south of the Caspian Sea, and of Eastern Siberia,^ but it is not found in India, the north of China, or the basin of the river Amur.^ * Sonnerat, Voy. Nouv. Guin., tab. 119, 120. 2 Thuuberg, Diss., n. p. 326 ; De Candolle, Prodi:, iii. p. 202 ; Hooker, Bot. Mag., tab. 2749 ; Hasskarl, Cat. Hort. Bogor. Alt., p. 2G1. ' Eoxburgh, Flora Iiidica, edit. 1832, voL ii. p. 194. ^ Alph. de Candolle, in Prodronms, vol. xvi., sect. 1, p. 29 ; Boissier, Fl. Orient., iv. Y). 1152 ', Hohenacker, Enum. Plant. TaZj/sc/i, j). 30 ; Buhse Aufzdhlung Transcaucasien^ p. 202. * An erroneous transcription of what Asa Gray (Botany of North. United States, edit. 5) says of tlie hemp, wronfrly attributed to the hop in Prodromus, and repeated in the French edition of this work, should PLANTS CULTIVATED FOR THEIR FLOWERS, ETC. 1G3 In spite of the entirely wild appearance of the hop in Europe in districts far from cultivation, it has been some- times asked if it is not of Asiatic orioin.-^ I do not think this can be proved, nor even that it is likely. The fact that the Greeks and Latins have not spoken of the use of the hop in making beer is easily explained, as they were almost entirely unacquainted with this drink. If the Greeks have not mentioned the plant, it is simply perhaps because it is rare in their country. From the Italian name liipulo it seems likely that Pliny speaks of it with other vegetables under the name luims salictarius.^ That the custom of brcAving with hops only became general in the Middle Ages proves nothing, except that other plants were formerly employed, as is still the case in some districts. The Kelts, the Germans, otlier peoples of the north and even of the soutli who had the vine, made beer ^ either of barley or of other fermented grain, addino; in certain cases different ve2;etable substances — the bark of the oak or of the tamarisk, for instance, or the fruits of Mi/rica gale} It is very possible that they did not soon discover the advantages of the hop, and that even after these were recognized, they employed wild hops before beginning to cultivate them. The first men- tion of hop-gardens occurs in an act of donation made by Pepin, father of Charlemagne, iji 7G8.^ In the fourteenth century it was an important object of culture in Germany, but it began in England only under Henry VIII.^ The common names of the hop only furnish negative indications as to its origin. There is no Sanskrit name,'^ be corrected. Humulus Lxtpuhus is indigenous in the east of the United States, and also in the island of Yeso, according to a letter from Maximowicz. — Author's Note, 1884. * Hehn, Nutzpflanzen und Hausthiere in ihren Uehergang aus Asien, edit. 3, p. 415. ^ Piiny, Hid., bk. 21, c. 15. He mentions asparagus in this con- nection, and the young shoots of the hop are s(nietinies eaten in this manner. ' Tacitus, Germania, cap. 25; Pliny, bk. 18, c. 7; Hehn, Kultur- pjlanzen, edit. 3, pp. 125-137. * Volz, Beitrage zur Cidfurgeschichte, p. 149. ® Ibid. ' Beckmann, Erfindungen, quoted by Vol?;. ' Piddiugton, hidex ; Pick, Worterl). Indo-Germ. Sprachen, i. ; Ur- spraolie. 1G4j origin of cultivated plants. and this agrees with the absence of the species in the regie n of the Himalayas, and shows that the early Aryan peoples had not noticed and employed it. I have quoted before ^ some of the European names, showing their diversity, although some few of them may be derived from a com- mon stock. Hehn, the philologist, has trea^ted of their etymology, and shown how obscure it is, but he has not mentioned the names totally distinct from humle, hopf or hop, and chneli of the Scandinavian, Gothic, and Slav races ; for example, Apini in Lette, Apivynis in Lithua- nian, tap in Esthonian, hliist in Illyrian,^ which have evidently other roots. This variety tends to confirm the theory that the species existed in Europe before the arrival of the Aryan nations. Several different peoples must have distinguished, knoA^Ti, and used this plant suc- cessively, which confirms its extension in Europe and in Asia before it w^as used in brewing. Carthamine — Carthamus ti ndori us, Linuc^us. The composite annual which produces the dye called carthamine is one of the most ancient cultivated species. Its flowers are used for dyeing in red or yellow, and the seeds yield oil. The grave-cloths w^hich wrap the ancient Egyptian mummies are dyed with carthamine,^ and quite recently fragments of the plant have been found in the tombs discovered at Deir el Bahari.* Its cultivation must also be ancient in India, since there are tw^o Sanskrit names for it, cusumhha and kmnalottarcn, of which the first has several derivatives in the modern lano^uao^es of tlie peninsula.^ The Chinese only received carthamine in the second century B.C., when Chang-kien brought it back from Bactriana.^ The Greeks and Latins were probably not acquainted with it, for it is very doubtful whether this is the plant wdiich they knew as ciiikos or cnicns? At a later period the Ai^abs contributed largely ^ A. de Candolle, Geogr. Bot. Rais., p. 857. ' Diet. MS., compiled from floras, Moritzi. ' Unger, Die Pjlanzen des Alien ^gyptens, p. 47. * Schweinf urth, in a letter to M. Boissiei-, 1882. ^ Piddiugton, Index. • Bret Schneider, Study and Value, etc., p. 15. ' See Targioni, Cenni Storici, p. 108. PLANTS CULTIVATED FOR THEIR FLOWERS, ETC. 165 to diffuse the cultivation of carthamine, which they named qortoUy kurturii, whence carthamine, or usfur, or ihridh, or movahu} a diversity indicating an ancient existence in several countries of Western Asia or of Africa. The progress of chemistry threatens to do away with the cultivation of this plant as of many others, but it still subsists in the south of Europe, in the East, and throughout the valley of the Nile.^ No botanist has found the carthamine in a really wild state. Authors doubtfully assign to it an origin in India or Africa, in Abyssinia in particular, but they have never seen it except in a cultivated state, or with every appearance of having escaped from cultivation.^ Mr. Clarke,^ formerly director of the Botanical Gardens in Calcutta, who has lately studied the CompositcB of India, includes the species only as a cultivated one. The summary of our modern knowledge of the plants of the Nile region, including Abyssinia, by Schweinfarth and Ascherson,^ only indicates it as a cultivated species, nor does the list of the plants observed by Rohlfs on his recent journey mention a wild carthamine.^ As the species has not been found wild either in India or in Africa, and as it has been cultivated for thousands of years in both countries, the idea occurred to me of seeking its origin in the intermediate region ; a method which had been successful in other cases. Unfortunately, the interior of Arabia is almost un- known. Forskal, who has visited the coasts of Yemen has learnt nothing about the carthamine ; nor is it mentioned among the plants of Botta and of Bove. But an Arab, Abu Anifa, quoted by Ebn Baithar, a thirteenth- century writer, expressed himself as follows : "^ — " Usfur, this plant furnishes a substance used as a dye ; there are two kinds, one cultivated and one wild, which both grow ^ Forslcal, Fl. Mgy%>t., p. 73 ; Ebn Baithar, Germ, trans., ii. pp. 196, 293 ; i. p. 18. ' * See Gasparin, Cours d'Agric, iv. p. 217. ' Boissier, Fl. Orient., in. p. 710 ; Oliver, Flora of Trop. Afr., iii. p. 439. * Clarke, Compositce Tndiccn, 1876, p. 241. * Schweinfurth and Ascherson, Aufzdhhmg. p. 28*^. " Rohlfs, Kufra, in 8vo, 1881. ^ eIl Baith ir, ii. p. 106. 166 ORIGIN OF CULTIVATED PLANTS. in Arabia, of wliicli the seeds are called eUcurthum" Abu Anifa was very likely right. Saffron — Crocus sativus, Linnaeus. The saffron was cultivated in very early times in the west of Asia. The Romans praised the saffron of Cilicia, which they preferred to that gi'own in Italy.^ Asia Minor, Persia, and Kashmir have been for a long time the countries wdiich export the most. India gets it from Kashmir^ at the present day. Roxburgh and WaDich do not mention it in their works. The two Sanskrit names mentioned by Piddington ^ probably applied to the substance saffron brought from the West, for the name kasruirajamma appears to indicate its origin in Kashmir. The other name is Icunkuma. The Hebrew word karJcom is commonly translated saffron, but it more probably applies to earth amine, to jiulge from the name of the latter in Arabic.^ Besides, the saffron is not cultivated in Egypt or in Arabia. The Greek name is krohos.^ Saffron, AAdiich recurs in all modern European languages, comes from the Arabic sahafaran^ zafran? The Spaniards, nearer to the Arabs, call it azafran. The Arabic name itself comes from assfar, yellow. Trustwortliy authors say that G. sativus is wild in Greece^ and in the Abruzzi mountains in Italy .^ Maw, who is preparing a monograph of the genus Crocus, based on a long series of observations in gardens and in herbaria, connects wdth C. sativus six forms which are found wild in mountainous districts from Italy to Kurdistan. None of these, he says,^^ are identical with the cultivated variety; but certain forms described under other names (0. Orisnii, G. Cartwrightianus, G. Thomasii), hardly differ from it. These are from Italy and Greece. * Pliny, bk. xxi. c. 6. ' Eoyle, III. IlimaL, p. 372. ^ Index, p. 25. * Accorcliug to Forskal, Dclile, Rejnicr, Scliweinfurth, and Ascherson. * Theophrastus, Hist., 1. 6, c. 6. « J. Bauhin, Hist, ii. p. 637. ^ Royle, III Himal. * Sibtliorp, Prodr. ; Fraas, Syn. Fl. Class., p. 292. ^ J. Gay, quoted by Babin^ton, Man. Brit. Fl. ^^ Maw, in the Gardener's Chron., 1881, vol. xvi. PLANTS CULTIVATED FOR THEIR FLOWERS, ETC. 167 The cultivation of saffron, of wliicli the conditions are given in the Coiirs cVAfjricidture by Gasparin, and in tiie Bulletin de la Societe d' Acclimatation for 1870, is becoming more and more rare in Europe and Asia.-^ It has sometimes had the effect of naturalizing the species for a few years at least in localities where it appears to be wild, * Jacquemont, Voyage, vol. iii. p. 238. CHAPTER lY. PLANTS CULTIVATED FOR THEIR FRUITS.^ Sweet Sop, Sugar Apple ^ — Aiiona squamosa, Linn?eus. (In British India, Custard Apple ; but this is the name of Anona niuricata in America.) The orioinal liome of this and other cultivated Anonacea3 has been the subject of doubts, wliich make it an interesting problem. I attempted to resolve them in 1855. The opinion at which I then arrived has been confirmed by the subsequent observations of travellers, and as it is useful to show how far probabihties based upon sound methods lead to true assertions, I will trans- cribe what I then said,^ mentioning afterwards the more recent discoveries. " Robert Brown proved in 1818 that all the species of the genus Anona, excepting Anona senegalensis, belong^ to America, and none to Asia. Aug. de Saint- Hilaire says that, according to Veliozo, A. squamosa was introduced into Brazil, that it is known there under the name of pinha, from its resemblance to a fir-cone, and of ata, evidently borrowed from the names attoa and atis, which are those of the same plant in Asia, and which belong to Eastern languages. Therefore, adds de * The word frait is here employed in the vulgar sense, for any fleshy part "vvhich enlarges after the flowering. In the strictly botanit-al sense, the Anonacese, strawberries, cashews, pine-apples, and breadfruit are not fruits. ^ A. squamosa is figured in Descourtilz, Flore des Antilles, ii. pi. 83 ; Hooker's Bot. Mag., 3095 ; and Tussac, Flore des Antilles, iii. pi. 4. ' A, de Candolle, Geogr. Bat. Rais., p. 859. PLANTS CULTIVATED FOR THEIR FRUITS. 169 Saint-Hilaire,^ the Portuguese transported A, squamosa from their Indian to their American possessions, etc." Having made in 1832 a review of the family of the Anonacese,^ I noticed how Mr. Brown's botanical argument was ever growing stronger; for in spite of the considerable increase in the number of described Anonacese, no Anona, nor even any species of Anonacese with united ovaries, had been found to be a native of Asia. I admitted^ the probability that the species came from the West Indies or from the neighbouring part of the American continent ; but I inadvertently attributed this opinion to Mr. Brown, who had merely indicated an American origin in general.^ Facts of different kinds have since confirmed this view. " Avona squamosa has been found wild in Asia, apparently as a naturalized plant ; in Africa, and espe- cially in America, with all the conditions of an indigenous plant. In fact, according to Dr. Boyle,^ the species has been naturalized in several parts of India ; but he only saw it apparently growing wild on the side of the moun- tain near the fort of Adjeegurh in Bundlecund, among teak trees. When so remarkable a tree, in a country so thoroughly explored by botanists, has only been discovered in a single locality beyond the limits of cultivation, it is most probable that it is not indigenous in the country. Sir Joseph Hooker found it in the isle of St. lago, of the Cape Verde group, forming woods on the hills which over- look the valley of St. Domingo.^ Since A. sqiiaraosa is only known as a cultivated plant on the neighbouring continent ; "^ as it is not even indicated in Guinea by Thonning,^ nor in Congo,^ nor in Senegambia,^^ nor in * Ansf. de Saint-Hilaire, Plantes usuelles des Br4silip.ns, bk. vi. p. 5. ' Alph. de Candolle, Mem. Soc. Phys. et d'Hist. Nat. de Geneve. ' Ihid., p. 19 of Mem. printed separately. * See Botany of Congo, and the German translation of Brown's works, which has alphabetical tables. * Royle, III. Bimal., p. 60. « Webb, in Fl. Nijr., p. 97. ' nid., p. 204. * Thonuiug, PI. Guin. ' Brown, Congo, p. 6. " Guillemin, Perrottet, and Richard, Tentamen FL Seneg. 170 ORIGIN OF CULTIVATED PLANTS. Abyssinia and Egypt, wliicli proves a recent introduction into Africa ; lastly, as the Cape Verde Isles have lost a great part of their primitive forests, I believe that this is a case of naturalization from seed escaped from gardens. Authors are agreed in considering the species wild in Jamaica. Formerly the assertions of Sloane^ and Brown^ might have been disregarded, but they are confirmed by Macfadyen.^ Martins found the species wild in the virgin forests of Para.* He even says, * Sylvescentciin in oiemorihiis ijaraensihus inveiii' whence it may be in- ferred that these trees alone formed a forest. S2:)litgerber^ found it in the forests of Surinam, but he says, 'An sjyontanea?' The number of localities in this part of America is significant. I need not remind my readers that no tree growing elsewhere than on the coast has been found truly indigenous at once in tropical Asia, Africa, and America.^ The result of my researches renders such a fact almost impossible, and if a tree were robust enough to extend over such an area, it would be extremely common in all tropical countries. "Moreover, historical and philological facts tend also to confirm the theory of an American origin. The details given by Rumphius "^ show that Anona squamosa was a plant newly cultivated in most of the islands of the Malay Archipelago. Forster does not mention the culti- vation of any Anonacea in the small islands of the Pacific.^ Rheede ^ says that A. squamosa is an exotic in Malabar, but was brought to India, first by the Chinese and the Arabs, afterwards by the Portuguese. It is cer- tainly cultivated in China and in Cochin-China,^^ and in the Philippine Isle.s,^^ but we do not know from wliat epoch. It is doubtful whether the Arabs cultivate it.^^ ^ Sloane, Jam., ii, p. 168. 2 p Bro-n-n, Jam., p. 257. ^ Macfadyen, Fl. Jam., p. 9. ■* Martius, Fl. Bras., fasc. ii. p. 15. * Splitgerber, Nederl. Kruidk. Arch., ii. p. 230. * A. de Candolle, Geogr. Bot. Bais., cliap, x. ' Rnmpliius, i. p. 139. « Forster, Plantce Esculenta;. « Eheede, Malabar, iii. p. 22. " Loureiro, Fl. Cochin., p. 427. " Blanco, Fl. Filip. " This depends upon the opinion formed with respect to A. glabra, ForskaJ {A. Asiatica, B. Dun. Anon., p. 71 ; A. Forslcalii, D. C. Syst., i. p. 472), which was sometimes cultivated in gardens in Egypt when PLANTS CULTIVATED FOR THEIR FRUITS. 171 It was cultivated in India in Roxburgh's day ; ^ he had not seen the wild plant, and only mentions one common name in a modern language, the Bengali ata, which is already in Rheede. Later the name gunda-gatra ^ was believed to be Sanskrit, but Dr. Royle ^ having consulted Wilson, the famous author of the Sanskrit dictionary, touching the antiquity of this name, he replied that it w^as taken from the Sahda Chanrika, a comparatively modern compilation. The names of ata, ati, are found in Rheede and Rumphius.* This is doubtless the founda- tion of Saint-Hilaire's argument ; but a nearly similar came is given to Anona squamosa in Mexico. This name is ate, ahate di Panucho, found in Hernandez^ with two similar and rather poor figures which may be attributed either to A. squamosa^ a,s Dunal^ thinks, or to A. cherimolia, according to Martins.'^ Oviedo uses the name anon.^ It is very possible that the name ata was introduced into Brazil from Mexico and the nei2:h- bouring countries. It may also, I confess, have come from the Portuguese colonies in the East Indies. Mar- tins says, however, that the species was imported from the West India Islands.^ I do not know whether he had any proof of this, or whether he speaks on the authority of Oviedo's work, w^hich he quotes and which I cannot consult. Oviedo's article, translated by Marcgraf,^^ describes A. sgita77ios(X without speaking of its origin. Forskal visited that country ; it was called Iceschta, that is, coagulated milk. The rarity of its cultivation and the silence of ancient authors fchowa that it was of modern, inti'oduction into Egypt. Ebn Baithar (Sondtheimer's German translation, in 2 vols., 1840), an Arabian physician of the thirteenth century, mentions no Anonacea, nor the name keschta. I do not see that Forskal's description and illustration (Descr., p. 102. ic. tab. 15) differ from A. squamosa. Coquebert's specimen, mentioned in the Systerna, agi-ees with Furskal's plate ; but as it is in flower while the plate shows the fruit, its identity cannot be proved. * Roxburgh, Fl. Lid., edit. 1832, v. ii. p. 657. 2 Piddington, Index, p. 6. ' Royle, III. Hvm., p. 60. * Rheede and RumphiuB, i. p. 139. * Hernandez, pp. 318, 454. * Dunal, Mem. Anon., p. 70. ' Martins, Fl. Bras., fuse. ii. p. 15. * Hence the generic name .4?2ona, which Linnoous changed to Annona (provision), because he did not wish to have any savage name, and did not mind a pun. ^ Martius, Fl, Bra.s., fasc ii. p. 15. ^^ Marcgraf, Brazil, p. 94. 172 ORIGIN OF CULTIVATED PLANTS. " The sum total of the facts is altogether in favour of an American origin. The locality where the species usually appears wild is in the forests of Para. Its culti- vation is ancient in America, since Oviedo is one of the first authors (1535) wlio has written about this country. No doubt its cultivation is of ancient date in Asia like- wise, and this renders the problem curious. It is not proved, however, that it was anterior to the discovery of America, and it seems to me that a tree of which the fruit is so agreeable would have been more widely diffused in the old world if it had always existed there. More- over, it would be difficult to explain its cultivation in America in the beginning of the sixteenth century, on the hypothesis of an origin in the old world." Since I wrote the above, I find the following facts published by difierent authors : — 1. The argument drawn from the fact that there is no Asiatic species of the genus Anona is stronger than ever. A. Asiatica, Linnaeus, was based upon errors (see my note in the Geogr. Bot., p. 8G2). A. ohtusifolia (Tussac, Fl des Antilles, i. p. 191, pi. 28), cultivated formerly in St. Domingo as of Asiatic origin, is also perhaps founded upon a mistake. I suspect that the drawing represents the fiower of one species (A. onuricata) and the fruit of another (-4. squamosa). No Anona has been discovered in Asia, but four or five are now known in Africa instead of only one or two/ and a larger number than formerly in America. 2. The authors of recent Asiatic floras do not hesi- tate to consider the Anonas, particularly A squamosa, which is here and there found apparently wild, as naturalized in the neighbourhood of cultivated ground and of European settlements.^ ^ See Baker, Flora of Ilaxiriiius, p. 3. The identity admitted by Oliver, Fl. Trap. Afr., i. p. 16, of the Anona palustris of America -vviLli that of Senegambia, appears to me very extraordinary, although it is a species which grows in marshes ; that is, having perhaps a very wide area. ^ Hooker, Fl. of Brit. Ind., i. p. 78 ; Miquel, Fl. Indo-Batava, i. part 2, p. 33 ; Kurz, Forest Flora of Brit. Burm.f i. p. 40 ; Stewart and Craudis, Forests of India, p. 6. PLANTS CULTIVATED FOR THEIR FRUITS. 173 3. In the new African floras already quoted, A. squamosa and the others of which I shall speak presently are always mentioned as cultivated species. 4. McNab, the horticulturist, found A. squamosa in tlie dry plains of Jamaica/ which confirms the asser- tions of previous authors. Eggers says ^ that the species is common in the thickets of Santa Cruz and Virgin Islands. I do not find that it has been discovered wild in Cuba. 5. On the American continent it is given as culti- vated.^ However, M. Andre sent me a specimen from a stony district in the Magdalena valley, which appears to belong to this species and to be wild. The fruit is want- ing, which renders the matter doubtful. From the note on the ticket, it is a delicious fruit like that of A. squa- mosa. Warming ^ mentions the species as cultivated at Lagoa Santa in Brazil. It appears, therefore, to be cultivated or naturalized from cultivation in Para, Guiana, and New Granada. In fine, it can hardly be doubted, in my opinion, that its original country is America, and in especial the West India Islands. Sour Sop — A nona muricata, Linnjeus. This fruit-tree,^ introduced into all the colonies in tropical countries is wild in the West Indies ; at least, its existence has been proved in the islands of Cuba, St. Domingo, Jamaica, and several of the smaller islands.^ It is sometimes naturalized on the continent of South America near dwellinos,? Andre brouGfht specimens from the district of Cauca in New Granada, » Grisebach, Fl. of Brit. W. I. Isles, p. 5. * Eggers, Flora of St. Croix and Virgin Isles, p. 23. ^ Triaua and Planchou, Prodr. Fl. Novo-Granatensis, p. 29; Sagot, Journ. Soc. d'Horfic, 1872. * Warming, Syinboloi ad. Fl. Bras., xvi. p. 434-. * Figured in DescourtiJz, Fl. Med. des. Antilles, ii. pi. 87, and in Tussao, Fl. des Antilles, ii. p. 24. ' Richard, Plantes Vasculaires de Cuhii, p. 29; Swarfcz, Ohs., p. 221; P. Brown, Jamaica, p. 255; Macfadyen, M. of Jam., -p. 7; Eggers, jPZ. of St. Croix, p. 23; Grisebach, PZ. Brit. W. /"., p. 4. 7 Martins, Fl. Brasil, fasc. ii. p. 4; Split^erber, PL de /Surinam, in Nederl. Kruidk. Arch., i. p. 226. 174! ORIGIN OF CULTIVATED PLANTS. but he does not say they were wild, and I see that Triana {Proclr. Fl. Graoiat.) only mentions it as culti- vated. Custard Apple in the West Indies, Bullock's Heart in the East Indies — Anona reticulata, Linnseus. This Anona, figured in Descourtilz, Flore Medicale des Antilles, ii. pi. 82, and in the Botanical Marjaziiie, pi. 2912, is wild in Cuba, Jamaica, St. Vincent, Guade- loupe, Santa Cruz, and Barbados,-^ and also in the island of Tobago in the Bay of Panama,^ and in the province of Antioquia in New Granada.^ If it is wild in the last- named localities as well as in the West Indies, its area probably extends into several states of Central America and of New Granada. Although the bullock's heart is not much esteemed as a fruit, the species has been introduced into most tropical colonies. Rheede and Rumphius found it in plantations in Southern Asia. According to Welwitsch, it has naturalized itself from cultivation in Angola, in Western Africa,* and this has also taken place in British India.^ Chirimoya — Anona Cherhnolia, Lamarck. The chirimoya is not so generally cultivated in the colonies as the preceding species, although the fruit is excellent. This is probably the reason that there is no illustration of the fruit better than that of Feuillee {Ohs., iii. pi. 17), while the flower is well represented in pi. 2011 of the Botanical Magazine^ under the name of A. tripetala. In 1855, I wrote as follows, touching the origin of the species:^ "The chirimoya is mentioned by Lamarck and Dunal as growing in Peru; but Feuillee, who was the first to speak of it,"^ says that it is cultivated. Mac- * Richarcl, MacfadyeD, Grisebach, Eggcrs, Swartz, Maycock, Fl. Barhad., p. 233. 2 Seemann, Bot. of the Herald, p. 75. ' Ti-iaua and Plancbon, Prodr. Fl. Novo-Granat-, p. 29. * Oliver, Fl. Trop. Aj'r., i. p. 15. * Sir J. Hooker, Fl. Brit. Ind., i. p. 78. " Tie Candolle, Geogr. Bot. Rais., p. 803. ^ Feuillee, Ob.<^., iii. p. 23, t. 17. PLANTS CULTIVATED FOR THEIR FRUITS. 175 fadyen^ says it abounds in the Port Ko3^al Mountains, Jamaica ; but he adds that it came originally from Peru, and must have been introduced lon^r asjo, whence it appears that the species is cultivated in the higher plantations, rather than wild. Sloane does not mention it. Humboldt and Bonpland saw it cultivated in Venezuela and New Granada ; Martins in Brazil,^ where the seeds had been introduced from Peru. The species is cultivated in the Cape Verde Islands, and on the coast of Guinea,^ but it does not appear to have been introduced into Asia. Its American origin is evident. I might even go further, and assert that it is a native of Peru, rather than of New Granada or Mexico. It will probably be found wild in one of these countries. Meyen has not brouo^ht it from Peru." * My doubts are now lessened, thanks to a kind com- munication from M. Ed. Andre. I may mention first, that I have seen specimens from Mexico gathered by Botteri and Bourgeau, and that authors often speak of finding the species in this region, in the West Indies, in Central America, and New Granada. It is true, they do not say that it is wild. On the contrary, they remark that it is cultivated, or that it has escaped from gardens and become naturalized.** Grisebach asserts that it is wild from Peru to Mexico, but he gives no proof. Andre gathered, in a valley in the south-west of Ecuador, specimens which certainly belong to the species as far as it can be asserted without seeing the fruit. He says nothing as to its wild nature, but the care with which he points out in other cases plants cultivated or perhaps escaped from cultivation, leads me to think that he regards these specimens as wild. Claude Gay says that the species has been cultivated in Chili from time im- memorial.'' However, Molina, who mentions several fruit- * Macfaclyen, Fl. Jam., p. 10. ^ Martius, Fl. Bras., fasc. iii. p. 15. ' Hooker, Fl. Nigr., p. 205. * Nov. Act. Nat. Cur., xix. suppl. 1. * Richard, Plant. Vase, de Cuba ; Grisebach, FL Brit. W. Ind. Is. ; Hemsley, Biologia Centr. Am., p. 118; Kunth, in Huniboldt and Bon- pland, Nova Gen., v. p. 57 ; Triana and Planchon, Prodr, Fl. Novo. Gruvat., p. 28. e G ly, Flora ChiL, i. p. G6, 176 ORIGIN OF CULTIVATED PLANTS. trees in the ancient plantations of tlie country, does not speak of it.^ In conclusion, I consider it most probable that the species is indigenous in Ecuador, and perhaps in the neighbouring part of Peru. Oranges and Lemons — Citrw^, Linnreus. The ditierent varieties of citrons, lemons, oranges, shaddocks, etc., cultivated in gardens have been the subject of remarkable works by several horticulturists, amonof which Gallesio and Risso^ hold the first rank. The difficulty of observing and classifying so many varieties was very great. Fair results have been obtained, but it must be owned that the method was wrong from the beginning, since the plants from which the observations were taken were all cultivated, that is to say, more or less artificial, and perhaps in some cases hybrids. Botanists are now more fortunate. Thanks to the discoveries of travellers in British India, they are able to distinofuish the Avild and therefore the true and natural species. According to Sir Joseph Hooker,^ who was himself a collector in India, the work of Brandis ^ is the best on the Citrus of this region, and he follows it in his flora. I shall do likewise in default of a mono- graph of the genus, remarking also that the multitude of garden varieties which have been described and fissured for centuries, ouojht to be identified as far as possible with the wild species.^ The same species, and perhaps others also, probably grow wild in Cochin-China and in China ; but this has not been proved in the country itself, nor by means of specimens examined by botanists. Perhaps the im- portant works of Pierre, now in course of publication, will ^ ISIolina, French trans. ^ Gallesio, Traite da Cifn's, in 8vo, Paris, 1811 ; Eisso and Poiteau, Histoire Naturelle ties Oramjers, 1818, in folio, 109 plates. ' Hooker, Fl. of Brit. Ind.j i. p. 515. * Brandis, Forest Flora, p. 50. ' For a work of this nature, the first step -wonld be to publish cfood figures of wild species, showing particularly the fruit, which is not seen in herbaria. It would then be seen which forms represented in the plates of Risso, Duhamel, and others, are nearest to the wild types. PLANTS CULTIVATED FOR THEIR FRUITS. 177 give information on this head for Cochin-China. With regard to China, I will quote the following passage from Dr. Bretschneider/ which is interesting from the special knowledge of the writer : — " Oranges, of which there are a great variety in China, are counted by the Chinese anion cr their wild fruits. It cannot be doubted that most of them are indigenous, and have been cultivated from very early times. The proof of this is that each species or variety bears a distinct name, besides being in most cases represented by a particular character, and is mentioned in the Shu-Jcing, Rh-ya, and other ancient works." Men and birds disperse the seeds of Aurantiacese, whence results the extension of its area, and its naturali- zation in all the warm regions of the two worlds. It was observed ^ in America from the first century after the conquest, and now groves of orange trees have sprung up even in the south of the United States. Shaddock — Citrus decumana, Willdenow. I take this species first, because its botanical character is more marked than that of the others. It is a lar2:er tree, and this species alone has down on the young- shoots and the under sides of the leaves. The fruit is spherical, or nearly spherical, larger than an orange, sometimes even as large as a man's head. The juice is slightly acid, the rind remarkably thick. Good illus- trations of the fruit may be seen in Duhamel, Traite cle8 Arhres, edit. 2, vii. pi. 42, and in Tussac, Flore des Antilles, iii. pis. 17, 18. The number of varieties in the Malay Archipelago indicates an ancient cultivation. Its original country is not yet accurately known, because the trees which appear indigenous may be the result of naturaliza- tion, following frequent cultivation. Roxburgh says that the species was brought to Calcutta from Java,^ and Rumphius * believed it to be a native of Southern China. ^ Bretschneider, On the Study and Value of Chinese Botanical Worlds, p. 55. - Acosta, Hist. Nat. des Indes, Fr. trans. 1598, p. 187. * Roxburgh, Flora Tndica, edit. 1832 iii. p. 393. * Rumphius, Hortus Arnheinensis, ii. p. 98. 178 ORIGIN OF CULTIVATED PLANTS. Neither he nor modern botanists saw it wild in the Malay Archipelago.-^ In China the species has a simple name, yu ; but its written character ^ appears too com- plicated for a truly indigenous plant. According to Loureiro, the tree is common in China and Cochin-China, but this does not imply that it is wild.^ It is in the islands to the east of the Malay Archipelago that the clearest indications of a wild existence are found. Forster * formerly said of this species, " ver}'- common in the Friendly Isles." Seemann ^ is yet more positive about the Fiji Isles. " Extremely common," he says, " and covering the banks of the rivers." It would be strange if a tree, so much cultivated in the south of Asia, should have become naturalized to such a degree in certain islands of the Pacific, while it has scarcely been seen elsewhere. It is probably indi- genous to them, and may perhaps yet be discovered wild in some islands nearer to Java. The French name, 'pompelmoiise, is from the Dutch pompehnoes. Shaddock was the name of a captain who first introduced the species into the West Indies.^ Citron, Lemon — Citrus Tnedica, Linnseus. This tree, like the common orange, is glabrous in all its parts. Its fruit, longer than it is wide, is surmounted in most of its varieties by a sort of nipple. The juice is more or less acid. The young shoots and the petals are frequently tinted red. The rind of the fruit is often rough, and very thick in some subvarieties.'' Brandis and Sir Joseph Hooker distinguish four cultivated varieties : — 1. Citrus medica proper {citron in English, cedra- tier in French, cedro in Italian), with large, not * Miqiiel, Flora Indo-Batava, i. pt. 2, p. 526. 2 Bretschneider, Study and Value, etc. ^ Loureiro, Fl. Cochin., ii. p. 572. For another species of the genus, he says that it is cultivated and non-cultivated, p. 509. * Forster, De Plantis Esculentis Oceani Australis, p. 35. * Seemann, Flora Vitiensis, p. 33. " Plukenet, Almagestes;, p. 239; Sloane, Jamaica, i. p. 41. ' Cedrat a gros fruit of Duhamel, Traite des Arbres, edit. 2, vii. p. G*. pi. 22. PLANTS CULTIVATED FOR THEIR FRUITS. 179 spherical fruit, whose highly aromatic rind is covered with lumps, and of which the juice is neither abundant nor very acid. According to Brand is, it was called vijajntra in Sanskrit. 2. Citrus niedica Limonuni (citronnier in French, lemon in English). Fruit of average size, not spherical, and abundant acid juice. 3. Citrus 7)iedica acida {C. acida, Roxburgh). Lime in English. Small flowers, fruit suiall and variable in shape, juice ver}^ acid. According to Brandis, the Sanskrit name was jamhira. 4. Citrus medica Limetta (C. Limetta and C. Liiinia of Risso), with flowers like those of the preceding variety, but with spherical fruit and sweet, non-aromatic juice. In India it is called the siveet lime. The botanist Wight aflirms that this last variety is wild in the Nilgherry Hills. Other forms, which answer more or less exactly to the three other varieties, have been found wild by several Anglo-Indian botanists^ in the warm districts at the foot of the Himalayas, from Garwal to Sikkim, in the south-east at Chittao^onof and in Burmah, and in the south-west in the western Ghauts and the Satpura Mountains. From this it cannot be doubted that the species is indigenous in India, and even under diflerent forms of prehistoric antiquity I doubt whether its area includes China or the Malay Archipelago. Loureiro mentions Citrus medica in Cochin- China only as a cultivated plant, and Bretschneider tells us that the lemon has Chinese names which do not exist in the ancient writings, and for which the written characters are complicated, indications of a foreign species. It may, he says, have been introduced. In Japan the species is only a cultivated one.^ Lastly, several of Rumphius' illustrations show varieties culti- vated in the Sunda Islands, but none of these are con- sidered by the author as really wild and indigenous to the country. To indicate the locality, he sometimes used * Roy le, IZZ. HimaL, -p. 129; Brandis, Forest Flora, p. 52; Hooker, Fl. of Brit. Ind., i. p. 514. * Franchet and Savatior, Emim, Plant, Jap., p. 129. 9 ISO ORIGIN OF CULTIVATED PLANTS. the expression " in Jiortis sylvestrihus!' which might be translated shrubberies. Speaking of his Lemon sussu (vol. ii. pi. 25), which is a Citrus meclica with ellipsoidal acid fruit, he says it has been introduced into Amboyna, but that it is commoner in Java, " usually in forests." This may be the result of an accidental naturalization from cultivation. Miquel, in his modern flora of the Dutch Indies,^ does not hesitate to say that Citrus meclica and C. Limoniim are only cultivated in the archipelago. The cultivation of more or less acid varieties spread into Western Asia at an early date, at least into Mesopo- tamia and Media. This can hardly be doubted, for two varieties had Sanskrit names ; and, moreover, the Greeks knew the fruit through the Medes, whence the name Citrus medica. Theophrastus ^ was the first to speak of it under the name of apple of Media and of Persia, in a phrase often repeated and commented on in the last two centuries.^ It evidently applies to Citrus medica ; but while he explains how the seed is first sown in vases, to be afterwards transplanted, the author does not say whether this was the Greek custom, or whether he was describing the practice of the Medes. Probably the citron was not then cultivated in Greece, for the Romans did not grow it in their gardens at the beginning of the Christian era. Dioscorides,^ born in Cilicia, and who wrote in the first century, speaks of it in almost the same terms as Theophrastus. It is supposed that the species was, after many attempts,^ cultivated in Italy in the third or fourth century. Palladius, in the fifth century, speaks of it as well established. The ignorance of the Romans of the classic period touching foreign plants has caused them to confound, under the name of lignum citreum, the wood of Citrus, with that of Cedrits, of which fine tables were made, and * Miquel, Flora Indo-Batava, i, pt. 2, p. 528, ^ Theophrastus, 1. 4, c. 4. ^ BodEens, in Theophrastus, edit. 1644, pp. 322, 343; Risso, Traite du Citrus, p, 198 ; Targioni, Cenni Storici, p. 196. ^ Dioscoride^, i. p 166. » Targioni, Cenni Storici. PLANTS CULTIVATED FOR THEIE FRUITS. 181 which was a cedar, or a Thuya, of the totally different family of Coniferse. The Hebrews must have known the citron before the Romans, because of their frequent relations with Persia, Media and the adjacent countries. The custom of the modern JeAVS of presenting themselves at the synagogue on the day of the Feast of Tabernacles, with a citron in their hand, o-ave rise to the belief that the word haclar in Leviticus signified lemon or citron ; but Risso has shown, by comparing the ancient texts, that it signifies a fine fruit, or the fruit of a fine tree. He even thinks that the Hebrews did not know the citron or lemon at the beginning of our era, because the Septuagint Version translates haclar by fruit of a fine tree. Nevertheless, as the Greeks had seen the citron in Media and in Persia in the time of Theophrastus, three centuries before Christ, it Avould be strano-e if the Hebrews had not become acquainted with it at the time of the Babylonish Captivity. Besides, the historian Josephus says that in his time the Jews bore Persian apples, malum persicum.j at their feasts, one of the Greek names for the citron. The varieties with very acid fruit, like Limonum and acida, did not perhaps attract attention so early as the citron, however the strongly aromatic odour mentioned by Dioscorides and Theophrastus appears to indicate them. The Arabs extended the cultivation of the lemon in Africa and Europe. According to Gallesio, they transported it, in the tenth century of our era, from the gardens of Oman into Palestine and Egypt. Jacques de Vitry, in the thirteenth century, well described the lemon which he had seen in Palestine. An author named Falcando mentions in 1260 some very acid " liimias " which were cultivated near Palermo, and Tuscany had them also towards the same period.-^ Orange — Citrus Aurantium, Linnaeus (excl. var. y) ; Citrus Attraniium, Risso. Oranges are distinguished from shaddocks (C. decu- miana) by the complete absence of down on the young shoots and leaves, by their smaller fruit, always spherical, 1 Targioni, p. 217. 182 ORIGIN OF CULTIVATED PLANTS. and by a thinner rind. They differ from lemons and citrons in their pure white flowers ; in the fruit, wdiich is never elongated, and without a nipple on the summit ; in the rind, smooth or nearly so, and adhering but lightly to the pulp. Neither Risso, in his excellent monograph of Citrus, nor modern authors, as Brandis and Sir Joseph Hooker, have been able to discover any other character than the taste to distinguish the sweet orange from more or less bitter fruits. This difference appeared to me of such slight importance from the botanical point of view, when I studied the question of origin in 1855, that I was inclined, with Risso, to consider these two sorts of orange as simple varieties. Modern Anglo-Indian authors do the same. They add a third variety, which they call Bergcnnia, for the bergamot orange, of wdiich the flower is smaller, and the fruit spherical or pyriform, and smaller than the common orange, aromatic and slightly acid. This last form has not been found wild, and appears to me to be rather a product of cultivation. It is often asked wdiether the seeds of sweet oranges yield sweet oranges, and of bitter, bitter oranges. It matters little from the point of view of the distinction into species or varieties, for w^e know that both in the animal and veojetable kino-doms all characters are more or less hereditary, that certain varieties are habitually so, to such a degree that they should be called races, and that the distinction into species must consequently be founded upon other considerations, such as the absence of intermediate forms, or the failure of crossed fertilization to produce fertile hybrids. However, the question is not devoid of interest in the present case, and I must answer that experiments have given results wdiich are at times contradictory. Gallesio, an excellent observer, expresses himself as follows : — " I have during a long series of years sown pips of sweet oranges, taken sometimes from the natural tree, sometimes from oranges grafted on bitter orange trees or lemon trees. The result has always been trees bearino* sweet fruit ; and the same has been observed for more than sixty years by all the gardeners of Finale. There PLANTS CULTIVATED FOR THEIR FRUITS. 183 is no instance of a bitter orange tree from seed of sweet oranges, nor of a sweet orange tree from the seed of bitter oranges. ... In 1709, the orange trees of Finale having been killed by frost, the practice of raising sweet orange trees from seed was introduced, and every one of these plants produced the sweet-juiced fruit." ^ Macfadyen,^ on the contrary, in his Flora of Jamaica, says, " lb is a well-established fact, familiar to every one who has been any length of time in this island, that the seed of the sweet orange very frequently grows up into a tree bearing the bitter fruit, numerous well-attested instances of which have come to my own knowledge. I am not aware, however, that the seed of the bitter orancce has ever grown up into the sweet-fruited variety. . . . We may therefore conclude," the author judiciously goes on to sav, " that the bitter orano-e was the orio-inal stock." He asserts that in calcareous soil the sweet orange may be raised from seed, but that in other soils it produces fruits more or less sour or bitter. Duchassaing says that in Guadeloupe the seeds of sweet oranges often yield bitter fruit,^ while, according to Dr. Ernst, at Caracas they sometimes yield sour but not bitter fruit.* Brandis relates that at Khasia, in India, as far as he can verify the fact, the extensive plantations of sweet oranges are from seed. These differences show the variable deo-ree of heredity, and confirm the opinion that these two kinds of orange should be considered as two varieties, not two species. I am, however, obliged to take them in succession, to explain their origin and the extent of their cultivation at different epochs. Bitter Orange — Arancio forte in Italian, higaradier in French, pomeranze in German. Citrus vulgaris, Risso ; G. aiirantium (var. bigaradia), Brandis and Hooker. It was unknown to the Greeks and Romans, as well as the sweet orange. As they had had communication * Gallesio, Traitc dit Citrus, pp. 32, 67, 355, 357. * Macfadyen, Flora of Jamaica, p. 129. * Quoted in Grisebach's Veget. Karaiben, p. 34. * Ei'nst, in Seemann, Journ. of Bat., 18G7, p. 272. 184; ORIGIN OF CULTIVATED PLANTS. with India and Ceylon, Gallesio supposed that these trees were not cultivated in their time in the west of India. He had studied from this point of view, ancient travellers and geographers, such as Diodorus Siculus, Nearchus, Arianus, and he finds no mention of the orange in them. However, there was a Sanskrit name for the orange — nagarunga, nagrunga} It is from this that the word orange came, for the Hindus turned it into naritn- gee (pron. naroudji) , Siccording to Royle,''/ier«7ir/a accord- ing to Piddington ; the Arabs into narunj, according to Gallesio, the Italians into naranzi, arangi, and in the medieeval Latin it was arancium, arangiuvi, afterwards aurantium.^ But did the Sanskrit name apply to the hitter or to the sweet orange ? The philologist Adolphe Pictet formerly gave me some curious information on this head. He had sought in Sanskrit works the de- scriptive names given to the orange or to the tree, and had found seventeen, which all allude to the colour, the odour, its acid nature (danta catha, harmful to the teeth), the place of growth, etc., never to a sweet or ao^reeable taste. This multitude of names similar to epithets show that the fruit had long been known, but that its taste was very different to that of the sweet orange. Besides, the Arabs, who carried the orange tree with them towards the West, were first acquainted with the bitter orange, and gave it the name narunj^ and their physicians from the tenth century prescribed the bitter juice of this fruit.* The exhaustive researches of Gallesio show that after the fall of the Empire the species advanced from the coast of the Persian Gulf, and by the end of the ninth century had reached Arabia, through Oman, Bassora, Irak, and Syria, according to the Arabian author Massoudi. The Crusaders saw the bitter oranixe tree in Palestine. It was cultivated in Sicily from the year 1002, probably a result of the incursions of the * Roxburgh, Fl. hidica, edit. 1832, voL ii. p. 392; Piddington, Index. ' Gallesio, p. 122. ^ In the modern languages of India the Sanskrit name has been applied to the sweet orange, so says Brandis, by one of those transposi- tions which are so common in popular language. * Gallesio, pp. 122, 217, 218. PLANTS CULTIVATED FOR THEIR FRUITS. 185 Arabs. It was they who introduced it into Spain, and most likely also into the east of Africa. The Portuguese found it on that coast when they doubled the Cape in 1498.^ There is no ground for supposing that either the bitter or the sweet orange existed in Africa before the Middle Ages, for the myth of the garden of Hesperides may refer to any species of the order Aitrantiacece, and its site is altogether arbitrary, since the imagination of the ancients was wonderfully fertile. The early Anglo-Indian botanists, such as Roxburgh, Royle, Griffith, Wight, had not come across the bitter orange wild; but there is every probability that the eastern region of India was its original country. Wallich mentions Silhet,^ but without asserting that the species was wild in this locality. Later, Sir Joseph Hooker ^ saw the bitter orange certainly wild in several districts to the south of the Himalayas, from Garwal and Sikkim as far as Khasia. The fruit was spherical or slightly flattened, two inches in diameter, bright in colour, and uneatable, of mawkish and bitter taste (" if I remember right," says the author). Citrus fusca, Loureiro,* similar, he says, to pi. 23 of Rumphius, and wild in Cochin-China and China, may very likely be the bitter orange whose area extends to the east. Sweet Orange — Italian, Arancio clolce ; German, A2')felsine. Citrus Aurantium sinense, Gallesio. Royle -^ says that sweet oranges giow wild at Silhet and in the Nilgherry Hills, but his assertion is not accompanied with sufficient detail to give it importance. According to the same author. Turner's expedition gathered " delicious " wild oranges at Buxedwar, a locality to the north-east of Rungpoor, in the province of Bengal. On the other hand, Brandis and Sir Joseph Hooker do not mention the sweet orange as wild in ^ Gallesio, p. 210. Goeze, Beifrag ziir Kenniniss dcr Orongenrjeii-acln^e, 1874, p. 13, quotes early Portuguese ti'avellers on this head. 2 Wallich, Catalogue, No. 6384. * Hooker, Fl. of Brit. Ind., i. p. 515. * Loureiro, Fl. Cochin., p. 571. ' Royle, Illustr. of Himal., p. 129. He quotes Turner, Journey to Thibet, pp. 20, 387. 186 ORIGIN OF CULTIVATED PLANTS. British India ; they only give it as cultivated. Kurz does not mention it in his forest flora of British Burmah. Further east, in Cochin-China, Loureiro ^ describes a G. Atirantiuni, with bitter-sweet (acido-duJcis) pulp, which appears to be the sweet orange, and which is found both wild and cultivated in China and Cochin-China. Chinese authors consider orange trees in general as natives of their country, but precise information about each species and variety is wanting on this head. From the collected facts, it seems that the sweet oranoe is a native of Southern China and of Cochin- China, with a doubtful and accidental extension of area by seed into India. By seeking in what country it was first cultivated, and how it was propagated, some light may be thrown upon the origin, and upon the distinction between the bitter and sweet orange. So large a fruit, and one so agreeable to the palate as the sweet orange, can hardly have existed in any district, without some attempts having been made to cultivate it. It is easily raised from seed, and nearly always produces the wished-for quality. Neither can ancient travellers and historians have necflected to notice the introduction of so remark- able a fruit tree. On this historical point Gallesio's study of ancient authors has produced extremely in- terestino- results. He first proves that the orange trees brought from India by the Arabs into Palestine, Egypt, the south of Europe, and the east coast of Africa, were not the sweet- fruited tree. Up to the fifteenth century, Arab books and chronicles only mention bitter, or sour oranges. However, when the Portuguese arrived in the islands of Southern Asia, they found the sweet orange, and ap- parentty it had not previously been unknown to them. The Florentine who accompanied Vasco de Gama, and who published an account of the voyage, says, " Sonvi melarancie asscd, ona tuite doici" (there are plenty of oranges, but all sweet.) Neither this writer nor subsequent travellers expressed surprise at the joleasant taste of the ' Loureiro, Fl. Cochin., p. 5G9. PLANTS CULTIVATED FOR THEIR FRUITS. 187 fruit. Hence Gallesio infers that the Portnixuese were not the first to brings the sweet oran^-e from India, which they reached in 3498, nor from China, which they reached in 1518. Besides, a number of writers in the beginning of the sixteenth century speak of the sweet orange as a fruit already cultivated in Spain and Italy. There are several testimonies for the years 1523, and 1525. Gallesio goes no further than the idea that the sweet orange was introduced into Europe towards the beginning of the fifteenth century ; ^ but Targioni quotes from Valeriani a statute of Fermo, of the fourteenth century, referring to citrons, sweet oranges, etc. ; ^ and the information recently collected from early authors by Goeze,^ about the introduction into Spain and Portugal, agrees with this date. It therefore appears to me prob- able that the oranges imported later from China by the Portuguese were only of better quality than those already known in Europe, and that the common expres- sions, Portugal and Lisbon oranges, are due to this cir- cumstance. If the sweet orange had been cultivated at a very early date in India, it would have had a special name in Sanskrit; the Greeks would have known it after Alexander's expedition, and the Hebrews would have early received it through Mesopotamia. This fruit wovild certainly have been valued, cultivated, and propagated in the Roman empire, in preference to the lemon, citron, and bitter orange. Its existence in India must, there- fore, be less ancient. In the Malay ArcTiipelago the sweet orange was believed to come from China.^ It was but little diffused in the Pacific Isles at the time of Cook's voyages.^ We come back thus by all sorts of ways to the idea that the sweet variety of the orange came from China » Gallesio, p. 321. 2 The date of tins statuto is given by Tarf^ioni, on p. 205 of the Cenni Storici, as 1379, and on p. 213 as 1309. The errata do not notice this discrepancy. ' Goeze, Ein Beitrag zur Kenntniss der Orangeyigeivdchse. Hamburg, 1874, p. 26. * Rumphius, Amhoin., ii. c. 42. Forster, Plantis Esculentisy p. 3~>. 188 ORIGIN OF CULTIVATED PLANTS. and Cochin- China, and that it spread into India perhaps towards the beginning of the Christian era. It may have become naturalized from cultivation in many parts of India and in all tropical countries, but we have seen that the seed does not always yield trees bearing sweet fruit. This defect in heredity in certain cases is in support of the theory that the sweet orange was derived from the bitter, at some remote epoch, in China or Cochin-China, and has since been carefully propagated on account of its horticultural value. Mandarin — Citrus nohilis, Loureiro. This species, characterized by its smaller fruit, uneven on the surface, spherical, but flattened at the top, and of a peculiar flavour, is now prized in Europe as it has been from the earliest times in China and Cochin-China. The Chinese call it kan} Rumpliius had seen it culti- vated in all the Sunda Islands,^ and says that it was introduced thither from China, but it had not spread into India. Roxburgh and Sir Joseph Hooker do not mention it, but Clarke informs me that its culture has been greatly extended in the district of Khasia. It was new to European gardens at the beginning of the present century, when Andrews published a good illustration of it in the Botanist's Repository (pi. 008). According to Loureiro,^ this tree, of average size, grows in Cochin-China, and also, he adds, in China, although he had not seen it in Canton. This is not very precise information as to its wild character, but no other origin can be supposed. According to Kurz,^ the species is only cultivated in British Burmah. If this is confirmed, its area would be restricted to Cochin-China and a few provinces in China. Mangosteen — Garcinia onangostana, Linnaeus. There is a good illustration in the Botanical Magazine, pi 4847, of this tree, belonging to the order Guttifene, of which the fruit is considered one of the best in existence. * Bretsclmeider, On the Study and Vahie, etc., p. 11. * Kumpliius, Anihoin., ii. pis. 3-i, 35, where, however, the form of the fruit is not that of our maadarin. * Loureiro, Fl. Cochin., p. 570. * Kurz, Forest Fl. of Brit. Bur. PLANTS CULTIVATED FOR THEIR FRUITS. 189 It demands a very hot climate, for Roxburgh could not make it grow north of twenty-three and a half degrees of latitude in India/ and, transported to Jamaica, it bears but poor fruit.^ It is cultivated in the Sunda Islands, in the Malay Peninsula, and in Ceylon. The species is certainly wild in the forests of the Sunda Islands ^ and of the Malay Peninsula.* Among cultivated plants it is one of the most local, both in its origin, habitation, and in cultivation. It belongs, it is true, to one of those families in which the mean area of the species is most restricted. Mamey, or Mammee Apple — Mammea Americana, Jacquin. This tree, of the order Guttifer?e, requires, like the mangosteen, great heat. Although much cultivated in the West Indies and in the hottest parts of Venezuela,^ its culture has seldom been attempted, or has met with but little success, in Asia and Africa, if we are to judge by the silence of most authors. It is certainly indigenous in the forests of most of the West Indies.^ Jacquin mentions it also for the neigh- bouring continent, but I do not find this confirmed by modern authors. The best illustration is that in Tussac's Flore des Antilles, iii. pi. 7, and this author gives a number of details respecting the use of the fruit. Ochro, or Gombo — Hibiscus esculentns, Linnaeus. The young fruits of this annual, of the order of Malvaceae, form one of the most delicate of tropical vegetables. Tussac's Flore des Antilles contains a fine plate of the species, and gives all the details a gourmet could desire on the manner of preparing the caloidou, so much esteemed by the Creoles of the French colonies. ^ Royle, HI. Himal., p. 133, and Roxburgh, Fl. Ind,, ii. p. 618. ' Macfadyen, Flora of Jamaica, p. 134. •* Rampliius, Amboin., i. p. 133 ; Miquel, PlantcB Junghun., i. p, 290 ; Flora Indo-Batava, i. pt. 2, p. 506. * Hooker, Flora of Brit. Ind., i. p. 260. ' Ernst in Seemann, Journal of Botany, 18G7, p. 273; Triaua and P:anchon, Prodr. Fl. Novo-G-ranat., p. 285. * Sloane, Jamaica, i. p. 123; Jacquin, Am.er., p. 263; Grisebach, Fl of Brit. W. Ind. Isles, p. 118. 190 ORIGIN OF CULTIVATED PLANTS. When I formerly ^ tried to discover vt^hence this plant, cultivated in the old and new worlds, came originally, the absence of a Sanskrit name, and the fact that the first writers on the Indian flora had not seen it wild, led me to put aside the hypothesis of an Asiatic origin. How- ever, as the modern flora of British India ^ mentions it as " probably of native origin," I was constrained to make further researches. Although Southern Asia has been thoroughly explored during the last thirty years, no locality is mentioned where the Gombo is wild or half wild. There is no indication, even, of an ancient cultivation in Asia. The doubt, therefore, lies between Africa and America, The plant has been seen wild in the West Indies by a good observer,^ but I can discover no similar assertion on the part of any other botanist, either with respect to the islands or to the American continent. The earliest writer on Jamaica, Sloane, had only seen the species in a state of cultivation. Marcgraf ^ had observed it in Brazilian plan- tations, and as he mentions a name from the Congo and Angola country, quilloho, which the Portuguese corrupted into quingombo, the African origin is hereby indicated. Schweinfurth and Ascherson ^ saw the plant wild in the Nile Valley in Nubia, Kordofan, Senaar, Abyssinia, and in the Baar-el-Abiad, where, indeed, it is cultivated. Other travellers are mentioned as having gathered speci- mens in Africa, but it is not specified whether these plants were cultivated or wild at a distance from habita- tions. We should still be in doubt if Fluckiger and Hanbury^ had not made a bibliographical discovery which settles the question. The Arabs call the fruit harayah, or bchniat, and Abul-Abas-Elnabati, who visited Egypt long before the discovery of America, in 1216, has * A. de Candolle, Geogr. Bot. Rats., p. 7G8. - Flora of Brit. Ind., i. p. 343. ' Jacquin, Ohservationes, iii. p. 11. * Marcgraf, Hist. Plant., p. 32, with illustrations. ' Scliweinfurtli and Ascherson, Aufzdldung, p. 265, under the name ahelmoschus. ^ FlUckiger and Hanburr, Pharmacograpliia, p. 86. The descrip- tion is in Ebn Baithar, Sondtheimer's ti'ans., i. p. 118. PLANTS CULTIVATED FOR THEIR FRUITS. 191 distinctly described the gomho then cultivated by the Egyptians. In spite of its undoubtedly African origin, it does not appear that the species was cultivated in Lower Egypt before the Arab rule. No proof has been found in ancient monuments, although Rosellini thought he recognized the plant in a drawing, which differs widely from it according to Unger.^ The existence of one name in modern Indian languages, according to Piddington, con- firms the idea of its propagation towards the East after the beiiinninor of the Christian era. Vine — Vitis vinifera, Linnseus. The vine grows wild in the temperate regions of Western Asia, Southern Europe, Algeria, and Marocco.^ It is especially in the Pontus, in Armenia, to the south of the Caucasus and of the Caspian Sea, that it grows with the luxuriant wildness of a tropical creeper, clinging to tall trees and producing abundant fruit without pruning or cultivation. Its vigorous growth is mentioned in ancient Bactriana, Cabul, Kashmir, and even -in Badak- khan to the north of the Hindu Koosh.^ Of course, it is a question whether the plants found there, as elsewhere, are not sprung from seeds carried from vineyards by birds. I notice, however, that the most trustworthy botanists, those who have most thoroughly explored the Transcaucasian provinces of Russia, do not hesitate to say that the plant is wild and indigenous in this region. It is as we advance towards India and Arabia, Europe and the north of Africa, that we frequently find in floras the expression that the vine is " subspontaneous," per- haps wild, or become wild (verwildert is the expressive German term). The dissemination by birds must have begun very early, as soon as the fruit existed, before cultivation, before the migration of the most ancient Asiatic peoples, * Unger, Die Pflanzen des Alien Mgyptens, p. 50. - Grisebach, Veget. cUi Olohe, French trans, by TchibatchffF, i. pp. 162, 163, 442; Munby, Catal. Alger; Ball, Fl. Maroc. Spicel, p. 392. » Adolphe Pictet, Origines Indo-Europ. edit. 2, vol. 1, p. 295, qnotes several tj-avellers for these regions, amdng others Wood's Journei/ to the Sources of the Oxus. 192 ORIGIN OF CULTIVATED PLANTS. perhaps before the existence of man in Europe or even in Asia. Nevertheless, the frequency of cultivation, and the multitude of forms of the cultivated grape, may have extended naturalization and introduced among wild vines varieties which originated in cultivation. In fact, natural agents, such as birds, winds, and currents, have always widened the area of species, independently of man, as far as the limits imposed in each age by geographical and physical conditions, together with the hostile action of other plants and animals, allow. An absolutely primitive habitation is more or less mythical, but habitations successively extended or i*estricted are in accordance with the nature of things. They constitute areas more or less ancient and real, provided that the species has maintained itself wild w^ithout the constant addition ot fresh seed. Concerning the vine, we have proofs of its great antiquity in Europe as in Asia. Seeds ot* the grape have been found in the lake-dwelUngs of Castione, near Parma, which date from the age of bronze,^ in a prehistoric settle- ment of Lake Varese,^ and in the lake-dwellino^s of Wangen, Switzerland, but in the latter instance at an un- certain depth.^ And, what is more, vine-leaves have been found in the tufa round Montpellier, where they were probably deposited before the historical epoch, and in the tufa of Meyrargue in Provence, which is certainly prehis- toric,^ though later than the tertiary epoch of geologists.^ A Russian botanist, Kolenati,*^ has made some very interesting observations on the different varieties of the vine, both wild and cultivated, in the country which may be called the central, and perhaps the most ancient home of the species, the south of the Caucasus. I consider his opinion the more important that the author has based * These are fig^nred in Heer's Pflanzen der Pfahlhauien, p. 24, fig. 11, * Eagazzoni, Rivista Arch, della Prov. di Conio, 1880, fasc. 17, p. 30. ^ Heer, ibid. * Planchon, Etude sur les Tufs de Montpellier, 1864, p. 63. * De Saporta, La Flore des Tufs Quaternaires de Provence, 1867, pp. 15, 27. * Kolenati, Bulletin de la SociCte Imp4riale des Naturalistes de Moscou, 1846, p. 279. PLANTS CULTIVATED FOR THEIR FRUITS. 193 liis classification of varieties with reference to the downy character and veining of the leaves, points absolutely indifferent to cultivatoi's, and which consequently must far better represent the natural conditions of the plant. He says that the wild vines, of which he had seen an immense quantity between the Black and Caspian Seas, may be grouped into two subspecies which he describes, and declares are recognizable at a distance, and which are the point of departure of cultivated vines, at least in Armenia and the neighbourhood. He recognized them near Mount Ararat, at an altitude where the vine is not cultivated, where, indeed, it could not be cultivated. Other characters — for instance, the shape and colour of the grapes — vary in each of the subspecies. We cannot enter here into the purely botanical details of Kolenati's paper, any more than into those of Kegels more recent work on the genus Vitis ; ^ but it is well to note that a species cultivated from a very remote epoch, and which has perhaps two thousand described varieties, presents in the district where it is most ancient, and probably presented before all cultivation, at least two principal forms, with others of minor importance. If the wild vines of Persia and Kashmir, of Lebanon and Greece, were observed with the same care, perhaps other sub- species of prehistoric antiquity might be found. The idea of collecting the juice of the grape and of alloAving it to ferment may have occurred to different peoples, principally in Western Asia, where the vine abounds and thrives. Adolphe Pictet,^ who has, in connnon with numerous authors, but in a more scientific manner, con- sidered the historical, philological, and even mythological questions relating to the vine among ancient peoples, ^ Re<]^el, Acta Horti Imp. Petrop., 1873. In this short review of the gonus, M. Kegel gives it as his opinion that Vitis vinij'era is a hybrid between two wild species, V. vulpina and V. lahrusca, modified by culti- vation ; but he gives no proof, and his characters of the two wild species are altogether unsatisfactory. It is much to be desired that the wild and cultivated vines of Europe and Asia should be compared with regard to their seeds, which furnish excellent distinctions, according to Englemann's observations on the American vines. 2 Ad. Fictet, Origines Indo-Eur., 2nd edit., vol. i. pp. 298-321. 194? OPJGIN OF CULTIVATED PLANTS. admits that both Semitic and Aiyan nations knew the use of wine, so that they may have introduced it into all the countries into which they migrated, into India and Egypt and Europe. This they were the better able to do, since they found the vine wild in several of these regions. The records of the cultivation of the grape and of the making of wine in Egypt go back five or six thousand years.^ In the West the propagation of its culture by the Phenicians, Greeks, and Romans is pretty well known, but to the east of Asia it took place at a late period. The Cliinese who now cultivate the vine in their northern provinces did not possess it earlier than the year 122 b.c.^ It is known that several wild vines exist in the north of China, but I cannot ao;ree with M. Re^el in consider- ^ CD it Flora of India, i. pi. 149 (representing the wild fruit, which is smaller thau that of the cultivated, plant) j Bruudis. ' Rheede, iv. pi. 141. * Piddington, Index. * Rumphius, Amhoyna, ii. pi. 36. ® Zizyphus ahyssinicus, Hochst, seems to be a different species. ^ Tussae, Flore des Antilles, iii. p. 55 (where there is an excellent figure, pi. 13). He says that it is an East Indian species, thus aggra- vating Linnaeus' mistake, who believed it to be Asiatic and American. PLANTS CULTIVATED FOR THEIR FRUITS. 199 I said on the subject in 1855,^ I find them occasionally reproduced. The French name Pommier cVacnjoii (mahogany apple tree) is as absurd as it is possible to be. It is a tree belonging to the order of Terehintacecc or Anacar- diacece, very different from the Rosaceee and the Meliace?e, to which the apple and the mahogany belong. The edible part is more like a pear than an apple, and botani- cally speaking is not a fruit, but the receptacle or sup- port of the fruit, which resembles a large bean. The two names, French and English, are both deriv^ed from a name given to it by the natives of Brazil, acaju, acajaiha, quoted by early travellers.^ The species is certainly wild in the forests of tropical America, and indeed occupies a wide area in that region ; it is found, for example, in Brazil, Guiana, the Isthmus of Panama, and the West Indies.^ Dr. Ernst * believes it is only indigenous in the basin of the Amazon River, although he had seen it also in Cuba, Panama, Ecuador, and New Granada. His opinion is founded upon the absence of all mention of the plant in Spanish authors of the time of the Conquest — a negative proof, which establishes a mere probability. Rheede and Rumphius had also indicated this plant in the south of Asia. The former says it is common on the Malabar coast.^ The existence of the same tropical arborescent species in Asia and America was so little probable, that it was at first suspected that there was a difierence of species, or at least of variety ; but this was not confirmed. Different historical and philological proofs have convinced me that its origin is not Asiatic.^^ Moreover, Rumi)hius, who is always accurate, s[)oke of an ancient introduction by the Portuguese into the Malay Archipelago from America. The Malay name he gives, » Gc'ogr. Bnf. Rais., p. 873, 2 Piso and Marcgraf, Hist. rer. Katur. Bradl^ 1648, p. 57. ' Vide Piso and Marcgraf ; Aublet, Guyane, p. 392 ; Seemann, Bnf. of the Herald, p. 106 ; Jacquin, Amer., p. 124 ; Macfadyen, PI. Jamaic, p. 119; Greisbach, Fl. of Brit. W. Ind., p. 176. * Ernst in Seemann, Journ. of Bot., 1867, p. 273. ' Rheede, Malabar, iii. pi. 54. • Kumphius, Herh. Amhoin., i. pp. 177; 178. 200 ORIGIN OF CULTIVATED PLANTS. cadjw, is American ; that used at Amboyna means Portugal fruit, that of Macassar was taken from the resemblance of the fruit to that of the jainhoscc. Kumphius says that the species was not widely diffused in the islands. Garcia ah Orto did not find it at Goa in 1550, but Acosta after- wards saw it at Couchin, and the Portuguese propagated it in India and the Malay Archipelago. According to Blume and Miquel, the species is only cultivated in Java. Rheede, it is true, says it is abundant {provenit ubiqiie) on the coast of Malabar, but he only quotes one name which seems to be Indian, kajpa onava ; all the others are derived from the American name. Piddinojton gives no Sanskrit name. Lastly, Anglo-Indian colonists, after some hesitation as to its origin, now admit the importation of the species from America at an early period. They add that it has become naturalized in the forests of British India. -^ It is yet more doubtful that the tree is indigenous in Africa, indeed it is easy to disprove the assertion. Loureiro ^ had seen the species on the east coast of this continent, but he supposed it to have been of American orio^in. Thonninsj had not seen it in Guinea, nor Brown in Congo.^ It is true that specimens from the last-named country and from the islands in the Gulf of Guinea were sent to the herbarium at Kew, but Oliver says it is cul- tivated there.^ A tree which occupies such a large area in America, and which has become naturalized in several districts of India within the last two centuries, would exist over a great extent of tropical Africa if it were indi- genous in that quarter of the globe. Mango — Mangifera indica, Linnreus. Belonging to the same order as the CasJieiu, this tree nevertheless produces a true fruit, something the colour of the apricot.^ It is impossible to doubt that it is a native of the south of Asia or of the Malay Archij)elago, when we see 1 Beddone, Flora Sylvaiica, t. 163 ; Hooker, Fl. Brit. Ind., ii. p. 20. ^ Loureiro, Fl. Cochin., p. 30i. ^ Brown, Congo, pp. 12, 49. * Oliver, Fl. of Trap. Afr., i. p. 4i3. ® See plate 4510 of the Botanical Magazine. PLANTS CULTIVATED FOR THEIR FRUITS. 201 the multitude of varieties cultivated in these countries, the number of ancient common names, in particular a Sanskrit name,^ its abundance in tlie gardens of Bengal, of the Dekkan Peninsula, and of Ceylon, even in Rheede's time. Its cultivation was less diti'used in the direction of China, for Loureiro only mentions its existence in Cochin-China. According to Ptumphius,^ it had been introduced into certain islands of the Asiatic Archipelago within the memory of living men. Forster does not mention it in his work on the fruits of the Pacific Islands at the time of Cook's expedition. TJie name common in the Philippine Isles, nianga,^ shows a foreign origin, for it is the Malay and Spanish name. The common name in Cevlon is amihe, akin to the Sanskrit antra, whence the Persian and Arab amh,'^ the modern Indian names, and perhaps the Malay, mangka, manga, manjyelaan, indicated by Ilum[)hius. There are, however, other names used in the Suuda Islands, in the Moluccas, and in Cochin-China. T]i(i variety of these names argues an ancient introduction into the East Indian Archipelago, in spite of the opinion of Rumphius. The Mangifeva which this author had seen wild in Java, and Mamgifera sylvatica which Roxburgh had discovered at Silhet, are other species ; but the true mango is indicated by modern authors as wild in the forests of Ceylon, the regions at the base of the Himalayas, especially towards the east, in Arracan, Pegu, and tlie Andaman Isles.^ Miquel does not mention it as wild in any of the islands of the Malay Archipelago. In spite of its growing in Ceylon, and the indications, less positive certainly, of Sir Joseph Hooker in the Flora of British India, the species is probably rare or only naturalized in the Indian Peninsula. The size of the stone is too great to allow of its being transported by * Eoxburgh, Flora Indira, edit. 2, vol. ii. p. 435 j Piddington, Index. ' Rumphius, Herh. Amhoiii., i. p. 95. ' Blanco, Fl. Filip., p. 181. * Eum])hius ; Forskal, p. cvii. * Thwaites, Enum, Plant. Cei/Z., p. 75; Brandis, Forest Flora, p. 12n Hooker, Fl.Brit. Ind., ii. p. 13 ; Kurz, Forest Flora Brit. Burmah, i. p. 304. 202 ORIGIN OF CULTIVATED PLANTS. birds, but the frequency of its cultivation causes a dispersion by man's agency. If the mango is only naturalized in the west of British India, this must have occurred at a remote epoch, as the existence of a San- skrit name shows. On the other hand, the peoples of Western Asia must have known it late, since they did not transport the species into Egypt or elsewhere towards the west. It is cultivated at the present day in tropical Africa, and even in Mauritius and the Seychelles, where it has become to some extent naturalized in the woods.^ In the new v/orld it was first introduced into Brazil, for the seeds were brousrht thence to Barbados in the middle of the last century.^ A French vessel was carrying some young trees from Bourbon to Saint Domingo in 1782, when it was taken by the English, who took them to Jamaica, where they succeeded won- derfully. AVhen the coffee plantations w^ere abandoned, at the time of the emancipation of the slaves, the mango, whose stones the negroes scattered everywhere, formed forests in ever}'- part of the islands, and these are now valued both for their shade and as a form of food.^ It was not cultivated in Cayenne in the time of Aublet, at the end of the eighteenth centur}^, but now there are mangoes of the finest kind in this colony. They are grafted, and it is observed that their stones produce better fruit than that of the orioinal stock.* Tahiti Apple — Spondias dulcis, Forster. This tree belongs to the family of the Anacardiacece, and is indigenous in the Society, Friendly, and Fiji Islands.^ The natives consumed quantities of the fruit at the time of Cook's voyage. It is like a large plum, of * Oliver, Flora of Trop.Afr., i. p. 442 ; Baker, Fl. of Maur. and Seych.^ p. 63. ^ IInj?he-5, Barhados, p. 177. ' Macfadyen, Fl. o/ Ja»i., p. 221 ; Sir J. Iloolcer, Speech at the Royal Institute. •* Sagofc, Jour, de la Soc. Centr. d*Aijric. de France, 1872. * Forster, De Plantis Escidentis Insularum Oceani Australts, p. 33 ; 8eoma,nn, Flora VUiensis, p. 51 j Nadaud, Enum. des Plantes de Ta'iti. V- 75. PLANTS CULTIVATED FOE THEIR FRUITS. 203 the colour of an apple, and contains a stone covered witli long hooked bristles.-^ The flavour, according to travel- lers, is excellent. It is not among the fruits most widely diff'ased in tropical colonies. It is, however, cultivated in Mauritius and Bourbon, under the primitive Polynesian name evi or hevi,^ and in the West Indies. It was in- troduced into Jamaica in 1782, and thence into Saint Domingo. Its absence in many of the hot countries of Asia and Africa is probably owing to the fact that the species was discovered, only a century ago, in small islands which have no communications with other countries. Strawberry — Fragaria vesca, Linnaeus. Our common strawberry is one of the most widely diffused plants, partly owing to the small size of its seeds, which birds, attracted by the fleshy part on which they are found, carry to great distances. It grows wild in Europe, from Lapland and the Shetland Isles ^ to the mountain ranges in the south ; in Madeira, Spain, Sicily, and in Greece.* It is also found in Asia, from Armenia and the north of Syria ^ to Dahuria. The strawberries of the Himalayas and of Japan,^ which several authors have attributed to this species, do not perhaps belong to it,'^ and this makes me doubt the assertion of a missionary ^ that it is found in China. It is wild in Iceland,^ in the north-east of the United States,^^ round Fort Cumberland, and on the north-west coast,-^^ perhaps even in the Sierra-Nevada of * There is a good coloured illustration in Tussac's Fl. des Antilles, iii. pi. 28. ' Boyer, ITortus Mauritianus, p. 81. ' H. C.Watson, Compendium Cyhele Brit., i. p. IGO ; Fries, Suvima Yeg. Scand., p. 44. * Lowe, Man. Fl. of Madeira, p. 246; Willkomm and Lange, Prodr. Fl. Hisp., iii. p. 224; Moris, Fl. Sardoa, ii. p. 17. * Boissier, Fl. Orient. * Ledebour, Fl. Ross., ii. p. 64. ' Gay; Hooker, Fl. Brit. Ind., ii. p. 344; Francliet and Savatier, Emini. PI. Japon., i. p. 129i ® Ferny, Propag. de la Foi, quoted in Decaisne's Jardin Fruitier du Mus., p. 27. Gay does not give China. * Babington, Journ. of Linncean Society, ii. p. 303 ; J. Gay. ^^ Asa Gray, Botany of the Northern States, edit. 1868, p. IGG. *^ Sir W, Ilooker, Fl, Bor, -Amer.y i. p. 184. 10 204! ORIGIN OF CULTIVATED PLANTS. California.^ Thus its area extends round the north pole, except in Eastern Siberia and the basin of the river Amur, since the species is not mentioned by IMaximowicz in his Prmiitice Florce Amurensis. In America its area is extended along the highlands of Mexico ; for Fragaria Qnexicana, cultivated in the Jardin cles Plantes, and examined by Gay, is F\ vesca. It also grows round Quito, according to the same botanist, who is an authority on this question.^ The Greeks and Romans did not cultivate the straw- berry. Its cultivation was probably introduced in the fifteenth or sixteenth century. Champier, in the six- teenth century, speaks of it as a novelty in the north of France,^ but it already existed in the south, and in England.^ Transported into gardens in the colonies, the straw- berry has become naturalized in a few cool localities far from dwellings. This is the case in Jamaica,^ in Mauritius,^ and in Bourbon, where some plants had been placed by Commerson on the table-land known as the Kaffirs' Plain. Bory Saint- Vincent relates that in 1801 he found districts quite red with strawberries, and that it was impossible to cross them without staining the feet red with the juice, mixed with volcanic dust.*^ It is probable that similar cases of naturalization may be seen in Tasmania and New Zealand. The genus Fragaria has been studied with more care than many others, by Duchesne (fils), the Comte de Lambertye, Jacques Gay, and especially by Madame Eliza Yilmorin, whose faculty of observation was worthy of the name she bore. A summary of their works, with excellent coloured plates, is published in the Jardin * A. Graj, Bot. Calif., i. p. 176. ' J. Gay, in Decaisne, Jardin Fruitier du Museum^ Fraisier, p. 30. ' Le Grand d'Aussy, Hist, de la Vie Privee des Fran^ais, i. pp. 233 and 3. * Olivier de Serres, Theatre d'Agric, p. 511 ; Gerard, from Phillips, Pomarium Britannicum, p. 334. * Purdie, in Hooker's LondoTi Journal of Botany, 18-14, p. 515. * Bojer, Hortus Mauritianus, p. 121. ' Bory Saint-Vincent, Comptes Rendus de VAcad. des. Sc. Nat., 1836, sem. ii. p. 109. PLANTS CULTIVATED FOR THEIR FRUITS. 205 Fruitier clw Museum by Decaisne. These authors have overcome srreat difficulties in cliBtino^uishino^ the varieties and hybrids which are multiplied in gardens from the true species, and in defining these by well-marked charac- ters. Some strawberries whose fruit is poor have been abandoned, and the finest are the result of the crossing of the species of Virginia and Chili, of which I am about to speak. Virginian Strawberry — Fragaria virginiana, Ehrarht. The scarlet strawberry of French gardens. This species, indigenous in Canada and in the eastern States of America, and of which one variety extends west as far as the Rocky Mountains, perhaps even to Oregon,^ was introduced into English gardens in 1629.^ It was much cultivated in France in the last century, but its hybrids with other species are now more esteemed. Chili Strawberry — Fragaria ChiloensiSf Duchesne. A species common in Southern Chili, at Conception, Valdivia, and Chiloe,^ and often cultivated in that country. It was brought to France by Frezier in the year 1715. Cultivated in the Museum of Natural History in France, it spread to England and elsewhere. The large size of the berry and its excellent flavour have produced by diflferent crossings, especially with F. virginiana, the highly prized varieties Ananas, Victoria, Trollope, Ruhis, etc. Bird-Cherry — Prunus avium, Linnaeus ; SilssJcirscJi- haum in German. I use the word cherry because it is customary, and has no inconvenience when speaking of cultivated species or varieties, but the study of allied wild species confirms the opinion of Linnaeus, that the cherries do not form a separate genus from the plums. All the varieties of the cultivated cherry belong to two species, wdiich are found wild : 1. Prunus avium, Linnaeus, tall, with no suckers from the roots, leaves * Asa Gray, Manual of Botany of the Northern States, edit. 1SG8, p. 155 ; Botany of California; i. p. 177. * Phillips, Romar. Brit., p. 335. ^ CI. Gay, Hist. Chili, Botanica, ii. p. 305. 206 ORIGIN OF CULTIVATED PLANTS. downy on the under side, the fruit sweet; 2. Prunns cerasus, Linnreus, shorter, with suckers from the roots, leaves glabrous, and fruit more or less sour or bitter. The first of these species, from which the white and black cherries are developed, is wild in Asia ; in the forest of Ghilan (north of Persia), in the Prussian provinces to the south of the Caucasus and in Armenia ; ^ in Europe in the south of Russia proper, and generally from the south of Sweden to the mountainous parts of Greece, Italy, and Spain.^ It even exists in Algeria.^ As we leave the district to the south of the Caspian and Black Seas, the bird-cherry becomes less common, less natural, and determined more peril aps by the birds which seek its fruit and carry the seeds from place to palace.* It cannot be doubted that it was thus naturalized, from cultivation, in the north of India,^ in many of the plains of the south of Europe, in Madeira,^ and here and there in the United States ; "^ but it is probable that in the greater part of Europe this took place in prehistoric times, seeing that the agency of birds was employed before the first migrations of nations, perhaps before there were men in Europe. Its area must have extended in this region as the glaciers diminished. The common names in ancient lano^uag-es have been the subject of a learned article by Adolphe Pictet,^ but nothing relative to the origin of the species can be , deduced from them ; and besides, the diflferent species and varieties have often been confused in popular nomencla- ture. It is far more important to know whether archae- ology can tell us anything about the presence of the bird-cherry in Europe in prehistoric times. * Ledebour, Fl. Ross., ii. p. 6 ; Boissier, Fl. Orient., ii. p. 649. * Ledebour, ibid. ; Fries, Summa Scand., p. 46 ; Nyman, Conspec. Fl. Fur., p. 213 J Boissier, ibid.; Willkomm aud Lange, Prodr. FL Hisp., in. p. 245. ' Munby, Catal. Alger. ^ edit. 2, p. 8. * As the cherries ripen after the season when birds mic^rate, they disperse the stones chiefly in the neighboui-hood of the plautacions. ' Sir J. Hooker, Fl. of Brit. India. * Lowe, Manual of Madeira, p. 235. ' Darlington, Fl. Cestrica, edit. 3, p. 73. ^ Ad. Pictet, Origines Indo-Europ., edit. 2, vol. i. p. 2Sl. PLANTS CULTIVATED FOR THEIR FRUITS. 207 Heer gives an illustration of the stones of Prunus avium, in his paper on the lake-dwellings of Western Switzerland.-'- From what he was kind enou^'h to write to me, April 14*, 1881, these stones were found in the peat formed above the ancient deposits of the age of stone. De Mortillet ^ found similar cherry-stones in the lake-dwellings of Bourget belonging to an epoch not very remote, more recent than the stone age. Dr. Gross sent me some from the locality, also comparatively recent, of Corcelette on Lake Neuchatel, and Strobel and Pigorini discovered some in the *' terramare " of Parma.^ All these are settlements posterior to the stone age, and perhaps belono'incr to historic time. If no more ancient stones of this species are found in Europe, it will seem probable that naturalization took place after the Aryan migrations. Sour Cherry — Primus cerasus, Linnaeus ; Cerasus vul- garis, Miller ; Bauimueischel, Sauerhirschen, in German. The Montmorency and griotte cherries, and several other kinds known to horticulturists, are derived from this species.* Hohenacker^ ssi^Y Prumcs cerasus at Lenkoran, near the Caspian Sea, and Koch^ in the forests of Asia Minor, that is to say, in the north-east of that country, as that was the region in which he travelled. Ancient authors found it at Elisabethpol and Erivan, according to Ledebour."^ Grisebach^ indicates it on Mount Olympus of Bithynia, and adds that it is nearly wild on the plains of Macedonia. The true and really ancient habitation seems to extend from the Caspian Sea to the environs of Constantinople ; but in this very region Prunus avium is more common. Indeed, Boissier and Tchihatchetf do not appear to have seen P. cerasus even in the * Heer, Pflanzen der PfaliJhauien, p. 21<, figs. 17, 18, and p. 26. * In Kerrin, Etudes Prehist. sur la Savoie, p. 22. ^ Atte Soc. Ital. Sc. Nat., vol. vi. * For the numerous varieties which have common names in France, varying with the different provinces, see Duhamel^ Traite des Arbres, edit. 2, vol. v., in which are good coloured illustrations. ' Hohenacker, Plantce Talxjsch., p. 128. ® Koch, Dendrologie, i. p. 110. ' Ledebour, Fl. R >ss., ii. p. 6. ® Grisebach, Sjoicil. Fl. Rumel., p. 86. 208 ORIGIN OF CULTIVATED PLANTS. Pontus, though they received or brought back several specimens of P. avium} In the north of India, P. cerasiis exists only as a cultivated plant.^ The Chinese do not appear to have been acquainted with our two kinds of cherry. Hence it may be assumed that it was not ver}^ early introduced into India, and the absence of a Sanskrit name confirms this. We have seen that, according to Grisebach, P. cerasus is nearly wild in Macedonia. It was said to be wild in the Crimea, but Steven ^ only saw it cultivated; and Rehmann * gives only the allied species, P. chaTiioi- cerasus, Jacquin, as wild in the south of Russia. I very much doubt its wild character in any locality north of the Caucasus. Even in Greece, where Fi-aas said he saw this tree wild, Heldreich only knows it as a cultivated species.^ In Dalmatia,^ a particular variety or allied species, P. Marasca, is found really wild; it is used in making Maraschino wine. P. cerasus is wild in mountainous parts of Italy '^ and in tlie centre of France,^ but farther to the west and north, and in Spain, the species is only found cultivated, and naturalized here and there as a bush. P. cerasus, more than the bird- cherry, evidently presents itself in Europe, as a foreign tree not completely naturalized. None of the often-quoted passages ^ in Theophrastus, Pliny, and other ancient authors appear to apply to P. cerasus}^ The most important, that of Theophrastus, belongs to Prunus avium, because of the height of the tree, a character which distinguishes it from P. cerasus, Kerasos being the name for the bird-cherry * Boissier, Fl. Orient., iL p. 619; TcUhatchefif, Asie Mineure, Bot, p. 198. 2 Sir J. Hooker, Fl. of Brit. India, ii. p. 313. ^ Steven, Verzeichniss Halhinselm, etc., p. 147. ■* Eehmann, Verhandl. Nat. Ver. Brunn, x. 1871. ^ Heldreich, Isutzpji. Griech., p. 69 j PfMnzen d'Attisch. Ehene., p. 477. " Viviani, Fl. Dalmat., iii. p. 258. ' Bertoloni, Fl. Ital., v. p. 131. ^ Lecoc and Lamotte, Catal. du Plat. Centr. de la France, p. 148. ^ Theophrastes, Hist. PL, lib. 3, c. 13 ; Pliny, lib. 15, c. 25, and others quoted in Lenz, Bot. der Alien Gr. and Eom., p. 710, *" Part of the description of Theophrastus shows a confusion with other trees. He says, for instance, that the nut is scft. PLANTS CULTIVATED FOE THEIR FRUITS. 209 in Theophrastus, as now kerasaia among the modern Greeks, I notice a linguistic proof of the antiquity of P. cerasus. The Albanians, descendants of the Pelas- gians, call the latter vyssine, an ancient name which reappears in the German Wechsel, and the Italian visciolo} As the Albanians have also the name kerasie for P. avium, it is probable that their ancestors very clearly distinguished the two species by different names, perhaps before the arrival of the Hellenes in Greece. Another indication of antiquity may be seen in Virgil {Geor. ii. 17)— " Pullulat ab radice aliis densissima silva Ut cerasis nlmisqne " — which applies to P. cerasus, not to P, avium. Two Y)aintings of the cherry tree were found at Pompeii, but it seems that it cannot be discovered to which of the two species they should be attributed.^ Comes calls them Prunus cerasus. Any archaeological discovery would be more con- vincing. The stones of the two species present a differ- ence in the furrow or groove, which has not escaped the observation of Heer and Sordelli. Unfortunately, only one stone of P. cerasus has been found in the pre- historic settlements of Italv and Switzerland, and what is more, it is not quite certain from what stratum it was taken. It appears that it was a non-archaeological stratum.^ From all these data, somewhat contradictory and sufficiently vague, I am inclined to admit that Prunus cerasus was known and already becoming naturalized at the beginning of Greek civilization, and a little later in Italy before the epoch when Lucullus brought a cherry tree from Asia Minor. Pages might be tran- scribed from authors, even modern ones, who attribute, after Pliny, the introduction of the cherry into Italy to * Ad. Pictet quotes forms of the same name in Persian, Turkish, and Russian, and derives from the same source the French word guigne, now used for certain varieties of the clierry. * Schouw, Die Erde, p. 44 ; Comes, III. delle Piante, etc., in 4to, p. 5G. ^ Sordelli, Piante della torhicra di Lagozza, p. 40. 210 OEIGIN OF CULTIVATED PLANTS. this rich Koman, in the year 65 B.C. Since this error is perpetuated by its incessant repetition in classical schools, it must once more be said that cherry trees (at least the bird-cherry) existed in Italy before Lucullus, and that the famous gourmet did not need to go far to seek the species with sour or bitter fruit. I have no doubt that he pleased the Eomans with a good variety cultivated in the Pontua, and that cultivators hastened to propagate it by grafting, but Lucullus' share in the matter was confined to this. From what is now known of Kerasunt and the ancient names of the cherry tree, I venture to maintain, contrary to the received opinion, that it was a variety of the bird-cherry of which the fleshy fruit is of a sweet flavour. I am inclined to think so because Kerasos in Theophrastus is the name of Prunus avium, which is far the commoner of the two in Asia Minor. The town of Kerasunt took its name from the tree, and it is probable that the abundance of Prunus avium in the neighbouring woods had induced the inhabitants to seek the trees which yielded the best fruits in order to plant them in their gardens. Certainly, if Lucullus brought fine white-heart cherries to Rome, his countrymen who only knew the little wild cherry may well have said, *' It is a fruit which Ave have not." Pliny alfirms nothing more. I must not conclude without suggesting a hypothesis about the two kinds of cherry. They difler but little in character, and, what is very rare, their two ancient habitations, which are most clearly proved, are similar (from the Caspian Sea to Western Anatolia). The two species have spread towards the West, but rmequally. That which is commonest in its ori spinal home and the stronger of the two (P. avium) has extended further and at an earlier epoch, and has become better naturalized. P. cerasus is, therefore, perhaps derived from the other in prehistoric times. I come thus, by a diflerent road, to an idea suggested by Caruel ; ^ only, instead of saying that it would perhaps be better to unite them Caruel, Flora Toscana, p. 48. PLANTS CULTIVATED FOR THEIR FRUITS. 211 now in one species, I consider tliem actually distinct, and content myself with supposing a descent, which for the rest it would not be easy to prove. Cultivated Plums. Pliny ^ speaks of the immense quantity of plums known in his time : ingens tiirba prunoritni. Horti- culturists now number more than three hundred. Some botanists have tried to attribute these to distinct wild species, but they have not always agreed, and judging from the specific names especially they seem to have had very different ideas. This diversity is on two heads ; first as to the descent of a given cultivated variety, and secondly as to the distinction of the wild forms into species or varieties. I do not pretend to classify the innumerable culti- vated forms, and I think that labour useless when deal i no- with the question of geographical origin, for the differ- ences lie principally in the shape, size, colour, and taste of the fruit, in characters, that is to say, which it has been the interest of horticulturists to cultivate when they occur, and even to create as far as it was in their power to do so. It is better to insist upon the distinction of the forms observed in a wild state, especially upon those from which man derives no advantage, and which have probably remained as they were before the existence of gardens. It is probably only for about thirty years that botanists have given really comparative characters for the three species or varieties which exist in nature.^ They may be summed up as follows : — Prunus domestica, Linnseus. Tree or tall shrub, with- out thorns ; young branches glabrous ; flowers appearing with the leaves, their peduncles usually downy; fruit pendulous, ovoid and of a sweet flavour. Prunus insititia, Linnaeus. Tree or tall shrub, with- out thorns ; young shoots covered with a velvet down ; flowers appearing with the leaves, with peduncles covered » Hist., lib. 15, c. 13. * Koch, Syn. Fl. Germ., edit. 2, p. 228 ; Cossou and Germain, Flore des Environs de Paris, i. p. 165. 212 ORIGIN OF CULTIVATED PLANTS. with a fine down, or glabrous ; fruit pendulous, round or slightly elliptical, of a sweet flavour. Primus spinosa, Linniieus. A thorny shrub, with branches spreading out at right angles ; young shoots downy ; flowers appearing before the leaves ; pedicles glabrous ; fruit upright, round, and very sour. This third form, so common in our hedges (sloe or blackthorn), is very different from the other two. There- fore, unless we interpret by hypothesis what may have happened before all observation, it seems to me im- possible to consider the three forms as constituting one and the same species, unless we can show transitions from one to the other in those organs which have not been modifled by cultivation, and hitherto this has not been done. At most the fusion of the two first cateofories can be admitted. The two forms with naturally sweet fruit occur in few countries. These must have tempted cultivators more than Prunus sjnnosay whose fruit is so sour. It is, therefore, in these that we must seek to find the originals of cultivated plums. For greater clearness I shall speak of them as two species.^ Common Plum — Prunus domestica, LinnjBus ; Zwet- cken in German. Several botanists ^ have found this variety wild throughout Anatolia, the region to the south of the Caucasus and Northern Persia, in the neighbourhood of Mount Elbruz, for example. I know of no proof for the localities of Kashmir, the country of the Kirghis and of China, which are men- tioned in some floras. The species is often doubtful, and it is probably rather Prunus insititia ; in other cases it is its true and ancient wild character which is un- certain, for the stones have evidently been dispersed from cultivation. Its area does not appear to extend as far as Lebanon, although the plums cultivated at Damascus (damascenes, or damsons) have a reputation which dates * Hudson, Fl. Anglic.^ 1778, p. 212, unites them nndcr the name Prunus communis. ' Ledebour, Fl. Ross.j ii. p. 5 ; BoiBfiPr, Fl. Orient, ii. p. G52 ; K Koch Dendrologie, i. p. 94; Boissier and Eiibs*-, Aufzahl Transcaucaslen, p. 80l PLANTS CULTIVATED FOR THEIR FRUITS. 213 from the cla3^s of Pliny. It is supposed that this was the species referred to by Dioscorides^ under the name of Syrian cocciiinelea, growing at Damascus. Karl Koch relates that the merchants trading on the borders of China told him that the species was common in the forests of the western part of the empire. It is true that the Chinese have cultivated different kinds of plums from time immemorial, but we do not know them well enough to judge of them, and we cannot be sure that they are indigenous. As none of our kinds of plum has been found wild in Japan or in the basin of the river Amur, it is very probable that the species seen in China are different to ours. This appears also to be the result of Bretschneider's statements.^ It is very doubtful if Prunus domestica is in- digenous in Europe. In the south, where it is given, it grows chiefly in hedges, near dwellings, with all the appearance of a tree scarcely naturalized, and maintained here and there by the constant bringing of stones from plantations. Authors who have seen the species in the East do not hesitate to say that it is " subspontaneous." Fraas ^ affirms that it is not wild in Greece, and this is confirmed as far as Attica is concerned by Heldreich.* Steven^ says the same for the Crimea. If this is the case near Asia Minor, it must be the more readily admitted for the rest of Europe. In spite of the abundance of plums cultivated formerly by the Romans, no kind is found represented in the frescoes at Pompeii.^ Neither has Prunus domestica been found among the remains of the lake-dwellings of Italy, Switzerland, and Savoy, where, however, stones of Prunus insititia and spinosa have been discovered. From these facts, and the small number of words at- tributable to this species in Greek authors, it may be ^ Dioscorides, p. 174. ^ Bretschneider, On the Study, etc., p. 10. ' Fraas, Syn. Fl. Class., p. 69. ^ Heldreich, Pjiayizen Attischen Eheve. * Steven, Verzeichniss Halbinseln, i. p. 172. ® Comes, III. Piante Pompeiane. 214 ORIGIN OF CULTIVATED PLANTS. inferred that its half- wild or half-naturalized state dates in Europe from two thousand years at most. Prunes and damsons are ranked with this species. BuUace — Pr units insititia, LinniTeus;^ Pjiauenhaum and Haferschlehen in German. This kind of plum grows wild in the south of Europe.^ It has also been found in Cilicia, Armenia, to the south of the Caucasus, and in the province of Talj^sch near the Caspian Sea.^ It is especially in Turkey in Europe and to the south of the Caucasus that it appears to be truly wild. In Italy and in Spain it is perliaps less so, although trustworthy authors who have seen the plant OTOwinof have no doubt about it. In the localities named north of the Alps, even as far as Denmark, it is probably naturalized from cultivation. The species is commonly found in hedges not far from dwellings, and apparently not truly wild. All this aOTees with archfeoloirical and historical data. The ancient Greeks distinguished the Coccumelea of their country from those of Syria,'* whence it is inferred that the former were Primus insititia. This seems the more likely that the modern Greeks call it coromeleia.^ The Albanians say coroinhile,^ which has led some people to suppose an ancient Pelasgian origin. For the rest, we must not insist upon the common names of the plum which each nation may have given to one or another species, perhaps also to some cultivated variety, without any rule. The names which have been much commented upon in learned works generally, a]:>pear to me to apply to any plum or plum tree without having any very defined meaning. No stones of P. insititia have yet been found in ^ Insititia = foreign, A curious name, since eveiy plant is foreign to all countries but its own. * Willkomm and Lange, Prodr. Fl. Hisp., iii. p. 214 ; Bertoloni, Fl. Ital., V. p. 135; Grisebach, 5j3iccL FL Rumel.,-p, 85 j Heldreich, Nutzjjfl. Griech., p. G8. ' Boissier, FZ. Orient., ii. p. 651 ; Ledebour, Fl. Ross., ii. p. 5 ; Ilolien- acker, PL Talysch, p. 128. * Dioscorides, p. 173 ; Fraas, Fl. Class., p. 69. * Heldreich, Nutzpjlanzen Griechenlands, p, 68. ' Ibid. PLANTS CULTIVATED FOR THEIR FRUITS. 215 the terra-mare of Italy, but Heer has described and given illustrations of some whicli were found in the lake- dwellings of Robenhausen.^ The species does not seem to be now indigenous in this part of Switzerland, but we must not forget that, as we saw in the history of flax, the lake-dwellers of the canton of Zurich, in the age of stone, had communications with Italy. These ancient Swiss were not hard to please in the matter of food, for they also gathered the berries of the blackthorn, which are, as we think, uneatable. It is probable that they ate them cooked. Apricot — Primus arinenia^ca, Linnoeus ; Arsenica vulgaris, Lamarck. The Greeks and Romans received the apricot about the beofinnina: of the Christian era. Unknown in the time of Theophrastus, Dioscorides ^ mentions it under the name of viailoii arineniacon. He says that the Latins called it praiholdon. It is, in fact, one of the fruits mentioned briefly by Pliny,^ under the name of prcecocium, so called from the precocity of the species.* Its Armenian origin is indicated by the Greek name, but this name might mean only that the species was cidtivated in Armenia. Modern botanists have long had good reason to believe that the species is wild in that country. Pallas, Guldenstadt, and Hohenacker say they found it in the neighbourhood of the Caucasus Mountains, on the north, on the banks of the Terek, and to the south between the Caspian and Black Seas,^ Boissier ^ admits all these localities, but without saying any tiling about the wild character of the species. He saw a specimen gathered by Hohenacker, near ElisabethpoL On the * Heer, PJlanzen der Pfahlhauten, p. 27, fi?. 16, c. ^ Dioscorides, lib. 1, c. 165. ' Pliny, lib. 2, ca;?. 12. * The Latin name lias passed into modern Greek {prilcol^kia). The Spanish and French names, etc. (alharicoque, ahricot), seem to be derived from arbor pracox, or prcecocium, while the old French word annegne, and the Italian armenilU, etc., come from mailon armeniacon. See further details about the names of the species in my Geographie Botuniiue Raisonnee, p. 880. 5 Ledebour, Fl. Ross.y ii. p. 3. * Boissier, Fl, Orient., ii. p. 652. 216 OJllGIN OF CULTIVATED PLANTS. other hand, TchihatchefF^ who has crossed Anatolia and Armenia several times, does not seem to have seen the Avild apricot; and what is still more significant, Karl Koch, who travelled through the region to the south of the Caucasus, in order to observe facts of this nature, expresses himself as follows : ^ " Native country unknown. At least, during my long sojourn in Armenia, I nowhere found the apricot wild, and I have rarely seen it even cultivated." A traveller, W. J. Hamilton.^ said he found it wild near Oro-ou and Outch Hisar in Anatolia : but this asser- tion has not been verified by a botanist. The supposed wild apricot of the ruins of Baalbek, described by Eusebe de Salle * is, from what he says of the leaf and fruit, totally different to the common apricot. Boissier, and the different collectors who sent him plants from Syria and Lebanon, do not appear to have seen the species. Spach ^ asserts that it is indigenous in Persia, but he gives no proof. Boissier and Buhse ^ do not mention it in their list of the plants of Transcaucasia and Persia. It is use- less to seek its origin in Africa. The apricots which Reynier "^ says he saw, " almost wild," in Upper Eg3'pt must have sprung from stones grown in cultivated ground, as is seen in Algeria,^ Schweinfurth and Ascherson,^ in their catalogue of the plants of Eg^^pt and Abyssinia, only mention the species as cultivated. Besides, if it had existed formerly in the north of Africa it would have been early known to the Hebrews and the Romans. Now there is no Hebrew name, and Pliny sa^s its intro- duction at Rome took place thirty years before he wrote. Carrying our researches eastward, we find that Anglo- ^ Tcln'hatcheff, Asie ITineure, Botanique, voL L * K. Koch, Dendrologie, i. p. 87. * Noiiv. Ann. des Voya'jes, Feb., 1839, p. 176. * E. de Salle, Voyage, i. p. 140. * Spach, Hist, des Veget. Phaner,, i. p. 389. * Boissier and Buhse, Atifzdhlung, etc., in 4to, 1860. ' Eeynier, jEconomie des Egyptiens, p. 371. 8 Miinby, Catal. Fl. d^ Alger., edit. 2, p. 49. ' Schweinfuith and Ascherson, Bcitrage z. Fl. ^fhiop., in 4to., 1807, p. 259. PLANTS CULTIVATED FOR THEIR FRUITS. 217 Indian botanists-^ are agreed in considering that the apricot, which is generally cultivated in the north of India and in Thibet, is not wild in those regions; but they add that it has a tendency to become naturalized, and that it is found upon the site of ruined villages. IMessrs. Schlagintweit brought specimens from the north- west provinces of India, and irom Thibet, which West- mael verified,^ but he was kind enough to write to me that he cannot affirm that it was wild, since the collector's label gives no information on that head. Roxburgh,^ who did not neglect the question of origin, says, speaking of the apricot, " native of China as well as the west of Asia." I read in Dr. Bretschneider's curious little work,^ draAvn up at Pekin, the following passage, which seems to me to decide the question in favour of a Chinese origin : — " Sing, as is well known, is the apricot (Prunus armeniaca). The character (a Chinese sign [)rinted on p. 10) does not exist as indicat- ing a fruit, either in the Shti-Jdng, or in the Shi-Jcing, Cihouli, etc., but the Shan-hai-king says that several sings grow upon the hills (here a Chinese character). Besides, the name of the apricot is represented by a particular sign which may show that it is indigenous in China." The Shan-hai-Jcing is attributed to the Emperor Yli, who lived in 2205-2198 B.C. Decaisne,^ who was the first to suspect the Chinese origin of the apricot, has recently received from Dr. Bretschneider some specimens accompanied by the following note : — " No. 24, apricot wild in the mountains of Pekin, where it grows in abundance ; the fruit is small (an inch and a quarter in diameter), the skin red and yellow ; the flesli salmon colour, sour, but eatable. No. 25, the stone of the apricot cultivated round Pekin. The fruit is twice as larue as * Royle, III. of Himalaya, p. 205 ; Aitchison, Catal. of Punjab and Sindh, -p. 56 ; Sir Joseph Jlooker, Fl. of Brit. Ind., ii. p. 313 j JBiandis, Fared Flora of N. W. and Central India, 191. 2 Westtnael, in Bull. Soc. Bot. Belgiq., viii., p. 219. ' Roxburgh, Fl. Ind., edit. 2, v. ii. p. 501. * Bretschneider, On the Study and Value, etc., pp. IC, 49. ' Decaisne, Jardin Fruitier du Museum, vol. viii., art. Ahricotier, 218 OPJGIN OF CULTIVATED PLANTS. that of the wild tree." ^ Decaisne adds, in the letter he was good enough to write to me, " In shape and surface the stones are exactly like those of our small apricots ; they are smooth and not pitted." The leaves he sent me are certainly those of the apricot. The apricot is not mentioned in Japan, or in the basin of the river Amoor.^ Perhaps the cold of the winter is too erreat. If we recollect the absence of communication in ancient times between China and India, and the assertions that the plant is indigenous in both countries, we are at first tempted to believe that the ancient area extended from the north-west of India to China. How- ever, if we wish to adopt this hypothesis, we must also admit that the culture of the apricot spread very late towards the West.^ For no Sanskrit or Hebrew name is knowm, but only a Hindu name, zard alu, and a Persian name, mischmisch, which has passed into Arabic.^ How is it to be supposed that so excellent a fruit, and one which grows in abundance in Western Asia, spread so slowly from the north-west of India towards the Grcieco- Roman world ? The Chinese knew it two or three thousand years before the Christian era. Changkien went as far as Bactriana, a century before our era, and he was the first to make the West known to his fellow- countrymen.^ It was then, perhaps, that the apricot was introduced in Western Asia, and that it was cultivated and became naturalized here and there in the north-west of India, and at the foot of the Caucasus, by the scatter- ing of the stones beyond the limits of the plantations. Almond — Amygdalus communis, Linnreus ; Fruni species, Baillon ; Frunus Amygdalus, Hooker. * Dr. Bretschneider confirms this in a recent work, Notes on Botanical Questions, p. 3. 2 Prv.nus armeniaca of Thnnbcrg is P. mume of Siebold and ZuccTia- rini. The apricot is not mentioned in the Eiiumeratio, etc., of Fjancbet and Savatier, ' Capus (A7in. Sc. Nat., sixth series, vol. xv. p. 206) found it wild in Tni'kestan at the height of four thousatid to seven thousand feet, which weakens the hyjiothosis of a solely Chinese origin. * Piddington, Index ; Roxburgh, Fl. Ind. ; Forskal, Fl. J^gyp. ; Delile, III. Egypt. * Bretschneider, On the Study and Valuer etc. PLANTS CULTIVATED FOR THEIR FRUITS. 219 The almond grows apparently wild or half wild in the warm, dry regions of the Mediterranean basin and of western temperate Asia. As the nuts from cultivated trees naturalize the species very easily, w^e must have recourse to v^arious indications to discern its ancient home. We may first discard the notion of its origin in Eastern Asia. Japanese floras make no mention of the almond. That which M. de Bonge saw cultivated in the north of China was the Persica Davidiana} Dr. Bretschneider,^ in his classical work, tells us that he has never seen the almond cultivated in China, and that the compilation entitled Pent-sao, published in the tenth or eleventh century of our era, describes it as a tree of the country of the Maliometans, which signifies the north- west of India, or Persia. Anglo-Indian botanists ^ say that the almond is culti- vated in the cool parts of India, but some add that it does not thrive, and that many almonds are brought from Persia.* No Sanskrit name is known, nor even any in the languages derived from Sanskrit. Evidently the north-west of India is not the original home of the species. On the other hand, there are many localities in the region extending from Mesopotamia and Turkestan to Algeria, where excellent botanists have found the almond tree quite wild. Boissier ^ has seen specimens gathered in rocky ground in Mesopotamia, Aderbijan, Turkestan, Kurdistan, and in the forests of the Anti-Lebanon. Karl Koch ^ has not found it wild to the south of the Caucasus, nor Tchibatcheff in Asia Minor. Cosson "' found natural woods of almond trees near Saida in Aliieria. It o * Bretschneifler, Early European Researches, p, 149. ^ Bretsclineider, Study and ValiiCy etc., p. 10] and Early Europ, Besear.^ p. 140. ' Brandis, Forest Flora; Sir J. Hoolrer, Fl. of Brit. Tnd., iii. p. 318. * Roxburgh, Fl. Ind., edit. 2, vol. ii. p. 500 ; Eoyle, 111. Rimal.^ p. 204. ' Boissier, Fl. Orien., iii. p. 641. * K. Koch, Dendrologie, i. p. 80; Tchihatcheff, Asie Ulijieure Bota. niqne, i. p. 108. ' Ann. des Sc. Nat, 3rd scries, vol. xix. j). 108. 220 OKIGIN OF CULTIVATED PLANTS. is also regarded as wild on the coasts of Sicily and of Greece ; ^ but there, and still more in the localities in which it occurs in Italy, Spain, and France, it is probable, and almost certain, that it springs from the casual dis- persal of the nuts from cultivation. The antiquity of its existence in Western Asia is proved by Hebrew names for the almond tree — schaked, luz or lus (which recurs m the Arabic loiiz), and sche- kedhn for the nut.^ The Persians have another name, badani, but I do not know how old this is. Theophras- tus and Dioscorides ^ mention the almond by an entirely different name, amugdalai, translated by the Latins into aonygdalus. It may be inferred from this that the Greeks did not receive the species from the interior of Asia, but found it in their own country, or at least in Asia Minor. The almond tree is represented in several frescoes found at Pompeii.^ Pliny ^ doubts whether the species was known in Italy in Cato's time, because it was called the Greek nut. It is very possible that the almond was in- troduced into Italy from the Greek islands. Almonds have not been found in the terra-mare of the neio-h- bourhood of Parma, even in the upper layers. The late introduction of the species into Italy, and the absence of naturalization in Sardinia and Spain,^ incline me to doubt whether it is really indigenous in the north of Africa and Sicily. In the latter countries it was more probably naturalized some centuries ago. In confirma- tion of this hypothesis, I note that the Berber name of the almond, talouzet^ is evidently connected with the Arabic loiiz, that is to say with the language of the conquerors who came after the Romans. In Western Asia, on the contrary, and even in some parts of Greece, * Gussone, Synopsis Florae Siculce, i. p. 552 ; Heldreich, Niitzpjlanzen Grlechenlandi^, p. 67. ^ Hiller, Hierophyton, i. p. 215 j Rosenmiiller, Handh. Bihl. Alterth.y iv. p. 2(33. 3 Theophrastus, Eist., lib. 1, c. 11, 18, etc. ; Dioscorides, lib. 1, c. 170. * Schouw, Die Erde, etc.; Comes, III. Piante nei dipinti Pomp., p. 13. * Pliny, Hist., lib. 16, c. 22. * Moris, Flora Sardoa, ii. p. 5 ; Willkoimn and Lange, Prodr. Fl: Hisp,, iii. p. 243. ' Dictionnaire Fran^ais B< rbere, 1844. PLANTS CULTIVATED FOR THEIR FRUITS. 221 it may be regarded as indigenous from prehistoric time. I do not say primitive, for everything was preceded by something else. I remark finally that the difference be- tween bitter and sweet almonds was known to the Greeks and even to the Hebrews. Peach — Awygdalus persica, Linn?eus ; Persica vul- garis, Miller ; Prunus persica, Bentham and Hooker. I will quote the article in which I formerly^ attributed a Chinese origin to the peach, a contrary opinion to that which prevailed at the time, and which people who are not on a par with modern science continue to reproduce. I will afterwards give the facts discovered since 1855. '' The Greeks and Romans received the peach shortly after the be^inninfy of the Christian era. The names persica, Tiialum persicum, indicate whence they had it. I need not dwell upon those well-known facts.^ Several kinds of peach are now cultivated in the north of India,^ but, what is remarkable, no Sanskrit name is known ; * whence we may infer that its existence and its cultivation are of no great antiquity in these regions. Roxburgh, who is usually careful to give the modern Indian names, only mentions Arab and Chinese names. Piddington gives no Indian name, and Royle only Persian names. The peach does not succeed, or requires the greatest care to ensure success, in the north-east of India.^ In China, on the contrary, its cultivation dates from the remotest antiquity. A number of superstitious ideas and of legends about the properties of its diflerent varieties exist in that country.^ These varieties are very * Alph. de Candolle, Geogr. Bot. Bais., p. 881. ^ Theoplirastus, Hist., iv. c. 4 ; Dioscorides, lib. 1, c. 164 ; Plinj, Geneva edit., bk. 15, c. 13. 3 Royle, III. Him., p. 204. * Roxbui'gh, Fl. Intl., 2nd. edit., ii. p. 500; Piddington, Index; Royle, ibid. ' Sir Joseph Hooker, Journ. of Bot., 1850, p. 54. * Rose, the head of the French trade at Canton, collected these from Chinese mannscripts, and Noisette (Jard. Fruit., i. p. 76) has transcribed a part of his article. The facts are of the following nature. The Chinese believe the oval peaches, which are very red ou one side, to be a symbol of a long life. In consequence of this ancient belief, peaches are used iu all ornaments in painting and sculpture, and in congratulatoi'j pre. 222 ORIGIN OF CULTIVATED PLANTS. numerous ; ^ and m particular the singular variety with compressed or flattened fruit,^ which appears to be further removed than any other from the natural state of the peach ; lastly, a simple name, to, is given to the common peach.^ " From all these facts, I am inclined to believe that the peach is of Chinese rather than of western Asiatic origin. If it had existed in Persia or Armenia from all time, the knowledge and cultivation of so pleasant a fruit would have spread earlier into Asia Minor and Greece. The expedition of Alexander probably was the means of making it known to Theophrastus (332 B.C.), who speaks of it as a Persian fruit. Perhaps this vague idea of the Greeks dates from the retreat of the ten thousand (401 B.C.) ; but Xenophon does not mention the peach. Nor do the Hebrew writings speak of it. The peach has no Sanskrit name, yet the peoples who spoke this language came into India from the north-west; that is to say, from the generally received home of the species. On this hypothesis, how are we to account for the fact that neither the Greeks of the early times of Greece, nor the Hebrews, nor the Sanskrit-speaking peoples, who all radiated from the upper part of the Euphrates valley or communicated with it, did not cultivate the peach ? On the other hand, it is very possible that the stones of a fruit tree cultivated in China from the remotest times, should have been carried over the mountains from the centre of Asia into Kashmir, Bokliara, and Persia. The Chinese had very early discovered this route. The im- portation would have taken place between the epoch of the Sanskrit emigrations and the relations of the Persians with the Greeks. The cultivation of tlie peach, once sents, etc. According to the -work of Chin-nong-king, the peach Yu prevents death. If it is not eaten in time, it at least preserves tlie body from decay until the end of the -^vorld. The peach is always mentioned among the fruits of immortality, with which were entertained the hopes of Tsinchi-Hoang, Vouty, of the Hans and other emperors who pretended to immortality, etc. * Lindley, Trans. Hort. Soc, v. p. 121. * Trans. Hort. Soc. Lond., iv. p. 512, tab. 19, ' Hoxburgh, Fl. Ind. PLANTS CULTIVATED FOU THEIR FRUITS. 223 established in Persia, would have easily spread on the one side towards the west ; on the other, through Cabul towards the north of India, where it is not so very ancient. " In confirmation of the hypothesis of a Chinese origin, it may be added that the peach Avas introduced into Cochin-China from China,^ and that the Japanese give the Chinese name Tao ^ to the peach. M. Stanislas Julien was kind enough to read to me in French some passages of the Japanese encyclopaedia (bk. Ixxxvi. p. 7), in which the peach tree tao is said to be a tree of Western countries, which should be understood to mean the interior of China as compared to the eastern coast, since the passage is taken from a Chinese author. The tao occurs in the writino^s of Confucius in the fifth century before the Christian era, and even in the Ritual in the tenth century before Christ. Its wild nature is not specified in the encyclopaedia of which I have just spoken ; but Chinese authors pay little attention to this point." After a few details about the common names of the peach in different languages, I went on to say, " The absence of Sanskrit and Hebrew names remains the most important fact, whence we may infer an introduction into Western Asia from a more distant land, that is to say, from China. "The peach has been found wild in different parts of Asia ; but it is alwaj^s a question whether it is indige- nous there, or whether it sprang from the dispersion of stones produced by cultivated trees. The question is the more necessary since the stones germinate easily, and several of the modifications of the peach are hereditar}^^ Apparently wild peach trees have often been found in the neighbourhood of the Caucasus. Pallas * saw several on the banks of the Terek, where the inhabitants give * LoTireiro, Fl. Cochin., p. 386. * Kaempfer, J.mcB«-., p. 798; Thunberg, FL Jap., p. 199. Kaempfer and Thunberg also give the name momu, but Siebold (Fl. Jap., i. p. 29) attributes a somewhat similar name, mume, to a plum tree, Prunus murne, Sieb. and Z. ' Noisette, Jard. Fr., p. 77; Trans. 8oc. Hort. Lond., iv. p. 513- * Pallas, Fl. Rossica, p. 13. 224 ORIGIN OF CULTIVATED PLANTS. it a name which he calls Persian, scheptcda} It fruit is velvety, sour, not very fleshy, and hardly larger than a walnut ; the tree small. Pallas suspects that this tree has degenerated from cultivated peaches. He adds that it is found in the Crimea, to the south of the Caucasus, and in Persia; but Marshall, Bicberstein, Meyer, and Hohenacker do not give the wild peach in the neigh- bourhood of the Caucasus, Early travellers, Gmelin, Guldenstadt, and Georgi, quoted by Ledebour, mentioned it. C. Koch ^ is the only modern botanist who said he found the peach tree in abundance in the Caucasian provinces. Ledebour, however, prudently adds, Is it wild ? The stones which Brugniere and Olivier brought from Ispahan, which were sown in Paris and yielded a good velvety peach, were not, as Bosc ^ asserted, taken from a peach tree wild in Persia, but from one growing in a garden at Ispahan.^ I do not know of any proof of a peach tree found wild in Persia, and if travellers mention any it is always to be feared that these are only sown trees. Dr. Royle^ says that the peach grows wild in several places south of the Himalayas, notably near Mussouri, but we have seen that its culture is not ancient in these regions, and neither Roxburgh nor Don's Flora JS^epalensis mention the peach. Bunge ^ only found cul- tivated trees in the north of China. This country has hardly been explored, and Chinese legends seem some- times to indicate wild peaches. Thus the Choit-y-Jd, according to the author previously quoted, says, ' Who- soever eats of the peaches of Mount Kouoliou shall obtain eternal life.' For Japan, Thunberg "^ says, Crescit uhique vulgaris^ prcecipue juxta Kagasahi. In omni liorto colitur oh elegantiam florum. It seems from this passage that the species grows both in and out of gardens, but perhaps in the first case he only alludes to peaches growing in the open air and without shelter. * BTiuft aloo is, according to Royle {El. Him. p. 204), the Persian name for the nectarine. * Ledebour, Fl. Ross.,i. p. 3. Spc p. 228, the subsequent opinion of Koch. 3 Rose, Diet. d'Agric.^ ix. p. 481. * Thouin, Ami. Mus., viii. p. 433. 6 Rovle, III. Him., p. 204. « Buiigc, Enum. PL Chin., p. 23. ^ Thuuberg, Fl. Jap. 199. PLANTS CULTIVATED FOR THEIR FRUITS. 225 "I have said nothing hitherto of the distinction to be established between the different varieties or species of the peach, since most of them are cultivated in all countries — at least the clearly defined kinds, which may be considered as botanical species. Thus the great dis- tinction between the downy and smooth-skinned fruits (peaches proper and nectarines), on which it is proposed to found two species {Persica vulgaris, Mill, and P. Levis, D. C), exists in Japan ^ and in Europe, as in most of the intermediate countries.^ Less importance is attached to distinctions founded on the adherence or non-adherence of the skin, on the white, yellow, or red colour of the flesh, and on the general form of the fruit. The great division into peaches and nectarines presents most of these modifications in Europe, in Western Asia, and probably in China. It is certain that in the latter country the form of the fruit varies more than else- where ; for there are as in Europe oval peaches, and also the peaches of which I spoke just now, which are quite flattened, in which the top of the stone is not even covered with flesh.^ The colour also varies greatly.^ In Europe the most distinct varieties, nectarines and peaches, freestones and clingstones, existed three centuries ago, for J. Bauhin enumerates them very clearly ; ^ and before him Dalechamp, in 1587, also gave the principal ones.'' At that time nectarines were called Nucipevsica, because of their resemblance in shape, size, and colour to the walnut. It is in the same sense that the Italians call them pescanoce. " I have sought in vain for a proof that the nectarine existed in Italy in the time of ancient Rome. Pliny ,'^ who confounds in his compilation peaches, plums, the Lauriis Persea^ and perhaps other trees, says nothing 1 Thunberg, Fl. Jap., 199. * The accounts about China which I have consulted do not mention the nectarine ; but as it exists in Japan, it is extremely probable that it dots also in China. 3 Noisette, Jard. Fr., p. 77} Trans. Hort. Soc, iv. p. 512, tab. 19. * Lindley, Trans. Hort, Soc.,v. -p. 122. ^ J. Bauhin, lfi6-f,, i. pp. 1G2,1G3. ' Dalechamp, Hist., i. p. 295. ^ Pliny, lib. xv. cap. 12 and 13. ^ Pliny, Be Div. Gen. Malorum, lib. ii. cap. 14. 220 ORIGIN OF CULTIVATED PLANTS. which can apply to such a fruit. Sometimes people have thouo^ht thev recosfinzed it in the tuheres of which he speaks. It was a tree imported from Syria in the time of Augustus. There were both red and white tuheres. Others (tuheres ? or mala ?) of the neighbourhood of Verona were downy. Some graceful verses of Petronus, quoted by Dalechamp/ clearly prove that the tuheres of the Romans in Nero's time were a smooth-skinned fruit; but this might be the jujube (Zlzijphus), Diospyros, or some Crataegus, just as well as the smooth- skinned peach. Each author in the time of the Renais- sance had his opinion on this point, or criticized that of the others.^ Perhaps there were two or three species of tuheres, as Pliny says, and one of them which was grafted on plum trees was the nectarine (?) ^ but I doubt whether this question can ever be cleared up.^ " Even admitting that the NiLCvpersica was only intro- duced into Europe in the Middle Ages, we cannot help remarking that in European gardens for centuries, and in Japan from time unknown, there was an intermix- ture of all the principal kinds of peach. It seems that its different qualities were produced everywhere from a primitive species, which was probably the downy peach. If the two kinds had existed from the beginning, either they would have been in different countries, and their cultivation would have been established separately, or they would have been in the same country, and in this case it is probable that one kind would have been anciently introduced into this country and the other into that." I laid stress, in 1855, on other considerations in support of the theory that the nectarine is derived from the common peach ; but Darwin has given such a large number of cases in which a branch of nectarine has * Dalechamp, Sist., i. p. 358. 2 Dalechamp, ihid.; Mattliioli, p. 122 ; Cocsalpinus, p. 107; J. Bauhin, p. 163, etc. ^ Pliny, lib. xvii. cap. 10. * I havj not been able to discover an Italian name for a p^labrous or other fruit derived from /u6e •, or tuheres, which is sinpriilar, as tlio ancient names of frn'ts are usually preserved under some form or other. PLANTS CULTIVATED FOR THEIR FRUITS. 227 unexpectedly appeared upon a peach tree, that it is useless to insist longer upon this point, and I will only add that the nectarine has every appearance of an arti- ficial tree. Not only is it not found wild, but it never becomes naturalized, and each tree lives for a shorter time than the common peach. It is, in fact, a weakened form. " The facility," I said, " with which our peach trees are multiplied from seed in America, and have produced fleshy fruits, sometimes very fine ones, without the resource of grafting, inclines me to think that the species is in a natural state, little changed by a long cultivation or by hybrid fertilization. In Virginia and the neighbouring states there are peaches grown on trees raised from seed and not grafted, and their abundance is so great that brandy is made from them.^ On some trees the fruit is magnificent.^ At Juan Fernandez, says Bertero,^ the peach tree is so abundant that it is impossible to form an idea of the quantity of fruit which is gathered ; it is usually very good, although the trees have reverted to "a wild condition. From these instances it would not be surprising if the wild peaches with indifferent fruit found in Western Asia were simply naturalized trees in a climate not wholly favourable, and that the species was of Chinese origin, where its cultivation seems most ancient." Dr. Bretschneider,* who at Pekin has access to all the resources of Chinese literature, merely says, after reading the above passages, " Tao is the peach tree. De Candolle thinks that China is the native country of the peach. He may be right." The antiquity of the existence of the species and its wild nature in Western Asia have become more doubtful since 1855. Anglo-Indian botanists speak of the peach solely as a cultivated tree,^ or as cultivated and becoming naturalized and apparently wild in the north-west of India.^ Boissier^ mentions specimens gathered in Ghilan * Braddick, Trans. Bort. Soc. Lond., ii. p. 205. * Ihid., pL 13. * Bertero, Annales 8c. Nat., xxi. p. 350. * Bretscbneider, On the Studij and Value, etc., p. 10. fi Sir J. Hooker, Flora of Brit. Ind,, ii. p. 313, « Brandis, Forest Flora, etc,, p, 191, ' Buissier, Fl. Orient. ^ ii. p. G40. 11 228 ORIGIN OF CULTIVATED PLANTS. and to the south of the Caucasus, but he says nothing as to their wild nature ; and Karl Koch/ after tiavelling through this district, saj^s, speaking of the peach, " Country unknown, perhaps Persia. Boissier saw trees growing in the gorges on Mount H^miettiis, near Athens." The peach spreads easily in the conntries in which it is cultivated, so that it is hard to say whether a given tree is of natural origin and anterior to cultivation, or whether it is naturalized. But it certainly was first culti- vated in China ; it was spoken of there two thousand years before its introduction into the Greco-Roman world, a thousand years perhaps before its introduction into the lands of the Sanskrit-speaking race. The group of peaches (genus or subgenus) is composed of five forms, which Decaisne^ regards as species, but which other botanists are inclined to call varieties. The one is the common peach ; the second the nectarine, which we know to be derived ; the third is the flattened peach (P. platycarpa, Decaisne) cultivated in China ; and the two last are indigenous in China (P. simonii, Decaisne, and P. Davidii, Carriere). It is, therefore, essentially a Chinese group. It is diflicult, from all these facts, not to admit the Chinese origin of the common peach, as I had formerl}^ inferred from more scanty data. Its arrival in Italy at the beginning of the Christian era is now confirmed by the absence of peach stones in the terra-miare or lake- dwellings of Parma and Lorn bard y, and by the represen- tations of the peach tree in the yjaintings on the walls of the richer houses in Pompeii.^ I have yet to deal with an opinion formerly expressed by Knight, and supported by several horticulturists, that the peach is a modification of the almond. Darwin * collected facts in support of this idea, not omitting to mention one which seems opposed to it. They may be concisely put as follows : — (1) Crossed fertilization, which ^ K. Koch, Bendrologie, i. p. 83. * Decaisne, Jard. Fr. du Mni^., Peckers, p. 42. ^ Comes, Illns. Piante nei Dipinti Pompeinni, p. ]4. * Darwin, Variation of Plants and Animals, etc., i. p. 338. PLANTS CULTIVATED FOR THEIR FRUITS. 229 presented Knight with somewhat doubtful results ; (2) intermediate forms, as to the fleshiness of the fruit and the size of the nut or stone, obtained by sowing peach stones, or by chance in plantations, forms of which the almond-peach is an example which has long been known. Decaisne ^ pointed out differences between the almond and peach in the size and length of the leaves indepen- dently of the fruit. He calls Knight's theory a " strange hypothesis." Geographical botany opposes his hypothesis, for the almond tree has its orio^in in Western Asia ; it was not indio'enous in the centre of the Asiatic continent, and its introduction into China as a cultivated species was not anterior to the Christian era. The Chinese, however, ha.d already possessed for thousands of years different varieties of the common peach besides the two wild forms I have just mentioned. The almond and the peach, starting from two such widely separated regions, can hardly be considered as the same species. The one was established in China, the other in Syria and in Anatolia. The peach, after being transported from China into Central Asia, and a little before the Christian era into Western Asia, cannot, therefore, have produced the almond, since the latter existed already in Syria. And if the almond of Western Asia had produced the peach, how could the latter have existed in China at a very remote period while it was not known to the Greeks and Latins ? Pear — Pyrus coinimmis, Linnseus. The pear grows wild over the whole of temperate Europe and Western Asia, particularly in Anatolia, to the south of the Caucasus and in the north of Persia,^ per- haps even in Kashmir,^ but this is very doubtful. Some authors hold that its area extends as far as China. This opinion is due to the fact that they regard Pyrus sinensis, Lindley, as belonging to the same species. An examination of the leaves alone, of which the teeth are ' Decaisne, uhi supra, p. 2. 2 Lcdebour, Fl. Ross., ii. p. 94 ; B oissier, Fl. Orient., ii. p. 653. He has verified several specimens. " Sir J. Hooker, FL Brit. Ind., ii. p. 374. 230 ORIGIN OF CULTIVATED PLANTS. covered with a fine silky down, convinced me of the specific difference of the two trees.^ Our wild pear does not differ much from some of the cultivated varieties. Its fruit is sour, spotted, and narrowing towards the stalk, or nearly spherical on the same tree.^ With many other cultivated species, it is hard to distingruish the individuals of wild oricrin from those which the chance transport of seeds has produced at a distance from dwellings. In the present case it is not diflicult. Pear trees are often found in woods, and they attain to a considerable height, with all the con- ditions of fertility of an indigenous plant.^ Let us examine, however, whether in the wide area they occupy a less ancient existence may be suspected in some coun- tries than in others. No Sanskrit name for the pear is known, whence it may be concluded that its cultivation is of no long stand- ing in the north-west of India, and that the indication, which is moreover very vague, of wild trees in Kashmir is of no importance. Neither are there any Hebrew or Aramaic names,* but this is explained by the fact that the pear does not flourish in the hot countries in which these tongues were spoken. Homer, Theophrastus, and Dioscorides mention the pear tree under the names ochnai, apios, or achras. The Latins called it pyrus or pirus^ and cultivated a great * p. sinensis described by Lindley is badly drawn with regard to the indentation of the leaves in the plate in the Botanical Register, and very well in that of Decaisne's Jardin Fruitier die Museum. It is the same species as P. v.ssuriensis, Maximowicz, of Eastern Asia. * Well drawn in Duhamel, Traite des Arbres, edit. 2, vi. pi. 59 ; and in Decaisne, Jard. Frui. du Mus., pi. 1, figs. B and C. P. halansoi, pi. 6 of the same work, appears to be identical, as Boissier observes. ' This is the case in the forests of Lorraine, for instance, according to the observations of Godron, De VOrijine Probable des Poiriers Cultives, 8vo pamphlet, 1873, p. 6. * Rosenmiillcr, Bibl. Alterth. ; how, Aramaeische PJlanzennamen, 1881. ' The spelling Pyrus, adopted by Linnasus, occurs in Pliny, Historia, edit. 1631, p. 301. Some botanists, purists in spelling, write pirus, so that in referring to a modern work it is necessary to look in the index for both forms, or run the risk of believing that the pears are not in the work. In any case the ancient name was a common name; but the true botanical name is that of Linnaeus, funnier of the received nomen- clature, and Linnajus wrote Pyrus. PLANTS CULTIVATED FOR THEIR FRUITS. 231 number of varieties, at least in Pliny's time. The mural paintings at Pompeii frequently represent the tree with its fruit.^ The lake-dwellers of Switzerland and Italy gathered wild apples in great quantities, and among their stores pears are sometimes, but rarely, found. Heer has given an illustration of one which cannot be mistaken, found at Wangen or Kobenhausen. It is a fruit narrowing towards the stalk, 28 mm. (about an inch and a half) long by 19 mm. (an inch) wide, cut longitudinally so as to show the small quantity of pulp as compared to the cartilaginous central part.^ None have been found in the lake-dwellings of Bourget in Savoy. In those of Lombardy, Professor Raggazzoni ^ found a pear cut length- ways, 25 mm. by 16. This was at Bardello, Lago di Varese. The wild pears hgured in Duhamel, Traitedes Arhres, edit. 2, are 30 to 33 by 30 to 32 mm.; and those of Laristan, figured in the Jar din Fruitier du Museum under the name P. halansce, which seem to me to be of the same species, and undoubtedly wild, are 26 to 27 mm. by 24 to 25. In modern wild pears the fleshy part is a little thicker, but the ancient lake-dwellers dried their fruits after cutting them lengthways, which must have caused them to shrink a little. No knowledge of metals or of hemp is shown in the settlements where these were found; but, con- sidering their distance from the more civilized centres of antiquity, especially in the case of Switzerland, it is possible that these remains are not more ancient than the Trojan war, or than the foundation of Rome. I have mentioned three Greek and one Roman name, but there are many others ; for instance, jyauta in Armenian and Georgian ; vatzhov in Hungarian ; in Slav languages griisclia (Russian), hrussha (Bohemian), hvusha (lUyrian). Names similar to the Latin j^l/^^us recur in the Keltic languages ; j^ei?' in Erse, ^^ei* in Kymric and Armorican.* I leave philologists to conjecture the Aryan * Comes, III. Piante nei Dipinti Pompeiani, p. 59. 2 Heer, Pfahlhaiiten, pp. 24, 26, fig. 7. ' Sordelli, Notizie Stat. Lacustre di Lagozza. * Nemnicb, PoJyglott. Lex. Naturgesch. ; Ad. Pictet, Origines Indo- Europ., i. p. 277 j and my manuscx'ipt dictionary of common names. 232 ORIGIN OF CULTIVATED PLANTS. origin of some of these names, and of the German Bivn; I merely note their number and diversity as an indica- tion of the very ancient existence of the species from the Caspian Sea to the Atlantic. The Aryans certainly did not carry pears nor pear pips with them in their wander- ings westward; but if they found in Europe a fruit they knew, they would have given it the name or names they were accustomed to use, while other earlier names may have survived in some countries. As an example of the latter case, I may mention two Basque names, udarea and madaria^ which have no analogy with any known European or Asiatic name. The Basques being probably the descendants of the conquered Iberians who were driven back to the Pyrenees by the Kelts, the antiquity of their language is very great, and it is clear that their names for the species in question were not derived from Keltic or Latin. The modern area of the pear extending from the north of Persia to the western coast of temperate Europe, principally in mountainous regions, may therefore be con- sidered as prehistoric, and anterior to all cultivation. It must be added, however, tliat in the north of Europe and in the Biitish Isles an extensive cultivation must have extended and multiplied naturalizations in comparatively modern times which can scarcely be now distinguished. I cannot accept Godron's hypothesis that the numerous cultivated varieties come from an unknown Asiatic species.^ It seems that they may be ranked, as Decaisne sa3'S, either with P. comiiniinis or P. nivalis of which I am about to speak, taking into account the effect of accidental crossing, of cultivation, and of long- continued selection. Besides, Western Asia has been explored so thoroughly that it is probable it contains no other species than those already described. Snow Pear — Pyrus nivalis, Jacquin. This variety of pear is cultivated in Austria, in the north of Italy, and in several departments of the east and * From a list of plant-natnes sent hj M. d'Abadie to Professor Clos, of Toulouse. * Godron, ubi supra, p. 28. PLANTS CULTIVATED FOR THEIR FRUITS. 233 centre of France. It was named Pyriis nivalis by Jacquin^ from the German name ScJtiieehirn, given to it because the Austrian peasants eat the fruit when the snow is on the ground. It is called in France Poirier' sauger, because the under side of the leaves is covered with a white down which makes them like the sage (Fr. sauge). Decaisne^ considered all the varieties of P. nivalis to be derived from P. hotscliyana, Boissier,^ which grows wild in Asia Minor. The latter in this case should take the name oi nivalis, which is the older. The snowy pears cultivated in France to make the drink called perry have become wild in the woods here and there.* They constitute the greater number of the so-called " cider pears," which are distinguished by the sour taste of the fruit independent of the character of the leaf. The descriptions of the Greeks and Romans are too imperfect for us to be certain if they possessed this species. It may be presumed that they did, however, since they made cider.^ Sandy Pear, Chinese Pear — Pyrus sinensis, Lindley.^ I have already mentioned this species, which is nearly allied to the common pear. It is wild in Mongolia and Mantchuria,"^ and cultivated in China and Japan. Its fruit, large rather than good, is used for preserving. It has also been recently introduced into European gardens for experiments in crossing it with our species. This will very likely take place naturally. Apple — Pyrus Mains, Linnaeus. The apple tree grows wild throughout Europe ^ Jacquin, Flora Austriaca, ii. pp. 4, 107. "^ Decaisne, Jardin Fruitier du Museum, Poiriers, pi. 21. ' Decaisne, ibid., p. 18, and Introduction, p. 30. Several varieties of this species, of which a few bear a large fruit, are figured in the same work. * Boreau, Fl. dii Centre de la France, edit. 3, vol. ii. p. 236. * Palladius, Be re Rmfica, lib. 3, c. 25. For this purpose " ijira sylvestria vel asperi generis" were used. ® The Chinese quince had been called by Thonin P}jrus sinensis. Lindley has unfortunately given the same name to a true pyrus. ^ Decaisne {Jardin Fruitier d\h Musexun, Poiriers, pi. 5) saw speci- mens from both countries. Franchet and Savatiur give it as only cultivated in Japan. 234! ORIGIN OF CULTIVATED PLANTS. (excepting in the extreme north), in Anatolia, the south of the Caucasus, and the Persian province of Ghilan.^ Near Trebizond, the botanist Bourgeau saw quite a small forest of them.^ In the mountains of the north-west of India it is " apparently wild," as Sir Joseph Hooker writes in his Flora of British India. No author men- tions it as growing in Siberia, in Mongolia, or in Japan.^ There are two varieties wild in Germany, the one with glabrous leaves and ovaries, the other with leaves downy on the under side, and Koch adds that this down varies considerably.^ In France accurate autliors also give two wild varieties, but Avith characters which do not tally exactly with those of the German flora.^ It would be easy to account for this difference it' the wild trees in certain districts spring from cultivated varieties whose seeds have been accidentally dispersed. The question is, therefore, to discover to what degree the species is probably ancient and indigenous in different countries, and, if it is not more ancient in one country than another, how it was gradually extended by the accidental sowing of forms changed by the crossing of varieties and by cultivation. The country in which the apple appears to be most indio-enous is the rei^ion lying: between Trebizond and Ghilan. The variety which there grows wild has leaves downy on the under side, short peduncles, and sweet fruit,^ like Mains communis of France, described by Boreau. This indicates that its prehistoric area extended from the Caspian Sea nearly to Europe. Piddinoton o-ives in his Index a Sanskrit name for the apple, but Adolphe Pictef^ informs us that this * Nyman, Coni^pectus Florce EuropecB, p. 210; Ledebour, Flora Rossica, ii. p. 96; Boissier, Flora Orientalis, ii. p. 656; De-aisue, Nouv. Arch. Mus., X. p. 153. * Boissier, ibid. ^ Maximowicz, Pyi)n. JJssur.; Ee??el, Opit. FZon', etc., on the plants of Hic Ussuri collected bjMaak; Schmidt, Reisen Am'ir. Franchet and Savatier do not Tuention it in their Enum. Jap. Bretschneider quotes a Chinese name which, he says, applies also to other species. * Koch, Syn. Fl. Germ., i. p. 261. * Boreau, FL du Centre de la France, edit. 3, vol. ii. p. 236. * Boissier, uhi supra. ' Oi'i'j- Indo-Eur., i. p. 276. PLANTS CULTIVATED FOK THEIR FRUITS. 285 name seha is Hindustani, and comes from the Persian seb, sef. The absence of an earlier name in India argues that the now common cultivation of the apple in Kashmir and Thibet, and especially that in the north-west and central provinces of India, is not very ancient. The tree was probably known only to the western Aryans. This people had in all probability a name of which the root was ab, af, av, ob, as this root recurs in several European names of Aryan origin. Pictet gives ciball, ubliall, in Erse ; afal in Kymric ; aval in Armorican ; aphal in old High German ; appel in old English ; aj9^i in Scandinavian ; obolys in Lithuanian ; iahluko in ancient Slav ; iabloko in Russian. It would appear from this that the western Aryans, finding the apple wild or already naturalized in the north of Europe, kept the name under which they had known it. The Greeks had niailea or maila, the Latins mains, malitux, words whose origin, according to Pictet, is very uncertain. The Albanians, descendants of the Pelasgians, have mole'} Theophrastus ^ mentions wild and cultivated rtiaila. Lastly, the Basques (ancient Iberians) have an entirely different name, sagara, which implies an existence in Europe prior to the Aryan invasions. The inhabitants of the terra-mare of Parma, and of the palafittes of the lakes of Lombard y, Savoy, and Swit- zerland, made great use of apples. They always cut them lengthways, and preserved them dried as a provision for the winter. The specimens are often carbonized by fire, but the internal structure of the fruit is only the more clearly to be distinguished. Heer,^ who has shown great penetration in observing these details, distinguishes two varieties of the apple known to the inhabitants of the lake-dwellings before they poss3ssed metals. The smaller kind are 15 to 24? mm. in their longitudinal diameter, and about 3 mm. more across (in their dried and carbonized state) ; the larger, 29 to 32 mm. length- ways by 36 wide (dried, but not carbonized). The latter * Heldreich, Nutzpflanzen Grieclienlands, i. p. 64. ^ Theophrastus, Be Causis, lib. 6, cap. 24. ^ Heer, Pfahlhauten, p. 24, figs. 1-7. 236 OEIGIN OF CULTIVATED PLANTS. corresponds to an apple of German-Swiss orcl ar Js, now called campaney\ The English wild apple, figured in English Botany, pi. 179, is 17 mm. long by 22 wide. It is possible that the little apples of the lake-dwellings were wild ; hoAvever, their abundance in the stores makes it doubtful. Dr. Gross sent me two apples from tlie more recent palafittes of Lake Neuchatel; the one is 17 the other 22 mm. in longitudinal diameter. At Lagozza, in Lombardy, Sordelli^ mentions two apples, the one 17 mm. by 19, the other 19 mm. by 27. In a prehistoric deposit of Lago Varese, at Bardello, Ragazzoni found an a})ple in the stores a little larger than the others. From all these facts, I consider the apple to have existed in Europe, both wild and cultivated, from pre- historic times. The lack of communication with Asia before the Aryan invasion makes it probable that the tree was indigenous in Europe as in Anatolia, the south of the Caucasus, and Northern E;Ussia, and that its culti- vation began early everywhere. duince — Cydonia vulgaris, Persoon. The quince grows wild in the woods in the north of Persia, near the Caspian Sea, in the region to the south of the Caucasus, and in Anatolia.^ A few botanists have also found it apparently wild in the Crimea, and in the north of Greece;^ but naturalization may be suspected even in the east of Europe, and the further we advance towards Italy, especially towards the south-west of Europe and Algeria, the more it becomes probable that the species was naturalized at an early period round villages, in hedges, etc. No Sanskrit name is known for the quince, whence it may be inferred that its area did not extend towards the centre of Asia. Neither is there any Hebrew name, though the species is wild upon Mount Taurus.^ The Persian name is haivali,^ but I do not know whether * Sordelli, SuUe Piante della Stazi me di Logozza, p. 35. * Boissier, Fl. Orient., ii. p. 656 ; Ledebour, Fl. Ross., ii. p. 55. ' Steven, Verzeichniss Taurierif p. 150 j Sibtliorp, Prodr. FL Grcecx, i. p. 344. * Boissier, ihid. * Nemnich, Pohjglott Lexicon, PLANTS CULTIVATED FOR THEIR FRUITS. 237 it is as old as Zend. The same name, aiva, exists in Russian for the cultivated quince, while the name of the wild plant is armud, from the Armenian arinuda} The Greeks grafted upon a common variety, strution, a superior kind, which came from Cydon, in Crete, whence Kv^ujviov, translated by the Latin mcdiim cotoneuin, by cydonia, and all the European names, such as codogno in Italian, coudougner, and later coing in French, quitte in German, etc. There are Polish, pigiua, Slav, tiinja^ and Albanian (Pelasgian ?), ftua^ names which differ entirely from the others. This variety of names points to an ancient knowledge of the species to the west of its original country, and the Albanian name may even indicate an existence prior to the Hellenes. Its antiquity in Greece may also be gathered from the superstition, mentioned by Pliny and Plutarch, that the fruit of the quince was a preservation from evil influences, and from its entrance into the marriasfe rites prescribed by Solon. Some authors go so far as to main- tain that the apple disputed by Hera, Aphrodite, and Athene was a quince. Those who are interested in such questions will find details in Comes's paper on the plants represented in the frescoes at Pompeii.* The quince tree is figured twice in these, which is not sur- prising, as the tree was known in Cato's time.^ It seems to me probable that it was naturalized in the east of Europe before the epoch of the Trojan war. The quince is a fruit which has been little modified l)y cultivation; it is as harsh and acid when fresh as in the time of the ancient Greeks. Pomeg^ranate — Piinica granatumi, Linnaeus. The pomegranate grows wild in stony ground \n Persia, Kurdistan, Afghanistan, and Beluchistan.^ Burnes saw groves of it in Mazanderan, to the south of the Caspian Sea.'' It appears equally wild to the south ' ISTemnich, Po^V- T^ex. ^ Hid. ^ Heklreich, Nuts. Griech., p. 64 * In 4to, Napoli, 1879. ^ jj^ ^.g Eiu^tica, lib. 7, cap. 2. « Boissier, Fl. Orient., ii. p. 737; Sir J. Hooker, FL of Brit. Ind., ii. p. 581. ' Quoted from Royle, Illus. HimaL, p. 208. 238 OPJGIN OF CULTIVATED PLANTS. of the Caucasus.^ Westwards, that is to say, in Asia Minor, in Greece, and in the Mediterranean basin gene- rally, in the north of Africa and in Madeira, the species appears rather to have become naturalized from cultiva- tion, and by the dispersal of the seeds by birds. Many floras of the south of Europe speak of it as a " subspon- taneous" or naturalized species. Desfontaines, in his Atlantic Flora, gives it as wild in Algeria, but subsequent authors think ^ rather it is naturalized.^ I doubt its being wild in Beluchistan, where the traveller Stocks found it, for An^lo-Indian botanists do not allow it to be indi- genous east of tiie Indus, and I note the absence of the species in the collections from Lebanon and Syria which Boissier is always careful to quote. In China the pomegranate exists only as a cultivated plant. It was introduced from Samarkhand by Chang- Kien, a century and a half before the Christian era.* The naturalization in the Mediterranean basin is so general that it may be termed an extension of the original area. It probably dates from a very remote period, for the cultivation of the species dates from a very early epoch in Western Asia. Let us see whether historical and philological data can give us any information on this head. I note the existence of a Sanskrit name, darhnha, whence several modern Indian names are derived.^ Hence we may conclude that the species had long been known in the regions traversed by the Aryans in their route towards India. The pomegranate is mentioned several times in the Old Testament, under the name of Tmimon,^ whence the Arabic i^uininan or riiman. It was one of the fruit trees of the promised land, and the Hebrews had learnt to appreciate it in Eg;y'ptian gardens. Many localities in Palestine took their name from this * Ledebour, Fl. Ross., ii. p. 104. 2 Munby, FL Alger., p. 49 ; Spicilegium Flora Maroccanae, p. 458. ' Boissier, ihid. * Bretschneider, On Study and Value, etc., p. IG. * Piddington, Index. " Ilosenmullei", Bill. Katurge., i. p. 273 ; Ilauiilton, La Bot. de la Bible, Nice, 1871, p. 48. PLANTS CULTIVATED FOR THEIR FRUITS. 239 shrub, but the Scriptures only mention it as a cultivated species. The flower and the fruit figured in the religious rites of the Phoenicians, and the goddess Aphrodite had herself planted it in the isle of Cyprus/ which implies that it was not indigenous there. The Greeks were acquainted with the species in the time of Homer. It is twice mentioned in the Odyssey as a tree in the gardens of Pheeacia and Phrygia. They called it roia or roa, which philologists believe to be derived from the Syrian and Hebrew name,^ and also sidai,^ which seems to be Pelasgic, for the modern Albanian name is sige.^ There is nothing to show that the species was wild in Greece, where Fraas and Heldreich affirm that it is now only naturalized.^ The pomegranate enters into the myths and religious ceremonies of the ancient Romans.^ Cato speaks of its properties as a vermifuge. According to Pliny j*^ the best pomegranates came from Carthage, hence the name Malum punicuTii ; but it should not be supposed, as it has been assumed, that the species came originally from Northern Africa. Very probably the Phoenicians had introduced it at Carthasfe lono- before the Romans had anything to do with this town, and it was doubtless cultivated as in Egypt. If the pomegranate had formerly been wild in Northern Africa and the south of Europe, the Latins would have had more original names for it than granatum (from granum ?) and Malum inmicum. We should have perhaps found local names derived from ancient Western tongues ; whereas the Semitic name rimmon has prevailed in Greek and in Arabic, and even occurs, throuo-h Arab influence, among the Berbers.^ It must be admitted that the African origin is one of the errors caused by the erroneous popular nomenclature of the Romans. Leaves and flowers of a pomegranate, described by * Hehn, Ctiltur und Hansthiere axis Asicn, edit. 3, p. 106. 2 Hehn, ibid. ^ Lenz, Bot. der Alien Grie. und Rom., p. GSl. * Heldreich, Die Nutzpflanzen Griechenlands, p. 64. * Fraas, Fl. Class., p. 79 ; Heldreich, ibid. " Hehn, ibid. 7 Pliny, lib. 13, c. 19. * Dictionnaire Franpais-Berbdre, published by the French Government. 240 ORIGIN OF CULTIVATED PLANTS. Saporta^ as a variety of the rnodern Punica granatum, have been discovered in the pliocene strata of the environs of Meximieux. The species, therefore, existed under this form, before our epoch, along with several species, some extinct, others still existing in the south of Europe, and others in the Canaries, but the continuity of existence down to our own day is not thereby proved. To conclude, botanical, historical, and philological data agree in showing that the modern species is a native of Persia and some adjacent countries. Its cultivation began in prehistoric time, and its early extension, first towards the west and afterwards into China, has caused its naturalization in cases which may give rise to errors as to its true origin, for they are frequent, ancient, and enduring. I arrived at these conclusions in 1869,^ which has not prevented the repetition of the erroneous African orio^in in several works. Rose Apple — Eugenia Jamhos, Linnseus; Jamhosa vulgaris, de Candolle. This small tree belongs to the family of Myrtaceae. It is cultivated in tropical regions of the old and new worlds, as much perhaps for the beauty of its foliage as for its fruit, of which the rose-scented pulp is too scanty. There is an excellent illustration and a good description of it in the Botanical Magazine, pi. 335G. The seed is poisonous.^ As the cultivation of this species is of ancient date in Asia, there was no doubt of its Asiatic oricrin ; but the locality in which it grew wild was formerly unknown. Loureiro's assertion that it grew in Cochin- China and some parts of India required confirmation, which has been aftbrded by some modern writers.* The jamhos is wild in Sumatra, and elsewhere in the islands of the Malay Archipelago. Kurz did not meet with it in the forests of British Burmah, but when Rheede saw this tree in gardens in Malabar he noticed that it was. called Malacca-schaonbu, which shows that it came origi- » De Saporta, Bull. Soc. GeoL de France, April 5, ISGJ, pp. 767-769. Geogr. Bot. Eais., p. 191. ' Descourtilz, Flore Medicale des AnHllea, v. pL 315. * Miqiie\ Syrmtra, p. 118; Flora IndlcB-Batavce, i. p. 425 Bliimc; Museum Lugd.-Bat., i. p. 9j. PLANTS CULTIVATED FOR THEIR FRUITS. 241 nally from the Malay Peninsula. Lastly, Brandis says it is wild in Sikkim, to the north of Bengal. Its natural area probably extends from the islands of the Malay Archipelago to Cochin-China, and even to the north-east of India, where, however, it is probably naturalized from cultivation aud by the agency of birds. Naturalization has also taken place elsewhere — at Hong-kong, for in- stance, in the Seychelles, Mauritius, and Rodriguez, and in several of the West India Ishmds.^ Malay Apple — Eugenia Tiialaccensis, Linnreus; Jam- hosa malaccensis, de Candolle. A species allied to Eugenia jamhos, but differing from it in the ari-angement of its flowers, and in its fruit, of an obovoid instead of ovoid form ; that is to say, the smaller end is attached to the stalk. The fruit is more fleshy and is also rose-scented, but it is much^ or little ^ esteemed according to the country and varieties. These are numerous, ditfering in the red or pink colour of the flowers, and in tlie size, shape, and colour of the fruit. The numerous varieties show an ancient cultivation in the Malay Archipelago, where the species is indigenous. In confirmation, it must be noted that Forster found it established in the Pacific Islands, from Otahiti to the Sandwich Isles, at the time of Cook's voyages.* The Malay apple grows wild in the forests of the Malay Archipelago, and in the peninsula of Malacca.^ Tussac says that it was brought to Jamaica from Otahiti in 1793. It has spread and become naturalized in several of the West India Islands, also in Mauritius and the Seychelles.^ Guava — Psidium giiayava, Kaddi. Ancient authors, Linnaeus, and some later botanists, ^ Hooker, Fl. Brit. Ind., ii. p. 474 ; Baker, Fl. of Maurit., etc., p. 115; Grisebacb, Fl. of Brit. W. Ind. Isles, p. 235. ^ Eumphius, Amhoin, i. p. 121, t. 37. ' Tussac, Flore des Aniilles, iii. p. 89, pi. 25. * Forster, Plantis Esculentis, p. 36. * Blume, Museum Lnijd..Bat., i. p. 91 ; Miqucl, Fl. Inditv-Batav., i. p. 411 ; Hooker, Flo7-a of British India, ii. p. 472. « Grisebach, FL Brit. W. Indies, p. 235 ; Baker, Fl. of Mauritius, p. 115. 242 OlilGlN OF CULTIVATED PLANTS. admitted two species of this fruit tree of the family of Myrtacere, the one with elliptical or spherical fruit, with red flesh, Psidium j)om'ifcrum ; the other with a pyriform fruit and white or pink flesh, more agreeable to the taste. Such diversity is also observed in pears, apples, or peaches; so it was decided to consider all the Psidii as forming a single species. Raddi saw a proof that there was no essential difference, for he observed pyriform and round fruits growing on the same tree in Brazil.^ The majority of botanists, especially those who have observed the guava in the colonies, follow the opinion of Raddi,^ to which I was inclined, even in 1855, from reasons drawn from the geogTaphical distribution,^ Lowe,'* in his Flora of Madeira, maintains Avith some hesitation the distinction into two species, and asserts that each can be raised from seed. They are, therefore, races like those of our domestic animals, and of many cultivated plants. Each of these races comprehends several varieties.^ The study of the origin of the guava presents in the hio'hest deo^ree the difficultv which exists in the case of many fruit trees of this nature : their fleshy and some- what aromatic fruits attract omnivorous animals which cast their seeds in places far from cultivation. Those of the guava germinate rapidly, and fructify in the thii'd or fourth year. Its area has thus spread, and is still spreading by naturalization, principally in those tropical countries which are neither very hot nor very damp. In order to simplify the search after the origin of the species, I may begin by eliminating the old world, for it is suflicieutly evident that the guava came from America. * Eaddi, Bi Alcune Specie di Pero Indiano, in 4to, Bologrna, 1821, p. 1. * Martins, Syst. Nat. Mcdicm Bras., p. 32; lilume, Musernn Lugd.- Bat., i. p. 71 : Hasskarl, in jPZom, 1S14, p. 589; Sir J. Hooker, Fl. of Brit, hid., ii. p. 468. « Geogr, Bot. Rais., p. 893. * Lowe, Flora of Madeira, p. 26G, * See Biume, tJ;id. ; Descourtilz, JF7o?-e Mc'dicale des Antilles, ii. p. 20, in which there is a good illustration of the pyriform guava. Tussac, Flore des Antilles, gives a good plate of the round form. These two latter works furnish interesting details on the use of the guava, on the vegetation of the species, etc. PLANTS CULTIVATED I OR THEIR FRUITS. 243 Out of sixty species of the genus Psidium, all those which have been carefully studied are American, It is true that botanists from the sixteenth century have found plants of Fsidium guayava (varieties pomiferum and 2yyrifervb7)%) more or less wild in the Malay Archipelago and the south of Asia/ but everything tends to show that these were the result of recent naturalization. In each locality a foreign origin was admitted; the only doubt was whether this origin was Asiatic or American. Other considerations justify this idea. The common names in Malay are derived from the American word guiava. Ancient Chinese authors do not mention the guava, though Loureiro said a century and a half ago that they were growing wild in Cochin-China. Forster does not mention them among the cultivated plants of the Pacific Isles at the time of Cook's voyage, Avhich is significant when we consider how easy this plant is to cultivate and its ready dispersion. In Mauritius and the Seychelles there is no doubt of their recent intro- duction and naturalization.^ It is more diflicult to discover from what part of America the guava originally came. In the present century it is undoubtedly wild in the West Indies, in Mexico, in Central America, Venezuela, Peru, Guiana, and Brazil.^ But whether this is only since Europeans extended its cultivation, or whether it was previously difi'used by the agency of the natives and of birds, seems to be no more certain than when I spoke on the subject in 1855.* Now, however, with a little more experience in questions of this nature, and since the specific unity of the two varieties of guava is recognized, I shall endeavour to show what seems most probable. J. Acosta,^ one of the earliest authors on the natural history of the new world, expresses himself as follows, about the spherical variety of the guava : " There are * RumpLins, Amhoin, i. p. 141; Rheede, Hortus Malahariensis, iii. t. 3-1. ^ Bojer, Hortus Maiiritianus ; Baker, Flora of Mauritius, p. 112. ' All the floras, and Berg in Flora Braniliensis, vol. xiv. p. 196. ^ Geogr. Bof. Rais., p. 894. ^ Acosta, Hist. Nat. et Morale des Indes Orient, et Occid., French trans., 1598, p. 175. 21)4j origin of cultivated plants. mountains in San DomlnG^o and the other islands entirely covered with gnavas, and the natives say that there were no such trees in the islands before the arrival of the Spaniards, who brought them, I know not whence." The mainland seems, therefore, to have been the original home of the species. Acosta says that it grows in South America, adding that the Peruvian guavas have a white flesh superior to that of the red fruit. This argues an ancient cultivation on the main- land. Hernandez ^ saw both varieties wild in Mexico in the warm regions of the plains and mountains near Quauhnaci. He gives a description and a fair draw- ing of P. jpomiferiim. Piso and Marcgraf ^ also found the two guavas wild in the plains of Brazil ; but they remark that it spreads readily, Marcgraf saj^s that they were believed to be natives of Peru or of North America, by which he may mean the West Indies or Mexico. Evidently the species was wild in a great part of the continent at the time of the discovery of America. If the area was at one time more restricted, it must have been at a far more remote epoch. Different common names were given by the different native races. In Mexico it was xalxocotl ; in Brazil the tree was called araca-iha, the fruit araca guacu ; lastly, the name guajavos, or guajava, is quoted by Acosta and Hernandez for the o-uavas of Peru and San Dominjxo without any precise indication of origin. This diversity of names confirms the hypothesis of a very ancient and extended area. From what ancient travellers say of an origin foreign to San Domingo and Brazil (an assertion, however, which we may be permitted to doubt), I suspect that the most ancient habitation extended from Mexico to Columbia and Peru, possibly including Brazil before the discovery of America, and the West Indies after that event In its earliest state, the species bore spherical, highly coloured fruit, harsh to the taste. The other form is perhaps the result of cultivation. * Hernandez, Kovce Hispanice Thesaurus, p. 85. 2 Piso, Hist, Brasil, p. 74 j Marcgraf, ihid.y p. 105. PLANTS CULTIVATED FOR THEIR FRUITS. 245 Gourd,^ or Calabash — Lagenaria vulgaris, Seringe ; Cucurhita lagenaria, Linnaeus. The fruit of this Curciibitacea has taken different forms in cultivation, but from a general observ-ation of the other parts of the plant, botanists have lanked them in one species which comprises several varieties"^ The most remarkable are the ^^////rij^i's gourd, in the form of a bottle, the long-necked gourd, the trumpet gourd, and the calabash, generally large and without a neck. Other less common varieties have a flattened, very small fruit, like the snuff-box gourd. The species may always be recognized by its white flower, and by the hardness of the outer rind of the fruit, which allows of its use as a vessel for liquids, or a reservoir of air suitable as a buoy for novices in swimmino^. The flesh is sometimes sweet and eatable, sometimes bitter and even purgative. Linn?eus^ pronounced the species to be American. De Candolle^ thought it was probably of Indian origin, and this opinion has since been confirmed. Lagenaria vulgaris has been found wild on the coast of Malabar and in the humid forests of Deyra Doon.^ Roxburgh ^ considered it to be wild in India, although subsequent floras give it only as a cultivated species. Lastly, Rumphius "^ mentions wild plants of it on the sea- shore in one of the Moluccas. Authors generally note that the pulp is bitter in the.se wild plants, but this is sometimes the case in cultivated forms. The Sanskrit language already distinguished the common go\\vd,idavou, and another, bitter, hidou-toumbi, to which Pictet also attributes the name tiJdaka or tiktika.^ Seemann ^ saw * The word gourd is also used in English for Cucurhita inaximo. This is one of the examples of the confusiou in common names and the greater accuracy of scientific terms. * Naudin, Annales des Sc. Nat., 4th series, vol. xii. p, 91 ; Cogniaux, in our Monog. Phan€rog., iii. p. 417. ' Linnceus, Species Planiarum, p. 1434, under Cucurhita. * A. P. de CaudoUe, Flora Franqaise (1805), vol. iii. p. 002. ' Eheede, Ilalahar, iii. pis. 1, 5; Royle, III. Himal., p. 218. « Roxburgh, Fl. hid., edit. 1832, voL iii. p. 719. ^ Rumphius, Amhnin, vol. v. p. 397, t. 144. * Piddington, Jnde.r, at the word Cucurhita lagenaria; Ad. Pictet, Origines Indo-Europ., edit. 3, vol. i. p. 386. ® Seemann, Flora Vitiensis, p. 106, 24^6 ORIGIN OF CULTIVATED PLANTS. the species cultivated and naturalized in the Fiji Isles. Thozet gathered it on the coast of Queensland,^ but it had perhaps spread from neighbouring cultivation. The localities in continental India seem more certain and more numerous than those of the islands to the south of Asia. The species has also been found wild in Abyssinia, in the valley of Hieha by Dillon, and in the bush and stony ground of another district by Schimper.^ From these two regions of the old world it has been introduced into the gardens of all tropical countries and of those temperate ones where there is a sufficiently high temperature in summer. It has occasionally become naturalized from cultivation, as is seen in America.^ The earliest Chinese work which mentioned the gourd is that of Tchong-tchi-chou, of the first century before Christ, quoted in a work of the fifth or sixth century according to Bretschneider.* He is speaking here of cultivated plants. The modern varieties of the gardens at Pekin are the trumpet gourd, which is eatable, and the bottle gourd. Greek authoi's do not mention the plant, but Romans speak of it from the beginning of the empire. It is clearly alhuled to in the often-quoted lines ^ of the tenth book of Columella. After describinq* the different forms of the fruit, he says — " Dabit ilia capacem, Nariciae picis, ant Actaei mellis Hvmetti, Aut habilem lymphis hatnulam, Bacchove lagenam, Turn pueros eadem fluviis innare docebit." Pliny ^speaks of a Cucurhitacea, of which vessels and ' Bentham, Flora Australiensis, iii. p. 316. ^ Described first under the name Lagenaria idolatrica. A. Eichard, Tentamen Ft. Ahyss., i. p. 293, and later, Naudin and Cogniaux, recognized its identity with L. vulgaris. ^ Tori-ey and Gray, Fl. of N. Amer., i. p. 543 ; Grisebach, Flora of Brit. W. hid. Is., p. 288. * Bretschneider, letter of the 23rd of i\ugust, 1881. * Tragus, Stirp., p. 285 j Ruellius, De Natura Siirpium, p. 498; Nau- din, ihid. '' Pliny, Hist. Plant., 1. 19, c. 5. PLANTS CULTIVATED FOR THEIR FRUITS. 247 flasks for wine were made, which can only apply to this species. It does not appear that the Arabs were early ac- quainted with it, for Ibn Ala warn and Ibn Baithar say nothing of it.-"- Commentators of Hebrew works attri- bute no name to this species with certainty, and yet the climate of Palestine is such as to popularize the use of gourds had they been known. From this it seems to me doubtful that the ancient Egyptians possessed this plant, in spite of a single figure of leaves observed on a tomb which has been sometimes identified with it.^ Alexander Braun, Ascherson, and Magnus, in their learned paper on the Egyptian remains of plants in the Berlin Museum,^ indicate several Cucurbitacese without mentioning this one. The earliest modern travellers, such as Rauwolf,^ in 1574, saw it in the gardens of Syria, and the so-called pilgrim's gourd, figured in 1539 by Brunfels, was probably knoAvn in the Holy Land from the Middle Ages. All the botanists of the sixteenth century give illus- trations of this species, which was more generally culti- vated in Europe at that time than it is now The common name in these older writings is Camieraria, and three kinds of fruit are distinguished. From the white colour of the flower, which is always mentioned, there can be no doubt of the species. I also note an illustration, certainly a very indifferent one, in which the flower is wanting, but with an exact representation of the fruit of the pilgrim's gourd, Avhich has the great interest of having appeared before the discovery of America. It is pi. 216 of Ilerbarius Fatavice hnpvessvis, in 4to, 1485 — a rare work. In spite of the use of similar names by some authors, I do not believe that the gourd existed in America be- fore the ai'rival of the Europeans. The Taqwera of Piso ^ ^ Ibn Alawam, in E. Meyer, Geschichte der Botanik, iii. p. GO ; Ibn Baithar, Sondtheimer's translation. ^ Unger, Pjlanzen des Alien JEgyptens, p. 59 ; Pickering, Chronol. Arrang., p. 137. ' In 8vo, 1877, p. 17. ■• Rauwolf, Fl. Orient.^ p. 125. * Piso, Indice Utriusque.t etc., edit. 1658, p. 2G4. 248 ORIGIN OF CULTIVATED PLANTS. and Cucurhita lagencBforma of Marcgraf^ are per- haps Lagenaria vulgaris as monographs say,^ and the specimens from Brazil which they mention should be certain, but that does not prove that the species was in the country before the voyage of Amerigo Vespucci in 1504<. From that time until the voyages of these two botanists in 1G37 and 1638, a much longer time elapsed than is needed to account for the introduction and dif- fusion of an annual species of a curious form, easy of cultivation, and of which the seeds long retain the faculty of germination. It may have become naturalized from cultivation, as has taken place elsewhere. It is still more likely that Citcurhita siceratia, Molina, attributed sometimes to the species under consideration, sometimes to Citcurhita onaxmia,^ ii^ay have been introduced into Chili between 1588, the date of the discovery of that country, and 1787, the date of the Italian edition of Molina. Acosta^ also speaks of calabashes which the Peruvians used as cups and vases, but the Spanish edition of his book appeared in 1591, more than a hundred years after the Conquest. Among the first naturalists to mention the species after the discovery of America (1492) is Oviedo,^ who had visited the main- land, and, after dwelling at Vera Paz, came back to Europe in 1515, but returned to Nicaragua in 1539.^ According to Ramusio's compilation '' he s^^oke of zueche, freely cultivated in the West India Islands and Nicaragua at the time of the discovery of America, and used as bottles. The authors of the floras of Jamaica in the seventeenth century say that the species was cultivated in that island. P. Brown,^ however, mentions a large cultivated gourd, and a smaller one with a bitter and purgative pulp, which was found wild. » Marcsjraf, Hist. Nat. BradUce, 161-8, p. 44. "^ Nandin, Hid.; Cognian.v, f'/ora BrasiL, fasc. 78, p. 7; and de Candollc, Monogr. Phaner.y iii. p. 418. 3 CI. Gay, Flora Chilena, ii. p. 403. * Jos. Acosta, French trans., p. 167. * Pickering, Chronol. Arrang., p. SGI. '' Pickering, ihid. ' T^iimusio, voL iii. p. 112. * P. Brown, Jamaica^ edit. ii. p. 351. PLANTS CULTIVATED FOR THEIR FRUITS. 249 Lastly, Elliott ^ writes as follows, in 1824, in a work on the Southern States of America: "X. vulgaris is rarely found in the woods, and is certainly not indigenous. It seems to have been brought by the early inhabitants of our country from a warmer climate. The species has now become wild near dwellings, especially in islands." The expression, "inhabitants of our country," seems to refer rather to the colonists than to the natives. Between the discovery of Virginia by Cabot in 1497, or the travels of Raleigh in 1584, and the floras of modern botanists, more than two centuries elapsed, and the natives would have had time to extend the cultivation of the species if they had received it from Europeans. But the fact of its cultivation by Indians at the time of the earliest deal- ings with them is doubtful. Torrey and Gray^ mentioned it as certain in their flora published in 1830-40, and later the second of these able botanists,^ in an article on the Cucurhitacece known to the natives, does not mention the calabash, or Lagenaria. I remark the same omission in another special article on the same subject, published more recently.* [In the learned articles by Messrs. Asa Gray and Trumbull on the present volume {Ar)ieTican Journal of Science, 1883, p. 370), they give reasons for supposing the species known and indigenous in America previous to the arrival of the Europeans. Early travellers are quoted more in detail than I had done. From their testimony it apjiears that the inhabitants of Peru, Brazil, and of Paria possessed gourds, in Spanish calahazas, but I do not see that this proves that this was the species called by botanists Citcwrhita lagonaria. The only character in- dependent of the exceedingly variable form of the fruit is the white colour of the flowers, and this character is not mentioned. — Author's Note, 1884.] Gourd — Cucurhita maxima, Duchesne. In enumerating the species of the genus Cucur-hita, I ' Elliott, Sketch of the Botany of South Carolina and Georgia, ii. p. 663. ^ Torrey and Gray, Flora of N. America, i. p. 544. ^ Asa Gray, in the American Journal of Science, 1857, yol. xxiv p. 442. * Trumbull, in Biill. Torrey Bot. Club, vol. vi. p. 69. 250 ORIGIN OF CULTIVATED PLANTS. should explain that their distinction, formerly exceedingly difficult, has been established by M. Naudin ^ in a very scientific manner, bv means of an assiduous cultivation of varieties and of experiments upon their crossed fertiliza- tion. Those groups of forms which cannot fertilize each other, or of which the product is not fertile and stable, are regarded by him as species, and the forms which can be crossed and yield a fertile and varied product, as races, breeds, or varieties. Later experiments ^ showed him that the establishment of species on this basis is not without exceptions, but in the genus Cueurhita physio- loofical facts SL^ree Avith exterior differences. M. Naudin has established the true distinctive characters of C. Tnaxhiia and C. Pepo. The leaves of the first have rounded lobes, the peduncles are smooth and the lobes of the corolla are curved outwards ; the second has leaves with pointed lobes, the peduncles marked with ridges and furrows, the corolla narrowed towards the base and with lobes nearly always upright. The principal varieties of Cueurhita Tnaxima are the great yellow gourd, which sometimes attains to an enormous size,^ the Spanish gourd, the turban gourd, etc. Since common names and those in ancient authors do not agree with botanical definitions, we must mistrust the assertions formerly put forth on the origin and earl}'' cultivation of such and such a gourd at a given epoch in a given country. For this reason, when I considered the subject in 1855, the home of these plants seemed to me either unknown or very doubtful. At the present day it is more easy to investigate the question. According to Sir Joseph Hooker,^ Cueurhita maxima was found by Barter on the banks of the Niger in Guinea, apparently indigenous, and by Welwitsch in Angola without any assertion of its Avild character. In works on Abyssinia, Egypt, or other African countries in which the species is commonly cultivated, I find no * Naudin, Ann. 8c. Nat., 4th series, vol. vi. p. 5 ; vol. xii. p. 84. * Ihid.^ 4th series, vol. xviii. p. 160 ; vol. xix. p. 180. ^ As much as 200 lbs., according to the Bon Jardiniere 1850, p. 180. * Hooker, Fl. of Trop. Afr.y ii. p. 555. PLANTS CULTIVATED FOR THEIR FRUITS. 251 indication that it is found wild. The Abyssinians used the word duhha, which is applied in Arabic to gourds in general. The plant w^as long supposed to be of Indian origin, because of such names as Indian gourd, given by sixteenth- century botanists, and in particular the Pe2:)o onaximus indie as, figured b}^ Lobel,^ which answers to the modern species ; but this is a very insufficient proof, since popu- lar indications of origin are very often erroneous. The fact is that though pumpkins are cultivated in Southern Asia, as in other parts of the tropics, the plant has not been found wild.^ No similar species is indicated by ancient Chinese authors, and the modern names of gourds and pumpkins now grown in China are of foreign and southern origin.^ It is impossible to know to what species the Sanskrit name kurkarou belonged, although Roxburgh attributes it to Cucurhita Pepo ; and there is no less uncertainty w^ith respect to the gourds, pump- kins, and melons cultivated bv the Greeks and Romans. It is not certain if the species was known to the ancient Egyptians, but perhaps it was cultivated in that country and in the Gr?eco-Roman world. The Pepones, of which Charlemagne commanded the cultivation in his farms,* were perhaps some kind of pumpkin or marrow, but no figure or description of these plants which may be clearly recognized exists earlier than the sixteenth century. This tends to show its American origin. Its existence in Africa in a wild state is certainly an argument to the contrary, for the species of the family of Cucurhitacece are very local ; but there are arguments in favour of America, and I must examine them with the more care since I have been reproached in the United States for not having given them sufficient weight. In the first place, out of the ten known species of the genus Cucurhita, six are certainly wild in America ^ Lobel, IconeSf t. 641. The illustration is reproduced in Dalechamp's Hist., i. p. 626. » Clarke, Hooker's Fl. Brit. Ind., ii. p. 622. * Bretschneider, letter of Aug. 23, 1881. * The list is given by E. Meyer, Geschichte du Botanih, in. p. 401 . The Cucurbita of which be speaks must have been the gourd, Lagenaria. 12 252 ORIGIN OF CULTIVATED PLANTS. (Mexico and California) ; but these are perennial species' while the cultivated pumpkins are annuals. The plant called jurur}iic by the Brazilians, figured by Piso and Marcgraf ^ is attributed by modern writers to Cuciivhita maxima. The drawing and the short account by the two authors agree pretty well with this theory, but it seems to have been a cultivated plant. It may have been brought from Europe or from Africa by Europeans, between the discovery of Brazil in loO-i, and the travels of the above-named authors in 1637 and 1638. No one has found the species wild in North or South America, I cannot find in works on Brazil, Guiana, or the West Indies anv si^jn of an ancient cultivation or of wild growth, either from names, or from traditions or more or less distinct belief In the United States those men of science who best know the lano-uaofes and customs of the natives, Dr. Harris for instance, and more recently Trumbull,^ maintain that the Ciiciirhitaceai called 8(/ wo sA by the Anglo-Americans, and macoch, or cashaiu, cushaw, by early travel] eis in Viiginia, are ijwnphins. Trumbull says that squash is an Indian word. I have no reason to doubt the assertion, but neither the ablest linguists, nor the travellers of the seventeenth century, who saw the natives provided with fruits which they called gourds and pumpkins, have been able to prove that they were such and such species recognized as distinct by modern botanists. All that Ave learn from this is that the natives a century after the discovery of Virginia, and twenty to forty years after its colonization by Sir Walter Raleigh, made use of some fruits of the Cucurhitacece. The com- mon names are still so confused in the United States, that Dr. Asa Gray, in 1868, gives pumpkin and squash as answering to difierent species of Cucurhita^ while Darlington* attributes the r\^m.e pumpkin to the common C ucitrh ita Pepo, 3iU.d theit of squash to the varieties of the > Piso, Brazil, edit. 1658, p. 264; Marcgraf, edit. 164.S, p. 44. - Harris, American Journal, 1857, vol. xxiv. p. 4iL; Trumbull, Bull. of Torrey Bot. Chih, 1876, vol. vi. p. 69. ' Asa Gray, Botany of the Northern States, edit. 1868, p 186. * Darliugton, Flora Cestrica, 1853, p. 94. PLANTS CULTIVATED FOR THEIR FRUITS. 253 latter whicli correspond to the forms of Melopepo of early botanists. They attribute no distinct common name to Cucurhita nnaxima. Finally, without placing implicit faith in the indi- genous character of the plant on the banks of the Niger, based upon the assertion of a single traveller, I still believe that the species is a native of the old world, and introduced into America by Europeans. [The testimony of early travellers touching the ex- istence of Cucuvhitct "niaxima in America before the arrival of Europeans has been collected and supplemented by Messrs. Asa Gray and Trumbull {Amierican Journal of Science, 1883, p. 372). They confirm the fact ah^eady known, that the natives cultivated species of Cucurhita under American names, of which some remain in the modern idiom of the United States. None of these early travellers has noted the botanical characters by which Naudin established the distinction between G. maxima and G. Pepo, and consequently it is still doubtful to which species they referred. For various reasons I had already admitted that C. Pepo was of American origin, but I retain my doubts about G. inaxima. After a more attentive perusal of Tragus and Matthiolo than I had bestowed upon them, Asa Gray and Trumbull notice that they call Indian whatever came from America. But if these two botanists did not confound the East and West Indies, several others, and the public in general, did make this confusion, which occasioned errors touching the origin of species which botanists were liable to repeat. A further indication in favour of the American oricrin of G. mccxima is communicated by M. Wittmack, who in- forms me that seeds, certified by M. Naudin to belong to this species, have been found in the tombs of Ancon. This would be conclusive if the date of the latest burials at Ancon were certain. See on this head the article on PhaseoliLs vulgaris. — Author's Note, 1884.] Pumpkin — Gucurhita Pepo and G. Melopepo, Linn?eus. Modern authors include under the head of Gucurhita Pepo most of the varieties whicli Linn?eus designated by this name, and also those which he called G. Melopepo. 254 ORIGIN OF CULTIVATED PLANTS. These varieties are very different as to the shape of the fruit, which shows a very ancient cultivation. There is the Patagonian pumpkin, with enormous cylindrical fruit ; the sugared fiimphin, called Brazilian; the vegetable marrow, with smaller long-shaped fruit ; the Barberine, with knobby fruit; the Electors hat, with a curiously shaped conical fruit, etc. No value should be attached to the local names in this designation of varieties, for we have often seen that they express as many errors as varieties. The botanical names attributed to the species by Naudin and Cogniaux are numerous, on account of the bad habit which existed not long ago of describing as species purely garden varieties, without taking into account the wonderful effects of cultivation and selection upon the organ for the sake of which the plant is cultivated. Most of these varieties exist in the gardens of the warm and temperate regions of both hemispheres. The origin of the species is considered to be doubtful. I hesitated m 1855 ^ between Southern Asia and the Mediterranean basin. Naudin and Coo-niaux^ admit Southern Asia as probable, and the botanists of the United States on their side have given reasons for their belief in an American origin. The question requires careful investio^ation. I shall first seek for those forms now attributed to the species which have been found growing anywhere m a wild state. The variety Cucitrhita ovifera, Linnaeus, was formerly gathered by Lerche, near Astrakhan, but no modem botanist has confirmed this fact, and it is probable it was a cultivated plant. Moreover, Linnaeus does not assert it was wdld. I have consulted all the Asiatic and African floras without findincj the sliixhtest mention of a wild variety. From Arabia, or even from the coast of Guinea to Japan, the species, or the varieties attributed to it, are always said to be cultivated. In * Geogr. Bot. Raisonnee, p. 902. ' Naudin, Ann. Sc. Nat., 3rd series, vol. vi. p. 9 j Cogniaux, in de Candolle, Monogr. Phandr., in. p. 54^6. PLANTS CULTIVATED FOR THEIR FRUITS. 255 India, Roxburgh remarked this, and certainly Clarke, in his recent flora of British India, has good reasons for indicating no locality for it outside cultivation. It is otherwise in America, A variety, G. texana,^ very near to the variety ovata, according to Asa Gray, and which is now unhesitatingly attributed to G. Pepo, was found by Lindheimer " on the edges of thickets, in damp woods, on the banks of the upper Guadaloupe, apparently an indigenous plant." Asa Gray adds, how- ever, that it is perhaps the result of naturalization. However, as several species of the genus Gucarbita grow wild in Mexico and in the south-west of the United States, we are naturally led to consider the collector's opinion sound. It does not appear that other botanists found this plant in Mexico, or in the United States. It is not mentioned in Hemsley's Biologia Gentrali- Americana, nor in Asa Gray's recent flora of Cali- fornia. Some synonyms or specimens from South America, attributed to G. Fejpo, appear to me very doubtful. It is impossible to say what Molina ^ meant by the names G. Siceratia and G. mammeata, which appear, moreover, to have been cultivated plants. Two species briefly described in the account of the journey of Spix and Martins (ii. p. 53G), and also attributed to G. Fepo,^ are mentioned among cultivated plants on the banks of the Rio Francisco. Lastly, the specimen of Spruce, 2716, from the river Uaupes, a tributary of the Rio Negro, which Cogniaux^ does not mention having seen, and which he first attributed to the G. Fepo, and afterwards to the C. nioscliata, was per- haps cultivated or naturalized from cultivation, or by transport, in spite of the paucity of inhabitants in this country. Botanical indications are, therefore, in favour of a Mexican or Texan origin. It remains to be seen if * Asia Gray, Plantce Lindlieimeriancc, parfc ii. p. 193. " Molina, Hist. Nat. du Chili, p. 377. * Cogniaux, in Monogr. PhanSr. and Flora Brasil, fasc. 78, p. 21. * Cogniaux, Fl. Bras, and Mo7wgr. Fhandr., iii., p. 517. 256 ORIGIN OF CULTIVATED PLANTS. historical records are in agreement with or contrary to this idea. It is impossible to discover whether a given Sanskrit, Greek, or Latin name for the pumpkin belongs to one species rather than to another. The form of the fruit is often the same, and the distinctive characters are never mentioned by authors. There is no figure of the pumpkin in the Herhavius Patavicti hnpvessus of 1485, before the discovery of America, but sixteenth-century authors have published plates which may be attributed to it. There are three forms of Pepones figured on page 406 of Dodoens, edition 1557. A fourth, Pepo rotiindus major, added in the edition of 1616, appears to me to be C. onaxwia. In the drawing of Pejyo ohlongus of Lobel, I cones, 641, the character of the peduncle is clearly defined. The names given to these plants imply a foreign origin ; but the authors could make no assertions on this head, all the more that the name of '' the Indies " applied both to Southern Asia and America. Thus historical data do not gainsay the opinion of an American origin, but neither do they adduce anything in support of it. If the belief that it grows wild in America is con- firmed, it may be confidently asserted that the pumpkins cultivated by the Romans and in the Middle Ages were Ciicurhita onaxima, and those of the natives of North America, seen by different travellers in the seventeenth century, were Ciicurhita Pepo. Musk, or Melon Pumpkin — Cucurhita moschata, Duchesne. Tlie Bon Jarclinier quotes as the principal varieties of this species pumpkin niusccide cle Provence, pleine cle Naples, and cle Darharie. It is needless to say that these names show nothing as to origin. The species is easily recognized by its fine soft down, the pentagonal peduncle which supports the fruit broadening at the summit ; the fruit is more or less covered with a olaucous efflorescence, and the flesh is somewhat musk-scented. The lobes of the calyx are often terminated by a leafy PLANTS CULTIVATED FOR THEIR FRUITS. 257 border.^ Cultivated m all tropical countries, it is less successful than other pumpkins in temperate regions. Cogniaux ^ suspects that it comes from the south of Asia, but he gives no proof of this. I have searched through the floras of the old and new worlds, and I have nowhere been able to discover the mention of the species in a truly wild state. The indications which approach most nearly to it are : (1) In Asia, in the island of Bangka, a specimen verified by Cogniaux, and which Miquel '^ says is not cultivated ; (2) in Africa, in Angola, specimens which Welwitsch says are quite wild, but " probably due to an introduction ; " (3) in America, five specimens from Brazil, Guiana, or Nicaragua, mentioned by Cogniaux, without knowing whether they were cultivated, naturalized, or indigenous. These indications are very slight Rumphius, Blume, Clarke (Flora of British India) in Asia, Schweinfurth (Oliver's Flora of Trop. Africa) in Africa, only know it as a cultivated plant. Its cultivation is recent in China,* and American floras rarely mention the sp^icies. No Sanskrit name is known, and the Indian, Mala.y, and Chinese names are neither very numerous nor very original, although the cultivation of the plant seems to be more diffused in Southern Asia than in other parts of the tropics. It was already grown in the seventeenth century according to the Hortus Mala- haricuSy in which there is a good plate (vol. viii. pi. 2). It does not appear that this species was known in the sixteenth century, for Dalechamp's illustration (Hist., i. p. 616) which Seringe attributed to it has not its true cha- racters, and I can find no other figure which resembles it. Fig-leaved Pumpkin — Cucurhita ficifolia, Bouche ; Ciicurbita melanosperma, Braun. About thirty years ago this pumpkin with black or brown seeds was introduced into gardens. It differs * See the excellent plate in Wight's Icones, t. 507, tinder the erroneous name of Cucurhita maxima. * Cogniaux, in Monogr. Phaner., iii. p. 547. ' Miquel, Sumatra, under the name Gymnopetalum, p. 332. * Cogriaux, hi Monogr. Phaner. 258 ORIGIN OF CULTIVATED PLANTS. from other cultivated species in being perennial. It is sometimes called the Siamese melon. The Bon Jardinier says that it comes from China. Dr. Bretschneider does not mention it in his letter of 1 SSI, in which he enu- merates the pumpkins grown by the Chinese. Hitherto no botanist has found it wdld. I very much doubt its Asiatic origin as all the kno"wn perennial species of Cucurhita are from Mexico or California. Melon — Giicumis Melo, Linnaeus. The aspect of the question as to the origin of the melon has completely changed since the experiments of Naudin. The paper which he published in 1859, in the Annales cles Sciences Naturelles, 4th series, vol. ii., on the genus Cucuriiis, is as remarkable as that on the genus Cucurhita. He gives an account of the observations and experiments of several years on the variability of forms and the crossed fecundation of a multitude of species, breeds, or varieties coming from all parts of the world. I have already spoken (p. 250) of the physiological principle on which he believes it possible to distinguish those groups of forms which he terms species, although certain excep- tions have occurred wdiich render the criterion of fertili- zation less absolute. In spite of these excej^tional cases, it is evident that if nearly allied forms can be easily crossed and produce fertile individuals, as we see, for example, in the human species, they must be considered as constituting a single species. In this sense Citciiinis Melo, according to the ex- periments and observations made by Naudin upon about two thousand living plants, constitutes a species which comprehends an extraordinary number of varieties and even of breeds ; that is to say, forms w^hich are pre- served by heredity. These varieties or races can be ferti- lized by each other, and yield varied and variable products. They are classed by the author into ten groups, which he calls cantelowps, melons hrodes, sucrins, melons d'hiver, serpents y forme de concornbre, Chito, Dudaim, rouges de Perse, and sauvages, each containing varieties or nearly allied races. These have been named in twenty-five or thirty different ways by botanists, who, without noticing PLANTS CULTIVATED FOR THEIR FRUITS. 259 transitions of form, the faculty of crossing or of change under cultivation, have distinguished as species all the varieties which occur in a given time or place. Hence it results that several forms found wild, and which have been described as species, must be the types and sources of the cultivated forms ; and Naudin makes the very just observation that these wild forms, which differ more or less the one from the other, may have pro- duced different cultivated varieties. This is the more probable that they sometimes inhabit countries remote from each other as Southern Asia and tropical Africa, so that differences in climate and isolation may have created and consolidated varieties. The following are the forms which Naudin enume- rates as wild : 1. Those of India, which are named by Wildenow Cucumis puhescens, and by Roxburgh C. tur- hinatus or C. Tiiaderas-patanus, The whole of British India and Beluchistan is their natural area. Its natural wildness is evident even to non-botanical travellers.^ The fruit varies from the size of a plum to that of a lemon. It is either striped or barred, or all one colour, scented or odourless. The flesh is sweet, insipid, or slightly acid, differences which it has in common w^ith the cultivated Cantelopes. According to Roxburgh the Indians gather and have a taste for the fruits of G. tur- hinatus and of G. viaderas-patanus, though they do not cultivate it. Referring to the most recent flora of British India, in which Clarke has described the Gucurhitacece (ii. p. 619), it seems that this author does not agree with M. Naudin about the Indian wild forms, although both have examined the numerous specimens in the herbarium at Kew. The difference of opinion, more apparent than real, arises from the fact that the EnHish author attributes to a nearly and certainly wild allied species, C. trigomis, Roxburgh, the varieties which Naudin classes under G. Melo. Cogniaux,^ who afterwards saw the same speci- ' Gardener's Chronicle, articles signed *' I. II. II.," 1857, p. 153 j 1858, p. 130. * Cogniaux, Monogr. Phane'r., iii. p. 485. 2G0 ORIGIN OF CULTIVATED PLANTS. mens, attributes only G. tnrhinatus to trigonus. The specific difi'erence between C. Melo and C. trigonus is unfortunately obscure, from the characters given by these three authors. The principal difference is that G. Melo is an annual, the othar perennial, but this dura- tion does not appear to be very constant. Mr. Clarke says himself that G. Melo is perhaps derived by cultiva- tion from G. trigonus ; that is to say, according to him, from the forms which Naudin attributes to G. Melo. The experiments made during three consecutive years by Naudin^ upon the products of Givciiinis trigonus, fertilized by G. Melo, seem in favour of the opinion which admits a specific diversity ; for if fertilization took place the products were of different forms, and often reverted to one or other of the original parents. 2. The African forms. Naudin had no specimens in sufficiently good condition, or of which the wild state was sufficiently certain to assert positively the habitation of the species in Africa. He admits it with hesitation. He includes in the species cultivated forms, or other wild ones, of which he had not seen the fruit. Sir Joseph Hooker^ subsequently obtained specimens which prove more. I am not speaking of those from the Nile Valley,^ which are probably cultivated, but of plants gathered by Barter in Guinea in the sands on the banks of the Niger. Thonning * had previously found, in sandy soil in Guinea, a Gacumis to which he had given the name arenaHus ; and Cogniaux,^ after having seen a specimen brought home by this traveller, had classed it with G. Melo, as Sir J. Hooker thouoht. The nej^roes eat the fruit of the plant found by Barter. The smell is that of a fresh green melon. In Thonning's plant the fruit is ovoid, the size of a plum. Thus in Africa as in India the species bears small fruit in a wild state, as we might expect. The Dudaim amons: cultivated varieties is allied to it. * Naudin, Ann. Sc. Nat., 4th series, vol. xviii. p. 171, * Hookei*, in Oliver, Fl. of Trop. Afr., ii. p. 516. * Schweinfurth and Ascherson, Aufzdhlung, p. 267. * Schumacher and Thonning, Guineiske Planten., p. 423. * Cogniaux, in de Candolle, Monogr. Phaner., p. 483. PLANTS CULTIVATED FOR THEIR FRUITS. 261 The majority of the species of the genus Cucumis are found in Africa; a small minority in Asia or in America. Other species of Cucurhitacew are divided between Asia and America, although as a rule, in this family, the areas of species are continuous and restricted. Cu- cuviis Melo was once perhaps, like Citrullus Colocynthis of the same family, wild from the west coast of Africa as far as India without any break. I formerly hesitated to admit that the melon was indiofenous in the north of the Caucasus, as it is asserted by ancient authors — an assertion which has not been confirmed by subsequent botanists. Hohenacker, who was said to have found the species near Elisabethpolis, makes no mention of it in his paper upon the province of Talysch. M. Boissier does not include Cucii'tnis Melo in his Oriental flora. He merely says that it is easily naturalized on rubbish-heaps and waste ground. The same thing has been observed elsewhere, for instance in the sands of Ussuri, in Eastern Asia. This would be a reason for mistrusting the locality of the sands of the Niofer, if the small size of the fruit in this case did not recall the wild forms of India. The culture of the melon, or of different varieties of the melon, may have begun separately in India and Africa. Its introduction into China appears to date only from the eighth century of our era, judging from the epoch of the first work which mentions it.^ As the relations of the Chinese with Bactriana, and the north-west of India by the embassy of Chang-kien, date from the second century, it is possible that the culture of the species was not then widely diff*used in Asia. The small size of the wild fruit offered little inducement. No Sanskrit name is known, but there is a Tamul name, probably less ancient, molmn,^ wdiich is like the Latin Melo. It is not proved that the ancient Egyptians cultivated the melon. The fruit figured by Lepsius ^ is not recog- nizable. If the cultivation had been customary and * Bretschneider, letter of Au^. 26, 18S1. ' Piddington, Index. ** See tlie copy in Unger's Pfanzen des Alien 2Ej]jptens, fig. 25. 262 ORIGIN OF CULTIVATED PLANTS. ancient in that country, the Greeks and Romans would have early known it. Now, it is doubtful whether the Sikua of Hippocrates and Theophrastus, or the Pepon of Dioscorides, or the Melopepo of Pliny, was the melon. The passages referring to it are brief and insignificant ; Galen ^ is less obscure, when he says that the inside of the Melopepones is eaten, but not of the Pepones. There has been much discussion about those names,^ but we want facts more than words. The best proof w^hich I have been able to discover of the existence of the melon among the Romans is a very accurate representation of a fruit in the beautiful mosaic of fruits in the Vatican. Moreover, Dr. Comes certifies that the half of a melon is represented in a painting at Herculaneum.^ The species was probably introduced into the Gr^eco-Roman world at the time of the Empire, in the beginning of the Christian era. It was probably of indifferent quality, to judge from the silence or the faint praise of writers in a country where gourmets were not w^anting. Since the Renaissance, an improved cultivation and relations with the East have introduced better varieties into our gardens. We know, however, that they often degenerate either from cold or bad conditions of soil, or by crossing with inferior varieties of the species. Water-Melon — Citrullus vulgaris, Schrader; Cucur- hita Citrullus, Linnaeus. The oris^in of the water-melon was long mistaken or unknown. According to Linn?eus, it was a native of Southern Italy.* This assertion was taken from Matthiole, without observing that this author says it was a cultivated species. Seringe,^ in 1828, supposed it came from India and Africa, but he gives no proof. I believed it came from Southern Asia, because of its * Galen, De AHwentis, 1. 2, c. 5. * See all the Vei'gilian floras, and Nautlin, Ann. Sc. Kat., 4tli series, vol. xii. p. 111. ' Comes, III. Piarde nei Dipinti Pompeiani, in 4to, p. 20, in the Museo Nation., vol. iii. pi. 4. * Habitat in Apulia, Calabria, Sicilia (Linnaeus, Species, edit. 1763, p. 1435). * Seringe, in Prodromus, iii. p. 301, PLANTS CULTIVATED FOR THEIR FRUITS. 203 very general cultivation in this region. It was not known in a wild state. At length it was found indi- genous in tropical Africa, on both sides of the equator, which settles the question.^ Livingstone^ saw districts literally covered with it, and the savages and several kinds of wild animals eagerly devoured the wild fruit. They are sometimes, but not always, bitter, and this cannot be detected from the appearance of the fruit. The negroes strike it with an axe, and taste the juice to see whether it is good or bad. This diversity in the wild plant, growing in the same climate and in the same soil, is calculated to show the small value of such a character in cultivated Cucurhitacew. For the rest, the frequent bitterness of the water-melon is not at all extraordinary, as the most nearly allied species is Citrullus Colocynthis. Naudin obtained fertile hybrids from crossing the bitter water-melon, wild at the Cape, with a cultivated species winch confirms the specific unity suggested by the outward appearance. The species has not been found wild in Asia. The ancient Egyptians cultivated the water-melon, which is represented in their paintings.^ This is one reason for believing that the Israelites knew the species, and called it ahhatitchmi, as is said ; but besides the Arabic name, hattich, hatteca, evidently derived from the Hebrew, is the modern name for the water-melon. The French noxa.^.jjasteque, comes through the Arabic from the Hebrew. A proof of the antiquity of the plant in the north of Africa is found in the Berber name, tadeladt,^ which differs too widely from the Arabic name not to have existed before the Conquest. The Spanish names zan- dria, cindria, and the Sardinian smc^ria,^ which I cannot connect with any others, show also an ancient culture in the eastern part of the Mediterranean basin. Its '' Nanditi, Ann. s^c. Nat., 4tli series, vol. xii. p. 101 ; Sir J. Hooker, in Oliver, Flora of Trap. Afr., ii, p. 549. ^ French trans., p. 56. ' Uager has copied the figures from Lepsius' work in his memoir, Die Pjiir.zen des Alten yEgyptetis, figs. 30, 31, 32. * Dutionnaire Franrais-Berher, at the vford pasteque. * Mtris, Flora Sardoa. 264; ORIGIN OF CULTIVATED PLANTS. cultivation early spread into Asia, for there is a Sanskrit name, chai/ajnda,^ but the Chinese only received the plant in the tenth century of the Christian era. They call it si-kua, that is melon of the West.^ As the water-melon is an annual, it ripens out of the tropics wherever the summer is sufficiently hot. The modern Greeks cultivate it largely, and call it carpoiisia or carpousea,^ but this name does not occur in ancient authors, nor even in the Greek of the decadence and of the Middle Ages.* It is the same as the karpiis of the Turks of Constantinople,^ which we find again in the Russian arhas,^ and in Bengali and Hindustani as tarhiij twrhoiiz? Another Constantinople name, mentioned by Forskal, chimonico, recurs in Albanian chiraico^ The absence of an ancient Greek name which can wuth certainty be attributed to this species, seems to show that it was introduced into the Grseco-Roman world about the beginning of the Christian era. The poem Copa, attributed to Virgil and Pliny, perhaps mentions it (lib. 19, cap. 5), as Naudin thinks, but it is doubtful. Europeans have introduced the water-melon into America, where it is now cultivated from Chili to the United States. The jace of the Brazilians, of which Piso and Marcgraf have a drawing, is evidently in- troduced, for the first-named author says it is cultivated and partly naturalized.^ Cucumber — Gacumis sativus, Linn?eus. In spite of the very evident difference between the melon and cucumber, which both belong to the genus Cvbciimis, cultivators suppose that the species may be crossed, and that the quality of the melon is thus some- * Piddington, Index. 2 Bretsclmeider, Study and Value, etc., p. 17. 3 Heldreich, Pflanz, d. Attisch. Ebene., p. 501 ; Nutzpf,. Griechenl., p. 50. * Langkavel, Bot. der Spat. Griechen. * Forskal, Flora ^lEjypto-Arahica., part i. p. 34. * Nemriich, Pohjg. Lexic, i. p. 1309. ' Piddington, Index; Pickering, Chronol. Arrang., p, 72. ^ Heldreich, Nutzpji., etc., p. 50. * " Sativa phnita et tractu temporia quasi nativa factx*' (Piso, edit. 1658, p. 2dS). PLANTS CULTIVATED FOK THEIR FRUITS. 205 times spoilt. Naudin ^ ascertained by experiments that this fertilization is not possible, and has also shown that the distinction of the two species is well founded. The original country of Cticiiniis scttivus was un- known to Linnreus and Lamarck. In 1805, Wildenow^ asserted it was indigenous in Tartary and India, but without furnishing any proof Later botanists have not confirmed the assertion. When I went into the question in 1855, the species had not been anywhere found wild. For various reasons deduced from its ancient culture in Asia and in Europe, and especially from the existence of a Sanskrit name, soitkasa,^ I said, " Its original habitat is probabl}^ the north-west of India, for instance Cabul, or some adjacent country. Everything seems to show that it will one day be discovered in these regions which are as yet but little known." This conjecture has been realized if we admit, with the best-informed modern authors, that Cucuviis Hard- tvickii, Royle, possesses the characteristics of Ckccumis sativus. A coloured illustration of this cucumber found at the foot of the Himalayas may be seen in Royle's Illustrations of Himalayan Plants, p. 220, pi. 47. The stems, leaves, and flowers are exactly those of C. sativus. The fruit, smooth and elliptical, has a bitter taste; but there are similar forms of the cultivated cucumber, and we know that in other species of the same family, the water-melon, for instance, the pulp is svv^eet or bitter. Sir Joseph Hooker, after describing the remarkable variety which he calls the Sikkim cucumber,* adds that the variety Hardivickii, wild from Kumaon to Sikkim, and of which he has gathered specimens, does not differ more from the cultivated plant than certain varieties of the latter differ from others ; and Cogniaux, after seeing the plants in the herbarium at Kew, adopts this opinion.^ The cucumber, cultivated in India for at least three ^ Naudin, in Ann. Sc. Nat., 4th series, vol. xi. p. 31. ^ Wildenow, Species, iv. p. 615. * Piddiugton, Index, 4 Bot. Mag., pi. 6206. ^ Cogniaux, in de Candolle, Monogr. Phancr., ill. p. 499. 2G6 ORIGIN OF CULTIVATED PLANTS thousand 3^ears, was only introduced into China in tlie second century before Christ, when the ambassador Chang-kien returned from Bactriana.^ The species spread more rapidly towards the West. The ancient Greeks cultivated the cucumber under the name of sikuos,^ which remains as sikua in the modern language. The modern Greeks have also the name aggowria, from an ancient Aryan root which is sometimes applied to the water-melon, and which recurs for the cucumber in the Bohemian agiirka, the German Giirke, etc. The Albanians (Pelasgians ?) have quite a different name, kratsavets.^ which we recognize in the Slav Krastavak. The Latins called the cucumber cucwrtiis. These different names show the antiquity of the species in Europe. There is even an Esthonian name, uggurits, ukkurits, iirits} It does not seem to be Finnish, but to belonor to the same Aryan root as aggoiiria. If the cucumber came into Europe before the Aryans, there would perhaps be some name peculiar to the Basque language, or seeds would have been found in the lake-dwellingrs of Switzer- land and Savoy ; but this is not the case. The peoples in the neighbourhood of the Caucasus have names quite different to the Greek; in Tartar kiar, in Kalmuck chaja, in Armenian karan.^ The name cfdar exists also in Arabic for a variety of the cucumber.^ This is, therefore, a Turanian name anterior to the Sanskrit, whereby its cultm-e in Western Asia would be more than three thousand years old. It is often said that the cucumber is the kischschiiim, one of the fruits of Egypt regretted by the Israelites in the desert. '^ However, 1 do not find any Arabic name among the three given by Forskal which can be con- nected with this, and hitherto no trace has been found of the presence of the cucumber in ancient Egypt. 1 Bretschneider, letters of Aug. 23 and 26, 1881. 2 Tlieophrastus, Hist., lib. 7, cap. 4; Lenz, Bot. der Alien, p. 492. ' Heldreich, Nutzpfl. Griechen., p. 50. * Nemnich, Polygl. Lex., i. p. 1306. ^ Nemnich, ihid. * Forskal, FI. JRgypt., p. 76. ^ Rosenmiiller, Biblische Alterth., i. p. 97 ; Hamilton, Bot. de la Bille, p. 34. PLANTS CULTIVATED FOR THEIR FRUITS. 267 West Indian Gherkin — Cucumis Anguria, Linngeiis. This small species of cucumber is designated in the Bo7i Jardinier under the name of the cucumber Arada. The fruit, of the size of an egg, is very prickly. It is eaten cooked or pickled. As the plant is very produc- tive, it is largely cultivated in the American colonies. Descourtilz and Sir Joseph Hooker have published good coloured illustrations of it, and M. Cogniaux a plate with a detailed analysis of the flower.^ Several botanists affirm that it is wild in the West Indies. P. Browne,^ in the last century, spoke of the plant as the "little wild cucumber" (in Jamaica). Descourtilz said, " The cucumber grows wild everywhere, and principally in the dry savannahs and near rivers, whose banks afford a rich vcG^etation." The inhabitants call it the "maroon cucumber." Grisebach^ saw speci- mens in several other West India Isles, and appears to admit their wild character. M. E. Andre found the species growing in the sand of the sea-shore at Porto- Cabello, and Burchell in a similar locality in Brazil, and Riedel near Rio di Janeiro.* In the case of a number of other specimens gathered in the east of America from Brazil to Florida, it is unknown whether they were wild or cultivated. A wild Brazilian plant, badly drawn by Piso,^ is mentioned as belonging to the species, but I am very doubtful of this. Botanists from Tournefort down to our own dav have considered the Anguria to be of American origin, a native of Jamaica in particular. M. Naudin^ was the first to point out that all the other species of Cucumis are of the old world, and principally African. He wondered whether this one had not been introduced into America by the negroes, like many other plants which have become * Desconrtilz. Fl. Med. dcs Antilles, v. pi. 329 ; Ilooker, Bot. Mag., i. 5817 J Cogniaux, in Fl. Brasil, fasc. 78, pL 2. ^ Browne, Jamaica, edit. 2, p. 353. 3 Grisebach, FL of Brit. W. India Is., p. 288. * Cogniaux, uhi supra. * Guanerva-oha, in Piso, Brasil, edit. 1G58, p. 264 ; Maregraf, edit. 1648, p. 44, without illustration, calls it Cucumis sylvestris Bra^iticB. * Naudin, Ann. Sc. Kat., 4th series, vol. ii. p. 12. 2G8 OEIGIN OF CULTIVATED PLANTS. naturalized. However, unable to find any similar African plant, he adopted the general opinion. Sir Joseph Hooker, on the contiary, is inclined to believe that G. Aiigwria is a cultivated and modified form of some African species nearly allied to G. prophetarum and G. Figarei, although these are perennial. In favour of this h3q:>othesis, I may add : (1) The name unaroon cu- cumber, given in the French West India Islands, indicates a plant which has become wild, for this is the meaning of the word r^iaroon as applied to the negroes ; (2) its extended area in America from Brazil to the West Indies, always along the coast where the slave trade was most brisk, seems to be a pi oof of foreign oriiiin. If the species gi-ew in America previous to its discovery, it would, with such an extensive habitat, have been also found upon the west coast of America, and inland, which is not the case. The question can only be solved by a more complete knowledge of the African species of Gucumis, and by experiments upon fertilization, if any have the patience and abilit}'' necessary to do for the genus Cucumis what Naudin has done for the genus Gucurbita. Lastly, I would point out the absurdity of a common name for the Anoruria in the United States — Jerusalem Gucumher} After this, is it possible to take popular names as a o^uide in our search for oria^ins ? White Gourd-melon, or Benincasa — Benincasa hispida, Thunberg ; Benincasa cerifera, Savi. This species, wdiich is the only one of the genus Benincasa, is so like the pumpkins that early botanists took it for one,^ in spite of the waxy efflorescence on the surface of the fruit. It is very generally cultivated in tropical countries. It was, perhaps, a mistake to aban- don its cultivation in Europe after having tried it, for Naudin and the Bon Jardinier both recommend it. It is the ciiinhalam of Rheede, the camolenga of Rumphius, who had seen it cultivated in Malabar and the Sunda Islands, and give illustrations of it. * Darlington, Agric. Bot., p. 58. * Cucurhita Pepo of Loureiro and Eosburgh. PLANTS CULTIVATED FOR THEIR FRUITS. 2G9 From several works, even recent ones/ it might be supposed that it had never been found in a wild state, but if we notice the different names under which it has been described we shall find that this is not the case. Thus Gitcurhita hispida, Thunberg, and Lagenaria dasysteonon, Miquel, from authentic specimens seen by Cogniaux,^ are synonyms of the species, and these plants are wild in Japan.^ Cucurhita littoralis, Hass- karl,* found among shrubs on the sea-shore in Java, and Gi/mnopetalum septemlohum, Miquel, also in Java, are the Benincasa according to Cogniaux. As are also Cucurhita vacua, Mueller,^ and Cucurhita pruriens, Forster, of which he has seen authentic specimens found at Rockingham, in Australia, and in the Society Islands. jSfadeaud ^ does not mention the latter. Temporary naturalization may be suspected in the Pacific Isles and in Queensland, but the localities of Java and Japan seem quite certain. I am the more inclined to believe in the latter, that the cultivation of the Benincasa in China dates from the remotest antiquity^ Towel Gourd — Momordica cylindrica, Linnddus ; Luffa cylindrica, Roemer. Naudin ^ says, " Luffa cylindrica, which in some of our colonies has retained the Indian name petole, is probably a native of Southern Asia, and perhaps also of Africa, Australia, and Polynesia. It is cultivated by the peoples of most hot countries, and it appears to be naturalized in many places where it doubtless did not exist originally." Cogniaux ^ is more positive. " An indigenous species," he says, "in all the tropical regions » Clarke, in Fl. of Brit. Ind., n. p. 616. ^ Cogniaux, in de Candolle, Monogr. Phaner., lii- p. 513. ' Tlmnberg, Fl. Jap., p. 322*, Franchet and Savatier, Enum. PL Jap., i. p. 173 * Hasskarl, Catal. Horti, Bocjor. Alter., p. 190 ; Miquel, Flora Inclo- Batav. ^ Mueller, Fragm., vi, p. 186 ; Forster, Prodr. (no description) ; Seemann, Jour, of Bot., ii. p. 50. * Nadeaud, Plan. Ui9. 274 ORIGIN OF CULTIVATED PLANTS. Cuba, and is cultivated there," and Richard copies this phrase in the flora of R de La Sagra without adding any proof. Naudin says/ "a Mexican plant," but he does not give his reasons for asserting this. Cogniaux,^ in his recent monograph, mentions a great number of specimens gathered from Brazil to the West Indies with- out saying if he had seen any one of these given as wild. Seemann ^ saw the plant cultivated at Panama, and he adds a remark, important if correct, namely, that the name chayote, common in the isthmus, is the corruption of an Aztec word, chayotl. This is an indication of an ancient existence in Mexico, but I do not find the word in Hernandez, the classic author on the Mexican plants anterior to the Spanish conquest. The chayote was not cultivated in Cayenne ten years ago.* Nothing indicates an ancient cultivation in Brazil. The species is not mentioned by early writers, such as Piso and Marcgraf, and the name chttchu, given as Brazilian,^ seems to me to come from chocho, the Jamaica name, which is perhaps a corruption of the Mexican word. The plant is probably a native of the south of Mexico and of Central America, and was transported into the West India Islands and to Brazil in the eighteenth century. The species was afterwards introduced into Mauritius and Algeria, where it is very successful.^ Indian Fig, or Prickly Pear — Opuntia ficus indica, Miller. This fleshy plant of the Cactus family, which produces the fruit known in the south of Europe as the Indian fig, has no connection with the fig tree, nor has the fruit with the fia:. Its orio-in is not Indian but American. Everything is erroneous and absurd in this common name. However, since Linnaeus took his botanical name from it. Cactus ficus indica, afterwards connected with the genus Opiintia^ it was necessary to retain the specific * Naurlin, Ann. 8c. Nat, 4tli series, vol. xviii. p. 20.'>, * In Monogr. Phaner., iii. p. 902. ' Seemann, Bot. of Herald, p. 128. * Sagot, Journal de la 8oc. d'Hortic, de France, 1872. * Cogniaux, Fl. Brasil, fasc. 78. • Sagot, ibid. PLANTS CULTIVATED FOR THEIR FRUITS. 275 name to avoid changes which are a source of confusion, and to recall the popular denomination. The prickly forms, and those more or less free from spines, have been considered by some authors as distinct species, but an attentive examination leads us to regard them as one.-"- The species existed both wild and cultivated in Mexico before the arrival of the Spaniards. Hernandez ^ describes nine varieties of it, which shows the antiquity of its cultivation. The cochineal insect appears to feed on one of these, almost without thorns, more than on the others, and it has been transported with the plant to the Canary Isles and elsewhere. It is not known how far its habitat extended in America before man transported pieces of the plant, shaped like a racket, and the fruits, which are two easy ways of propagating it. Perhaps the wild plants in Jamaica, and the other West India Islands mentioned by Sloane,^ in 1725, were the result of its introduction by the Spaniards. Certainly the species has become naturalized in this direction as far as the climate permits ; for instance, as far as Southern Florida.* It was one of the first plants which the Spaniards in- troduced to the old world, both in Europe and Asia. Its singular appearance was the more striking that no other species belonging to the family had before been seen.^ All sixteenth-century botanists mention it, and the plant became naturalized in the south of Europe and in Africa as its cultivation was introduced. It was in Spain that the prickly pear was first known under the American name tuna, and it was probably the Moors who took it into Barbary Avhen they were expelled from the peninsula. They called it fig of the Christians.^ The custom of using the plant for fences, and the nourishing property of the fruits, which contain a large proportion of sugar, have determined its extension round the Mediterranean, and in general in all countries near the tropics. ^ Webb and Berthelot, Phytog. Canar., sect. 1, p. 208. * Hernandez, Theo. Novae Hisp., p, 78, ' Sloaug, Jamaica, ii, p. 150. * Chapman, Flora of Southern States, p. 144. ^ The cactos of the Greeks was quite a different plant, ® Steinbeil, in Boissier, Voyage JJot, en Esiiagne. i, p, 25. 13 276 ORIGIN OF CULTIVATED PLANTS. The cultivation of the cochineal, which was unfavour- able to the production of the fruit/ is dying out since the manufacture of colouring matters by chemical processes. Gooseberry — Rihes grossidaria and H. Vacrispa, Linnaeus. The fruit of the cultivated varieties is generally smooth, or provided with a few stiff hairs, while that of the wild varieties has soft and shorter hairs ; but inter- mediate forms exist, and it has been shown by experi- ment that bv sowing the seeds of the cultivated fruit, plants with either smooth or hairy fruit are obtained.'-^ There is, therefore, but one species, which has produced under cultivation one principal variety and several sub- varieties as to the size, colour, or taste of the fruit. The gooseberry grows wild throughout temperate Europe, from Southern Sweden to the mountainous regions of Central Spain, of Italy, and of Greece.^ It is also mentioned in Northern Africa, but the last published catalogue of Algerian plants* indicates it only in the mountains of Aures, and Ball has found a variety in the Atlas of Marocco.^ It grows in the Caucasus,^ and under more or less different forms in the western Himalavas.'^ The Greeks and Romans do not mention the species, which is rare in the South, and which is hardly worth planting where grapes will ripen. It is especially in Germany, Holland, and England that it has been culti- vated from the sixteenth century,^ principally as a seasoning, whence the English name, and the French groseille a niaqimreauoc (mackerel currant). A wine is also made from it. The frequency of its cultivation in the British Isles and in other places where it is found wild, which are ' Webb and Berthelot, Phytog. Canar.^ vol. iii. sect. 1, p. 208 * Robson, quoted in English Botany, pi. 2057. ' Nyman, Coyispectus Fl. Europece, p. 266 ; Boissier, Fl. Or., ii. p. 815. * Munbj, Catal., edit. 2, p. 15. * Ball, Spicilegium FL. Maroc, p. 449. * Ledebour, Fl. Ross., ii. p. 19i ; Boissier. iihi supra. ^ Clarke, in Hooker's Fl. Brit, Ind., ii. p. 410, * Phillips, Account of Fruits, p. 174. PLANTS CULTIVATED FOR THEIR FRUITS. 277 often near gardens, has suggested to some English botanists the idea of an accidental naturalization. This is likely enough in Ireland ; ^ but as it is an essentially European species, I do not see why it should not have existed in England, where the wild plant is more common, since the establishment of most of the species of the British flora ; that is to say, since the end of the glacial period, before the separation of the island from the continent. Phillips quotes an old English na,xne, feaberry or feahes, which supports the theory of an ancient exist- ence, and two Welsh names,^ of which I cannot, however, certify the originality. Red Currant — Rihes ruhrum, Linnaeus. The common red currant is wild throuj^hout Northern and Temperate Europe, and in Siberia ^ as far as Kamts- chatka, and in America, from Canada and Vermont to the mouth of the river Mackenzie.^ Like the preceding species, it was unknown to the Greeks and Romans, and its cultivation was only intro- duced in the Middle Ages. The cultivated plant hardly differs from the wild one. That the plant was foreign to the south of Europe is shown by the name of groseillier d'outremer (currant from beyond the sea), given in France^ in the sixteenth century. In Geneva the currant is still commonly called raisin de mare, and in the canton of Soleure oneei^trilhli. I do not know why the species was supposed, three centuries ago, to have come from be- yond seas. Perhaps this should be understood to mean that it was brought by the Danes and the Northmen, and that these peoples from beyond the northern seas introduced its cultivation. I doubt it, however, for the Ribes riibriim is wild in almost the whole of Great Britain^ and in Normandy; '' the English, who were in constant communication with the Danes, did not cultivate it as late as 1557, from a list of the fruits of that epoch ' Moore and More, Contrib. to the Cyhele Hyhernica, p. 113. ^ Davies, Welsh Botanologij, p. 24. ^ Ledebour, Fl. Ross., ii. p. 199. * Torrey and Gray, Fl. N. Anier., i. p. 150. * Dodoneus, p. 748. * Watson, Cyhele Brit. '' Brebisson, Flore de Normandie, p. 99. 278 ORIGIN OF CULTIVATED PLANTS. drawn up by Th. Tusser, and published by Phillips;^ and even in the time of Gerard, in 1597,^ its cultivation was rare, and the plant had no particular name.^ Lastly, there are French and Breton names which indicate a cultivation anterior to the Normans in the west of France. The old names in France are given in the dictionary by Menage. According to him, red currants are called at Kouen gardes, at Caen grades, in Lower Normandy gra- dilles, and in Anjou castilles. Menage derives all these names from ruhius, rubicus, etc., by a series of imaginary transformations, from the word ruber, red. Legonidec * tells us that red currants are also called Kastilez (1. liquid) in Brittany, and he derives this name from Castille, as if a fruit scarcely known in Spain and abundant in the north could come from Spain. These words, found both in Brittany and beyond its limits, appear to me to be of Celtic origin ; and I may mention, in support of this theory, that in Legonidec's dictionar}^ gardis means rough, harsh, pungent, sour, etc., which gives a hint as to the etymology. The generic name Rihes has caused other errors. It was thought the plant might be one which was so called by the Arabs; but the word comes rather from a name for the currant very common in the north, ribs in Danish,^ risp and resp in Swedish.^ The Slav names are quite different and in considerable number. Black Currant — Cassis ; Rihes nigrum, Linnaeus. The black currant grows wild in the north of Europe, from Scotland and Lapland as far as the north of France and Italy ; in Bosnia,'^ Armenia,® throughout Siberia, in the basin of the river Amur, and in the western Hima- * Phillips, Account of Fruits, p. 13G. 2 Gerard, Herbal, p. 1143. ' That of currant is a later introduction, given from the resemblance to the grapes of Corinth (Phillips, ibid.). ^ Legonidec, Diction. Celto-Breton. * Moritzi, Did. Inedii des Noms Vulgaires* ^ Linnaeus, Flora Suecica, n. 197. ' Watson, Compend. Cybele, i. p. 177 ; Fries, Summa Veg. Scand.y p. 39; Nyman, Conspect. Fl. Europ., p. 2C6. Loissicr, Fl. Or., ii. p. 815. 8 PLANTS CULTIVATED FOR THEIR FRUITS. 279 layas ; ^ it often becomes naturalized, as for instance, in the centre of France.^ This shrub was unknown in Greece and Italy, for it is proper to colder countries. From the variety of the names in all the languages, even in those anterior to the Aryans, of the north of Europe, it is clear that this fruit was very early sought after, and its cultivation was pro- bably begun before the Middle Ages. J. Bauhin^ says it was planted in gardens in France and Italy, but most sixteenth-century authors do not mention it. In the Histoire de la Vie Privee des Frangais, by Le Grand d'Aussy, published in 1872, vol. i. p. 232, the following curious passage occurs : " The black currant has been cultivated hardly forty years, and it owes its reputa- tion to a pamphlet entitled Culture du Cassis, in which the author attributed to this shrub all the virtues it is possible to imagine." Further on (vol. iii. p. 80), the author mentions the frequent use, since the publication of the pamphlet in question, of a liqueur made from the black currant. Bosc, who is always accurate in his articles in the Dictionnaire d' Agriculture, mentions this fashion under the head Currant, but he is careful to add, " It has been very long in cultivation for its fruit, which has a peculiar odour agreeable to some, disagreeable to others, and which is held to be stomachic and diuretic." It is also used in the manufacture of the liqueurs known as ratafia de Cassis.* Olive — Olea Europea, Linnaeus. The wild olive, called in botanical books the variety * Ledebour, Fl. Ross., p. 200 ; Maximowicz, PrimiticB Fl. Amur., p. 119 ; Clarke, in Hooker, FL Brit, hid., ii. p. 411. * Boreau, Flore dii Centre de la France, edit. 3, p. 2G2. ^ Bauhin, Hist. Plant., ii. p. 99. * This name Cassis is curious. Littre says tbat it seems to have been introduced late into the language, and that he does not know its origin. I have not met with it in botanical works eai'lier than the middle of the seventeenth century. My manuscript collection of common names, among more than forty names for this species in different languages or dialects has not one which resembles it. Buchoz, in his Dictionnnire desPlantes, 1770, i. p. 289, calls the plant the Cassis or Cassetier des Poifevins. The old French name was Poivrier or groseillier noir. Larousse's dictionary says that good liqueurs were made at Cassis in Provence. Can this be the origin of the name ? 280 ORIGIN OF CULTIVATED PLANTS. sylvestris or oleaster, is distinguished from the cultivated olive tree by a smaller fruit, of which the flesh is not so abundant. The best fruits are obtained by selecting the seeds, buds, or grafts from good varieties. The oleaster now exists over a wide area east and west of Syria, from the Punjab and Beluchistan ^ as far as Portusfal and even Madeira, the Canaries and even Marocco,^ and from the Atlas northwards as far as the south of France, the ancient Macedonia, the Crimea, and the Caucasus.^ If we compare the accounts of travellers and of the authors of floras, it will be seen that towards the limits of this area there is often a doubt as to the wild and indigenous (that is to say ancient in the country) nature of the species. Sometimes it offers itself as a shrub which fruits little or not at all ; and sometimes, as in the Crimea, the plants are rare as though they had escaped, as an exception, the destructive eflfects of winters too severe to allow of a definite estal)lishment. As resrards Alf]reria and the south of France, these doubts have been the subject of a discussion among competent men in the Botanical Society.* They repose upon the uncontestable fact that birds often transport the seed of the olive into uncultivated and sterile places, where the wild form, the oleaster, is produced and naturalized. The question is not clearly stated when we ask if such and such olive trees of a given locality are really wild. In a woody species which lives so long and shoots again from the same stock when cut off" by accident, it is impossible to know the origin of the individuals observed. They may have been sown by man or birds at a very early epoch, for olive trees of more than a thousand years old are known. The eflect of such sowing is a naturaliza- tion, which is equivalent to an extension of area. The point in question is, therefore, to discover what was the * Aitchison, Catalogue, p. 8G. * Lowe, Man. Fl. of Madeira, ii. p. 20 ; Webb and Berthelot, Hist. Nat. des Canaries^ GSog. Bat., p. 48 ; Ball, Spicil. Fl. Maroc, p. 5G5. ' Cosson, Bull. Soc. Bot. France, iv. p. 107, and vii. p. 31 ; Grisebach, Spicil. Fl. Ruinelicoi, ii. p. 71 ; Steven, Verzeich. der Tauriach. Halbins., p. 248 ; Ledebour, Fl. Ross., p. 38. * Bulletin, iv. p. 107. PLANTS CULTIVATED FOR THEIR FRUITS. 281 home of the species in very early prehistoric times, and how this area has grown larger by different modes of transport. It is not by the study of living olive trees that this question can be answered. We must seek in what coun- tries the cultivation began, and how it was propagated. The more ancient it is in any region, the more probable it is that the species has existed wild there from the time of those geological events which took place before the coming of prehistoric man. The earliest Hebrew books mention the olive salt, or zeit,^ both wild and cultivated. It was one of the trees promised in the land of Canaan. It is first mentioned in Genesis, where it is said that the dove sent out by Noah should bring back a branch of olive. If we take into account this tradition, which is accompanied by miracu- lous details, it may be added that the discoveries of modern erudition show that the Mount Ararat of the Bible must be to the east of the mountain in Armenia which now bears that name, and which was anciently called Masis. From a study of the text of the Book of Genesis, Fran9ois Lenormand ^ places the mountain in question in the Hindu Kush, and even near the sources of the Indus. This theory supposes it near to the land of the Aryans, yet the olive has no Sanskrit name, not even in that Sanskrit from which the Indian lano-ua^^es ^ are derived. If the olive had then, as now, existed in the Punjab, the eastern Aryans in their migrations towards the south would probably have given it a name, and if it had existed in the Mazanderan, to the south of the Cas- pian Sea, as at the present day, the western Aryans would perhaps have known it. To these negative indi- cations, it can only be objected that the wild olive attracts no considerable attention, and that the idea of extracting oil from it perhaps arose late in this part of Asia. * Rosentnuller, Handhuch der Bibl. Alterth., vol. iv. p. 258 ; Hamilton, Bot. de la Bible, p. 80, where the passages are indicated. * Fr, Lenormand, Manuel de I'Hist. Auc. de I'Orient., 1869, vol. i. p. 31. * Fick, Worterhuch, Piddington, Index, only mentions one Hindu name, jn/pai. 282 ORIGIN OF CULTIVATED PLANTS. Herodotus * tells us that Babylonia grew no olive trees, and that its inhabitants made use of oil of sesame. It is certain that a country so subject to inundation was not at all favourable to the olive. The cold excludes the higher plateaux and the mountains of the north of Persia. I do not know if there is a name in Zend, but the Semitic word sait must date from a remote antiquity, for it is found in modern Persian, seitiin^ and in Arabic, zeitiin, sjetun.^ It even exists in Turkish and among the Tartars of the Crimea, seitun,^ \vhich may signify that it is of Turanian origin, or from the remote epoch when the Turanian and Semitic peoples intermixed. The ancient Egyptians cultivated the olive tree, which they called tat.^ Several botanists have ascertained the presence of branches or leaves of the olive in the sarco- phagi.^ Nothing is more certain, tliough Hehn*^ has recently asserted the contrary, without giving any proof in support of his opinion. It would be interesting to know to what dynasty belong the most ancient mummy- cases in which olive branches have been found. The Egyptian name, quite ditferent to the Semitic, shows an existence more ancient than the earliest dynasties. I shall mention presenth^ another fact in support of this great antiquity. Theophrastus says ^ that the olive was much grown, and the harvest of oil considerable in Cyrenaica, but he does not say that the species Avas wild there, and the quantity of oil mentioned seems to point to a cultivated variety. The low-lying, very hot country between Egypt and the Atlas is little favourable to a naturalization of the olive outside the plantations. Kralik, a very accurate botanist, did not anywhere see on his journey 1 Herodotus, Hist., bk. i. c. 193. « Boissier, Fl. Oi-ient., iv. p. 3C. ' Ebn Baithar, Germ, trans., p. 569 ; Forskal, Plant. Egypt., p. 49. * Boissier, ihid. ; Steven, ibid. * Unger, Die Pfianz. der Alten. JF.gypt, p. 45. ^ De Candolle, Phydol. Veget., p. 696; Pleyte, qnoted by Eraun and Ascherson, Sifzher. Naturfor. Ges., May 15, 1877. ^ Hehn, Kidturpflanzen, edit. 3, p. 88, line 9. ^ Theophrastus, Hist. Plant., lib. iv. c. 3. PLANTS CULTIVATED FOR THEIR FRUITS. 28 o to Tunis and into Egypt tlio olive growing wild,* although it is cultivated in the oases. In Egypt it is only culti- vated, accordinoj to Schweinfurth and Ascherson,^ in their resume of the Flora of the Nile Valley. Its prehistoric area probably extended from Syria towards Greece, for the wild olive is very common along the southern coast of Asia Minor, where it forms reo-ular woods.^ It is doubtless here and in the archipelago that the Greeks early knew the tree. If they had not known it on their own territory, had received it from the Semites, they would not have given it a special name, elaia, whence the Latin olea. The Iliad and the Odyssey mention the hardness of the olive wood and the practice of anointing the body with olive oil. The latter was in constant use for food and lighting. Mythology attributed to Minerva the planting of the olive in Attica, which probably signifies the introduction of cultivated varieties and suitable processes for extracting the oil. Arist?eus introduced or perfected the manner of pressing the fruit. The same mythical personage carried, it was said, the olive tree from the north of Greece into Sicily and Sar- dinia. It seems that this may have been early done by the Phoenicians, but in support of the idea that the species, or a perfected variety of it, was introduced by the Greeks, I may mention that the Semitic name seit has left ..no trace in the islands of the Mediterranean. We find the Graeco-Latin name here as in Italy,* while upon the neighbouring coast of Africa, and in Spain, the names are Egyptian or Arabic, as I shall explain directly. The Romans knew the olive later than the Greeks. According to Pliny ,^ it was only at the time of Tarquin the Ancient, 627 B.C., but the species probably existed already in Great Greece, as in Greece and Sicily. Besides, Pliny was speaking of the cultivated olive. A remarkable fact, and one which has not been noted * Kralik, Bull. Soc. Bot. Fr., W. p. lOS. ^ Beitrage zur Ft. Mthio'piens, p. 28L 8 Balansa, Bull. Soc. Bot. de Fr., iv. p. 107. * Moris, Fl. Sard., iii. p. 9 ; Bertoloui, Fl. lial.y i. p. 4G. * riiiij, Hist., lib. XV. cap. 1. 284 ORIGIN OF CULTIVATED PLANTS. or discussed by philologists, is that the Berber name for the olive, both tree and fruit, has the root taz or tas, similar to the tat of the ancient Egyptians. The Kabyles of the district of Algiers, according to the French- Berber dictionary, published by the French Government, calls the wild olive tazehhoujt, tesettha, ovC zehhoiij, and the grafted olive tazemonourt, tasettka, ovu zemT,iou7\ The Touareo's, another Berber nation, call it tmnaldnet} These are strong indications of the antiquity of the olive in Africa. The Arabs having conquered this country and driven back the Berbers into the mountains and the desert, having likewise subjected Spain excepting the Basque country, the names derived from the Semitic zeit have prevailed even in Spanish. The Arabs of Algiers say zenhoiidje for the wild, zitoun for the cultivated olive,^ zit for olive oil. The Andalusians call the wild olive aze- huche, and the cultivated aceytimo? In other provinces we find the name of Latin origin, olivio, side by side with the Arabic words.* Tlie oil is in Spanish aceyte, which is almost the Hebrew name ; but the holy oils are called oleos Santos, because they belong to Rome. The Basques use the Latin name for the olive tree. Early voyagers to the Canaries, Bontier for instance, in 1403, mention the olive tree in these islands, where modern botanists regard it as indigenous.^ It may have been introduced by the Phoenicians, if it did not pre- viously exist there. We do not know if the Guanchos had names for the olive and its oil. Webb and Berthelot do not give any in their learned chapter on the language of the aborigines,^ so the question is open to conjecture. It seems to me that the oil would have played an impor- tant part among the Guanchos if they had possessed the olive, and that some traces of it would have remained in the actual speech of the people. From this point of view * Dnveyrier, Les Tovaregs du Nord (1864), p. 179. ' Munby, Flore de VAlgerie, p. 2 ; Debeaux, Catal. Boghar, p. G3. ' Boirsier, Voyage Bot. en Espagne, edit. 1, vol. ii. p. -107. * Willkomm and Laiige, Prod. Fl. Hispan., ii. p. 672. * Webb and Berthelot, Hist. Nat. des Canaries, Geog. Bot., pp. 47,43. ® Webb and Berthelot, ibid., Ethnographie, p 1S8, PLANTS CULTIVATED FOR THEIR FRUITS. 285 the naturalization in the Canaries is perhaps not more ancient than the Phoenician voyages. No leaf of the olive has hitherto been found in the tufa of the south of France, of Tuscany, and Sicily, where the laurel, the myrtle, and other shrubs now existing have been discovered. This is an indication, until the contrary is proved, of a subsequent naturalization. The olive thrives in dry climates like that of Syria and Assyria. It succeeds at the Cape, in parts of America, in Australia, and doubtless it will become wild in these places when it has been more generally planted. Its slow growth, the necessity of grafting or of choosing the shoots of good varieties, and especially the concurrence of other oil-producing species, have hitherto impeded its extension ; but a tree which produces in an ungrateful soil should not be indefinitely neglected. Even in the old w^orld, where it has existed for so many thousands of years, its productiveness might be doubled by taking the trouble to graft on wild trees, as the French have done in Algeria. Star Apple — Chrysophyllum Cainito, Linnreus. The star apple belongs to the family of the Sapotaceoe, It yields a fruit valued in tropical America, though Europeans do not care much for it. I do not find that any pains have been taken to introduce it into the colonies of Asia or Africa. Tussac gives a good illustration of it in his Flore des Antilles, vol. ii. pi. 9. Seemann ^ saw the star apple wild in several places in the Isthmus of Panama. De Tussac, a San Domingo colonist, considered it wild in the forests of the West India Islands, and Grisebach^ says it is both wild and cultivated in Jamaica, San Domingo, Antigua, and Tri- nidad. Sloane considered it had escaped from cultivation in Jamaica, and Jacquin says vaguely, "Inhabits Mar- tinique and San Domingo."^ Caimito, or Abi — Luciima Cainito, Alph. de Candolle. Tliis Peruvian Caimito must not be confounded with * ReeTnann, Bot. of the Herald., p. 1G6. « GL-isebach, Flora of Brit. W. hid. Isl., p. 398- ^ Sloane, Jamaica, ii. p. 170; Jacquin, Atner., p. 53. 286 ORIGIN OF CULTIVATED PLANTS. the Chrysopliylliim Ca'inito of the West Indies. Both belong to the family Sapotacese, but tlie flowers and seeds are different. There is a figure of this one in Ruiz and Pavon, Flora Peruviana, vol. iii. pi. 240. It has been transported from Peru, where it is cultivated, to Ega on the Amazon River, and to Para, where it is commonly called abi or ahiu} Ruiz and Pavon say it is wild in the warm reo-ions of Peru, and at the foot of the Andes. Marmalade Plum, or Mammee Sapota — Lucuma mani- mosa, Gsertner. This fruit tree, of the order Sapotace?e and a native of tropical America, has been the subject of several mistakes in works on botany^ There- exists no satis- factory and complete illustration of it as yet, because colonists and travellers think it is too well known to send selected specimens of it, such as may be described in herbaria. This neo-lect is common enouo-h in the case of cultivated plants. The mammee is cultivated in the West Indies and in some warm regions of America. Sagot tells us it is grown in Venezuela, but not in Cayenne.^ I do not find that it has been transported into Africa and Asia, the Philippines^ excepted. This is probably due to the insipid taste of the fruit. Hum- boldt and Bonpland found it wild in the forests on the banks of the Orinoco.^ All authors mention it in the West Indies, but as cultivated or without asserting that it is wild. In Brazil it is only a garden species. Sapodilla — Sapota achras, Miller. The sapodilla is the most esteemed of the order Sapotacese, and one of the best of tropical fruits. " An over-ripe sapodilla," says Descourtilz, in his Flore des Antilles, "is melting, and has the sweet perfumes of honey, jasmin, and lily of the valley." There is a very good illustration in the Botanical Magazine, pis. 3111 and 3112, and in Tussac, Flore des Antilles, i. pi. 5. It ' Flora BraHl., vol, vii. p. 8S. ^ See the sjuonyms in the Flora Brasiliensis, vol. vii. p. GQ, ' Sagot, JoxLvn. Soc. d'Hortic. de France, 1872, p. 3-17. * Blanco, Fl. de Filipinas, under the name Achras lucuma.. ^ Nova Genera, iii. p. 210. PLANTS CULTIVATED FOR THEIR FRUITS. 287 has been introduced into gardens in Mauritius, the Malay Archipelago, and India, from the time of Rheede and Rumphius, but no one disputes its American origin. Several botanists have seen it wild in the forests of the Isthmus of Panama, of Campeachy/ of Venezuela,^ and perhaps of Trinidad.^ In Jamaica, in the time of Sloane, it existed only in gardens.* It is very doubtful that it is wild in the other West India Islands, although perhaps the seeds, scattered here and there, may have naturalized it to a certain degree. Tussac says that the young plants are not easy to rear in the plantations. Aubergine — Solanum melongena, Linnaeus ; Solanum esculentuTn, Dunal. The aubergine has a Sanskrit name, vartta, and several names, which Piddington in his Index considers as both Sanskrit and Bengali, such as hong, hartaJcon, mahoti, hingoli. Wallich, in his edition of Roxburgh's Indian Flora, gives vartta, varttaJcoii, varttaha hicnguna, whence the Hindustani hungan. Hence it cannot be doubted that the species has been known in India from a very remote epoch. Rumphius had seen it in gardens in the Sunda Islands, and Loureiro in those of Cochin-China. Thunberg does not mention it in Japan, though several varieties are now cultivated in that country. The Greeks and Romans did not know the species, and no botanist mentions it in Europe before the beginning of the seven- teenth century,^ but its cultivation must have spread towards Africa before the Middle Ages. The Arab phy- sician, Ebn Baithar,^ who wrote in the thirteenth century, speaks of it, and he quotes Rhasis, who lived in the ninth century. Ran wolf had seen the plant in the gardens of Aleppo at the end of the sixteenth century. It was called melanzana and hedengianu This Arabic * Dampier and Lussan, iu Sloane's Jamaica, ii. p. 172; Seetnann, Botany of the Herald.., p. 166. * Jacquio, Amer., p. 59; Humboldt and Bonpland, Nova Genera, iii. p. 239. " Grisebach, Flora, of Brit. W. Ind., p. 309. * Sloane, uhi supra, ' Dunal, Hist, des Solanum, p. 209. * Ebn Baithar, Germ, trans., i. p. 116. ^ Rauwulf, Flora Orient., ed. Groniuguc, p. 26. 288 ORIGIN OF CULTIVATED PLANTS. name, wh^cli Forskal writes hadmjcon, is the same as the Hindustani hadanjan, which Piddington gives. A sign of antiquity in Northern Africa is the existence of a name, tahendjalts, among the Berbers or Kabyles of the province of Algiei's,^ which differs considerably from the Arab word. Modern travellers have found the aubergine cultivated in the whole of the Nile Yalley and on the coast of Guinea.^ It has been transported into America. The cultivated form of Solanum melongena has not hitherto been found wild, but most botanists are agreed in resfardingf Solanum insanum, Roxburgh, and >S^. incanum, Linnaeus, as belonging to the same species. Other synonyms are sometimes added, the result of a study made by Nees von Esenbeck from numerous speci- mens.^ >S'. insanum appears to have been lately found wild in the Madras presidency and at Tong-dong in Burmah. The publication of the article on the Sola- naceae in the Flora of British India will probably give more precise information on this head. Red Pepper — Capslcurii. In the best botanical works the genus Capsicum is encumbered with a number of cultivated forms, which have never been found wild, and which differ especially in their duration (which is often variable), or in the form of the fruit, a character whicli is of little value in plants cultivated for that special organ. I shall speak of the two species most often culti- vated, but I cannot refrain from stating my opinion that no capsicum is indigenous to the old world. I believe them to be all of American origin, though I cannot absolutely prove it. These are my reasons. Fruits so conspicuous, so easily grown in gardens, and so agreeable to the palate of the inhabitants of hot countries, would have been very quickly diffused through- out the old world, if they had existed in the south of Asia, as it has sometimes been supposed. They would have had names in several ancient languages. Yet ' Diet. Fr.-Berhcre, published by the French Government, * Thonning, under the name S. edule; Hooker, Niger Flora, p. 473. ^ Trans, of Linn. Soc, xvii. p. 48, Baker, Fl. of Maurit., p. 215. PLANTS CULTIVATED FOR THEIR FRUITS. 289 neither Romans, Greeks, nor evon Hebrews were ac- quainted with them. They are not mentioned in ancient Chinese books.-^ The islanders of the Pacific did not cultivate them at the time of Cook's voyages,^ in spite of their proximity to the Sunda Isles, where Rumphius mentions their very general use. The Arabian physician, Ebn Baithar, who collected in the thirteenth century all that Eastern nations knew about medicinal plants, says nothing about it. Roxburgh knew no Sanskrit name for the capsicums. Later, Piddington mentions a name for C. frutescens, hj^an-onaricha,^ which, he says is Sanskrit; but this name, which may be compared to that of black pepper (niuricha, Tiiurichung), is probably not really ancient, for it has left no trace in the Indian lanofuaofes which are derived from Sanskrit.^ The wild nature and ancient existence of the capsicum is always uncertain, owing to its very general cultivation ; but it seems to me to be more often doubtful in Asia than in South America. The Indian specimens described b}'' the most trustworthy authors nearly all come from the her- baria of the East India Company, in which we never know whether a plant appeared really wild, if it was found far from dwellings, in forests, etc. For the localities in the Malay Archipelago authors often give rubbish-heaps, hedges, etc. We pass to a more particular examination of the two cultivated species. Annual Capsicum — Capsicinn annuiion, Linnaeus, This species has a number of ditierent names in European languages,^ which all indicate a foreign origin and the resemblance of the taste to that of pepper. In French it is often called poivre de Guinee (Guinea pepper), but also 2^oivre du Brezil, d'Inde (Indian, Brazi- lian pepper), etc., denominations to which no importance can be attributed. Its cultivation was introduced into Europe in the sixteenth century. It was one of the peppers that Piso and Marcgraf^ saw gi'own in Brazil ^ Bretschueider, On the Study and Value, etc., p. 17. ^ Forstcr, De Plant is Escul. InsuL, etc. ' Piddington, Index. * Piddington, at the word Capsicum. ' Nemnich, Lexicon, gives twelve French and eight German names. 6 Piso, p. 107 ; Marcgraf , p. 39. 290 ORIGIN OF CULTIVATED PLANTS. under tlie name qidja or quiya. They say nothing as to its origin. The species appears to have been early culti- vated in the West Indies, where it has several Carib names.^ Botanists who have most thoroughly studied the genus Capsicum^ do not appear to have found in herbaria a single specimen which can be considered wild. I have not been more fortunate. The original home is probably Brazil. C. grossiiin^ Willdenow, seems to be a variety of the same species. It is cultivated in India under the name hafree viurich, and kafree chilly, but Roxburgh did not consider it to be of Indian origin.^ Shrubby Capsicum — Capsicum frutescens, Willdenow. This species, taller and with a more woody stock than C. aniuiiim, is generally cultivated in the warm regions of both hemisjDheres. The great part of our so-called Cayenne pepper is made from it, but this name is given also to the product of other peppers. Roxburgh, the author who is most attentive to the origin of Indian plants, does not consider it to be wild in India. Blume says it is naturalized in the Malay Archipelago in hedges.* In America, on the contrary, where its culture is ancient, it has been several times found wild in forests, apparently indigenous. De Martins brought it from the banks of the Amazon, Pcieppig from the province of Maynas in Peru, and Blanchet from the province of Bahia^ So that its area extends from Bahia to Eastern Peru, which ex- plains its diffusion over South America generally. Tomato — Ly coper sicuioi escidentum, Miller, The tomato, or love apple, belongs to a genus of the Solanese, of which all the species are American.^ It has no name in the ancient languages of Asia, nor even in modem Indian languages.*^ It was not cultivated in Japan in the time of Thunberg, that is to say a century * Desconrtilz, Flore Medicate des Antilles, vi. pi. 423. ' Fingerhuth, Monographia Gen. Capsici, p. 12 ; Sendtner, in Flora Brasil., voL x. p. 147. ' Eoxburgh, Fl. Ind.y edit. Wall, ii. p. 260 ; edit. 1832, ii. p. 574. * Blume, Bijdr., ii. p. 704. * Sendtner, in Fl. Bras., x. p. 143. ^ Alph. de Candolle, Prodr., xiii. part 1, p. 20. ^ Roxburgh, Fl. Ind., edit. 1832, vol. i. p. b'oo j Tiddington, Index, PLANTS CULTIVATED FOR THEIR FRUITS. 291 asfo, and the silence of ancient writers on China on th's head shows that it is of recent introduction there. Ruiii- phius ^ had seen it in gardens in the Mai ly Archipelago. The Malays called it tomatte, but this is an American name, for C. Bauhin calls the species tumatle A'nierica- oiorurn. Nothing leads us to suppose it was known in Europe before the discovery of America. The first names given to it by botanists in the six- teenth century indicate that they received the plant from Peru.^ It was cultivated on the continent of America before it was grown in the West India Islands, for Sloane does not mention it in Jamaica, and Hughes^ says it was brought to Barbados from Portugal hardly more than a century ago. Humboldt considered that the cul- tivation of the tomato was of ancient date in Mexico.^ I notice, however, that the earliest work on the plants of this country (Hernandez, Historia) makes no mention of it. Neither do the early writers on Brazil, Piso and Marcgraf, speak of it, although the species is now culti- vated throughout tropical America. Thus by the process of exhaustion we return to the idea of a Peruvian origin, at least for its cultivation. De Martins^ found the plant wild in the neigh- bourhood of Rio de Janeiro and Para, but it had per- haps escaped from gardens. I do not know of any botanist w^ho has found it really wild in the state in which it is familiar to us, with the fruit more or less large, lumpy, and with swelled sides ; but this is not the case with the variety with small spherical fruit, called L. cerasiforme in some botanical works, and considered in others (and rightly so, I think ^) as belonging to the same species. This variety is wild on the sea-shore of ' Enmpljius, Amhoinj v. p. 416. 2 Mala Peruviana, Pomi del Peru, in Bauhin's Uist., iii, p. C21. ' Hughes, Bai-hados, p. 148. * Huniboidt, Espagne, edit. 2, vol. ii. p. 472. * Fl. Brasil., vol. x. p. 126, ; * The proportions of the calyx and the corolla are the same as those of the cultivated tomato, but they are different in the allied species S. Humholdtii, of which the fruit is also eaten, according to Humboldt, who found it wild in Venezuela. 292 ORIGIN OF CULTIVATED PLANTS. Peru,^ at Tarapoto, in Eastern Peru,^ and on the frontiers of Mexico and of the United States towards California.^ It is sometimes naturalized in clearings near gardens.* It is probably in this manner that its area has extended north and south from Peru. Avocado, or Alligator Pear — Fersea gratissima, Gsertner. The avocado pear is one of the most highly prized of tropical fruits. It belongs to the order Laurineae. It is like a pear containing one large stone, as is well shown in Tussac's illustrations, Flore des Aiitilles, iii. pi. 3, and in the Botanical Magazine, pi. 4580. The com- mon names are absurd. The origin of that of alligator is unknown; avocado is a corruption of the Mexican ahnaca, or aguacate. The botanical name Fersea has nothing to do with the 'persea of the Greeks, which was a Cordia. Clusius,^ writing in 1601, says that the avo- cado pear is an American fruit tree introduced into a garden in Spain ; but as it is widely spread in the colo- nies of the old world, and has here and there become almost wild,^ it is possible to make mistakes as to its origin. This tree did not exist in the gardens of British India at the beginning of the nineteenth century. It had been introduced into the Sunda Isles '^ in the middle of the eighteenth century, and in 1750 into Mauritius and Bourbon.^ In America its actual area in a wild state is of un- common extent. The species has been found in forests, on the banks of rivers, and on the sea-shore from Mexico and the West Indies as far as the Amazon.^ It has not * Ruiz and Pavon, Flor. Peruv., ii. p. 37. ' Spruce, n. 4143, in Boissier's herbarium. ' Asa Gray, Bot. of Califor., i. p. 538. * Baker, Fl. of ilaurit., p. 216. * Clnsius, Hif'fona, p. 2. * For instance in Madeira, according to Grisebach, Fl. of Brit. W. Inch, p. 280; in Mauritius, the Seychelles and Rodriguez, according to Baker, Flora of Mauritius, p. 290. * It is not in Rumphius, ' Aublet, Guyane, i. p. 364. " Meissner, in de Candolle, Prodromus, vol, xv. part 1, p. 52 ; and Flora Brasil.^ vol. v. p. 158. For Mexico, Hernandez, p. 89 ; for Veneznela and Para, Nees, Laurineaif p. 129 j for Eastern Peru, Poeppig, Exsicc, seen hj Meissner. PLANTS CULTIVATED FOR THEIR FRUITS. 293 always occupied this vast region. P. Browne says dis- tinctly that the avocado pear was introduced from the Continent into Jamaica, and Jacquin held the same opinion as regards the West India Islands generally.^ Piso and Marcgraf do not mention it for Brazil, and Martins gives no Brazilian name. At the time of the discovery of America, the species was certainly wild and cultivated in Mexico, according to Hernandez. Acosta ^ says it was cultivated in Peru under the name of palto, which was that of a people of the eastern part of Peru, among whom it was abundant.^ I find no proof that it was wild upon the Peruvian littoral. Papaw — Carica Papaya, Linngeus ; Papaya vulgaris, de Candolle. The papaw is a large herbaceous plant rather than a tree. It has a sort of juicy trunk terminated by a tuft of leaves, and the fruit, which is like a melon, hangs down under the leaves.^ It is now grown in ail tropical coun- tries, even as far as thirty to thirty-two degrees of latitude. It is easily naturalized outside plantations. This is one reason why it has been said, and people still say that it is a native of Asia or of Africa, whereas Robert Bi"own and I proved in 18 i8 and 1855 its American origin.^ I repeat the arguments against its supposed origin in the eastern hemisphere. The species has no Sanskrit name. In modern Indian lano;uao'es it bears names derived from the American word papaya, itself a corruption of the Carib ahahai.^ Bumphius'^ says that the inhabitants of the Malay Arclii- pelago considered it as an exotic plant introduced by the Portuguese, and gave it names expressing its likeness to * P. Browne, Jamaica, p. 214; Jacquin, Ob.s., i. p. 38. * Acosta, Hist. Nat. des Indes., edit. 151)8, p. 176. ' Laet, Hist. Nouv. Monde, i. pp. 325, 341. * See the fine plates in Tussac's Flore des Antilles, iii. p. 45, pis. 10 and 11. The papaw belongs to the small family of the Fopayacece^Uxsed by some botanists into the Fassifiorce, and by others into the Bixacece. * R. Brown, Bat. of Congo, p. 52 ; A. de Candolle, Geogr. Bot. Rais.j p. 917. * Sagot, Journ. de la Soc. Centr- d'Hortic. de France^ 1872. ' Rumphius, Amhoin, i. p. 117. 294 ORIGIN OF CULTIVATED PLANTS. other species or its foreign extraction. Sloane,^ in the beginning of the eighteenth century, quotes several of his contemporaries, who mention that it was taken from the West Indies into Asia and Africa. Forster had not seen it in the plantations of the Pacific Isles at the time of Cook's voyages. Loureiro,^ in the middle of the eigh- teenth century, had seen it in cultivation in China, Cochin-China, and Zanzibar. So useful and so strikinff a plant would have been spread throughout the old world for thousands of years if it had existed there. Everything leads to the belief that it was introduced on the coasts of Africa and Asia after the discovery of America. All the species of the family are American. This one seems to have been cultivated from Brazil to the West Indies, and in Mexico before the arrival of the Europeans, since the earliest writers on the productions of the new world mention it.^ Marcgraf had often seen the male plant (always com- moner than the female) in the forests of Brazil, while the female plants were in gardens. Clusius, who was the first to give an illustration of the plant, says^ that his drawing was made in 1G07, in the bay of Todos Santos (province of Bahia). I know of no modern author who has confirmed the habitation in Brazil. Martins does not mention the species in his dictionary of the names of fruits in the language of the Tupis.^ It is not given as wild in Guiana and Columbia. P. Browne ^ asserts, on the other hand, that it is wild in Jamaica, and before his time Ximenes and Hernandez said the ^ame for St. Domingo and Mexico. Oviedo "^ seems to have seen the papaw in Central America, and he gives the common * Sloane, Jamaica^ p. 165. ^ Loureiro, Fl. Cock., p. 772. ' Marcgraf, Brasil., p. 103, and Piso, p. 159, for Brazil ; Ximenes in Marcgraf and Hernandez, Thesaurus, p. 99, for Mexico j and the last for St. Domingo and Mexico. * Clusius, Curce Posteriores, pp. 79, 80. ' Martins, Beitr. z. Ethnogr., ii. p. 418. * P. Browne, Jamaica, edit. 2, p. 360. The first edition is of 1756. ' The passage of Oviedo is translated into English by Correa de Mello and Spruce, in their paper on the Proceedings of the Linncean Society, x. p. 1. PLANTS CULTIVATED FOR THEIR FRUITS. 295 name olocoton for Nicaraofua. Yet Correa de Mello and Spruce, in their important article on the Papayacew, after having' botanized extensively in the Amazon region, in Peru and elsewhere, consider the papaw as a native of the West Indies, and do not think it is anywhere wild upon the Continent. I have seen ^ specimens from the mouth of the river Manatee in Florida, from Puebla in Mexico, and from Columbia, but the labels had no remark as to their wild character. The indications, it w^ill be noticed, are numerous for the shores of the Gulf of Mexico and for the West Indies. The habitation in Brazil which lies apart is very doubtful. Fis^ — Ficus carica, Linnaeus, The history of the fig presents a close analogy with that of the olive in point of origin and geographical limits. Its area as a wild species may have been extended by the dispersal of the seeds as cultivation spread. This seems probable, as the seeds pass intact through the diixestive oro-ans of men and animals. However, countries may be cited where the fig has been cultivated for a century at least, and where no such naturalization has taken place. I am not speaking of Europe noith of the Alps, where the tree demands particular care and the fruit ripens with difficulty, even the first crop, but of India for instance, the Southern States of America, Mauritius, and Chili, where, to judge from the silence of compilers of floras, the instances of quasi-wildness are rare. In our own day the fig tree grows wild, or nearly wild, over a vast region of which Syria is about the centre ; that is to say, from the east of Persia, or even from Afghanistan, across the whole of the Mediterranean region as far as the Canaries.^ From north to south this zone varies in width from the 25th to the 40th or 42nd parallel, according to local circumstances. As a rule, the fig stops like the olive at the foot of the Caucasus and the mountains of Europe which limit the Mediterranean * De Candolle, Prodr., xv. part 1, p. 414. * Boissier, Fl. Orient., iv. p. 1154 j Brandis, Forest Flora of India, p. 418; Webb and Berthelot, Hist. Nat. des Canaries, Botanuitie, iii. p. 257. 296 ORIGIN OF CULTIVATED PLANTS. basin, but it grows nearly wild on the south-west coast of France, where the winter is very mild.^ We turn to historical and philological records to see whether the area was more limited in antiquity. The ancient Egyptians called the fig teh,^ and the earliest Hebrew books speak of the fig, whether wild or culti- vated, under the name teenah^ which leaves its trace in the Arabic tin.^ The Persian name is quite different, unjir; but I do not know if it dates from the Zend. Piddington's Index has a Sanskrit name, itduinvara, which Roxburgh, who is very careful in such matters, does not give, and which has left no trace in modern Indian languages, to judge from four names quoted by authors. The antiquity of its existence east of Persia appears to me doubtful, until the Sanskrit name is verified. The Chinese received the fig tree from Persia, but only in the eighth century of our era.^ Herodotus ^ says the Persians did not lack figs, and Reynier, who has made careful researches into the customs of this ancient people, does not mention the fig tree. This only proves that the species was not utilized and cultivated, but it perhaps existed in a wild state. The Greeks called the wild fig erineos, and the Latins caprificus. Homer mentions a fig tree in the Iliad which grew near Troy.'' Hehn asserts ^ that the cultivated fig cannot have been developed from the wild fig, but all * Count Solms Laubach, in a learned discussion (RerTiunft, Domestica- tion, etc., des Feigenhaums, in 4to, 1882), has himself observed facts of this nature already indicated by various authors. He did not find the seed provided with embryos (p. 64), which he attributes to the absence of the insect (Blastophaga) , which generally lives in the wild fig, and facilitates the fertilization of one flower by another in the interior of the fruit. It is asserted, however, that fertilization occasionally takes place without the intervention of the insect. 2 Chabas, Melamjes EriyptoL, 3rd series (1873), vol. ii. p. 92. ' Kosenmuller, Bibl. Alterth., i. p. 285 ; Keynier, Econ. Puhl. des Arahes et des Juifs, p. 470. * Forskal, Fl. JE'jypto-Arah., p. 125. Lagarde (Revue Critique d'His- toire^ Feb. 27, 1882) says that this Semitic name is very ancient. * Bretschneider, in Solms, uhi snpra, p. 51. * Herodotus, i. 71. ' Lenz, Botanik der Griechen, p. 421, quotes four lines of Homer. See also Hehn, Culturpfianzen, edit. 3, p. 84. * Hehn, Culturpfianzen, edit. 3, p. 513. PLANTS CULTIVATED FOR THEIR FRUITS, 297 botanists hold a contrary opinion;^ and, without speaking of floral details on which they rely, I may say that Gussone obtained from the same seeds plants of the form caprijicus, and other varieties.^ The remark made by several scholars as to the absence of all mention of the cul- tivated fig sukai in the Iliad, does not therefore prove the absence of the fig tree in Greece at the time of the Trojan war. Homer mentions the sweet fig in the Odyssey, and that but vaguely. Hesiod, says Hehn, does not mention it, and Archilochus (700 B.C.) is the first to mention distinctly its cultivation by the Greeks of Paros. Accord- ing to this, the species grew wild in Greece, at least in the Archipelago, before the introduction of cultivated varieties of Asiatic origin. Theophrastus and Dioscorides mention wild and cultivated figs.^ Romulus and Remus, according to tradition, were nursed at the foot of a fig tree called ritniinalis, from rumen, breast or udder.^ The Latin name, jiciis, which Hehn derives, by an effort of erudition, from the Greek su^m,^also argues an ancient existence in Italy, and Pliny's opinion is positive on this head. The good cultivated varieties were of later introduction. They came from Greece, Syria, and Asia Minor. In the time of Tiberius, as now, the best figs came from the East. We learnt at school how Cato exhibited to the as- sembled senators Carthaginian figs, still fresh, as a proof of the proximity of the hated country. The Phoenicians must have transported good varieties to the coast of Africa and their other colonies on the Mediterranean, even as far as the Canaries, where, however, the wild fig may have already existed. For the Canaries we have a proof in the Guanchos * No importance slioulcl be attached to the exaggerated divisions made by Gasparini in Ficus carica, Linnceus. Botanists who have studied the fig tree since his time retain a single species, and name several varieties of the wild fig. The cultivated forms are numberless. ' Gussone, Enum. Plant. Inarimensium, p. 301. ' For the history of the fig tree and an account of the operation (of doubtful utility) which consists in planting insect-bearing Caprifici among the cultivated trees (caprification), see Solms' work. * Pliny, Hist., lib. xv. cap. 18. ^ Hehn, CuUurr'flanzen, edit. 3, p. 513. 298 ORIGIN OF CULTIVATED PLANTS. words, araliormaze and acJiormaze, green figs , tahare- onenen and teh'ihunemen, dried figs. Webb and Ber- thelot/ who quote these names, and who admit the common origin of the Guanchos and Berbers, would have noted with pleasure the existence among the Touaregs, a Berber people, of the word tahart, fig tree,^ and in the French-Berber dictionarj^, published since their time, the names tabeksist, green fig, and tagroiirt, fig tree. These old names, of more ancient and local origin than Arabic, bear witness to a very ancient habitation in the north of Africa as far as the Canaries. The result of our inquiry shows, then, that the prehistoric area of the fig tree covered the middle and southern part of the Mediterranean basin from Syria to the Canaries. We may doubt the antiquity of the fig in the south of France, but a curious fact deserves mention. Plan- chon found in the quaternary tufa of Montpellier, and de Saporta^ in those of Aygalades near Marseilles, and in the quaternary strata of La Celle near Paris, leaves and even fruit of the wild Ficus caHca, with teeth of Elephas prhnigeniios, and leaves of plants of which some no longer exist, and others, like Laiiriis canaricnsis, have survived in the Canaries. So that the fig tree perhaps existed in its modern form in this remote epoch. It is possible that it perished in the south of France, as it certainly did at Paris, and re- appeared later in a wild state in the southern region. Perhaps the fig trees which Webb and Berthelot had seen as old plants in the wildest part of the Canaries were descended from those which existed in the fourth epoch. Bread-Fruit — Artocarpus incisa, LinmBus. The bread-fruit tree was cultivated in all the islands of the Asiatic Archipelago, and of the great oceans near * Webb and Berthelot, Hist. Nat. des Canaries Ethnojr., p. 186; Phytorir.y iii. p. 257. * Duveyrier, Les Touaregs du Nord,, p. 193. ' Planchoti, Etude sur les tufs de Montpellier, p. 63 ; de Saporta, La flare des tufs quaternaires en Provence, in Co^iptes rendus de la 32e Session du Congres Scientijique de France; Bull. Soc. Geolog., 1873-74, p. U2. PLANTS CULTIVATED FOR THEIR FRUITS. 299 the equator, from Sumatra to the Marquesas Isles, when first Europeans began to visit them. Its fruit is con- stituted, like the pine-apple, of an assemblage of bracts and fruits welded into a fleshy mass, more or less spherical ; and as in the pine-apple, the seeds come to nothing in the most productive cultivated varieties.^ Sonnerat ^ carried the bread-fruit tree to Mauritius, where the Intendant Poivre took care to spread it. Captain Bligh was commissioned to introduce it into the English West Indian Isles. The mutiny of his crew prevented his succeeding the first time, but a second attempt proved more fortunate. In January, 1793, he landed 153 plants at St. Vincent, whence the species has been diffused into several parts of tropical America.^ Rumphius* saw the species wild in several of the Sunda Isles. Modern authors, less careful, or acquainted only with cultivated species, say nothing on this head. Seemann^ says for the Fiji Isles, "cultivated, and to all appearance wild in some places." On the continent of Asia it is not even cultivated, as the climate is not hot enough. The bread-fruit is evidently a native of Java, Am- boyna, and the neighbouring islands ; but the antiquity of its cultivation in the whole of the archipelago, proved by the number of varieties, and the facility of propa- gating it by buds and suckers, prevent us from knowing its history accurately. In the islands to the extreme east, like Otahiti, certain fables and traditions point to an introduction which is not very ancient, and the absence of seeds confirms this.^ Jack-Fruit — Artocarpus integrifolia, Linnaeus. The jack-fruit, larger than the bread-fruit, for it sometimes weighs as much as eighty pounds, hangs from ^ See the fine plates published in Tussac's Flore des Antilles, \6l. ii. pis. 2 and 3 ; and Hooker, Bot. Mag., t. 2869-2871. * Voyages a la Nouvelle Guine'e, p. 100. ' Hooker, uhi supra. * Rumphius, Herb. Amboinf i. p. 112, pi. 33. * Flora Vitiensis, p. 255. * Seemann, Fl. Vit, p. 255 ; Nadeand, Enum. des PI. Indig. de Taiti, p. 4i; Idem, PI. usuelles des Taitiens, p- 21, 14 300 OEIGIN OF CULTIVATED PLANTS. the branches of a tree thirty to fifty feet high.^ The common name is derived from the Indian names jaca, or tsjaka. The species has long been cultivated in southern Asia, from the Punjab to China, from the Himalayas to the Moluccas. It has not spread into the small islands more to the east, such as Otahiti, which leads us to sup- pose it has not been so long in the archipelago as upon the continent. In the north-west of India, also, its cultivation does not perhaps date from a very remote epoch, for the existence of a Sanskrit name is not abso- lutely certain. Roxburgh mentions one, piinusa, but Piddim^ton does not admit it into his Index. The Per- sians and the Arabs do not seem to have known the species. Its enormous fruit must, however, have struck them if the species had been cultivated near their fron- tiers. Dr. Bretschneider does not speak of any Arto- carpus in his work on the plants known to the ancient Chinese, whence it may be inferred that towards China, as in otiier directions, the jack-fruit was not diffused at a very early epoch. The first statement as to its exist- ence in a wild state is given by Rheede in ambiguous terms : " This tree grows everywhere in Malabar and throughout India." He perhaps confounded the planted tree with the wild one. After him, however, Wight found the species several times in the Indian Peninsula, notably in the Western Ghauts, with every appearance of a wild and indigenous tree. It has been extensively planted in Ceylon ; but Thwaites, the best authority for the flora of this island, does not recognize it as wild. Neither is it wild in the archipelago to the south of India, according to the general opinion. Lastly, Brandis found it growing in the forests of the district of Attaran, in Burmah, but, he adds, always in the neighbourhood of abandoned settlements. Kurz did not find it wild in British Buimah.^ * See Tussac's plates, Flore des Antilles, pi. 4; and Hooker, Bof. Mag., t. 2833, 2834. * Eheecle, Malabar, iii. p. 18 ; Wight, Icones, ii. No. 678 ; Brandis, Forest Flora of India, p. 426; Kurz, Forest Flora of Brit. Burmah, p. 432. PLANTS CULTIVATED FOR THEIR FRUITS. SOI The species is, therefore, a native of the region lying at the foot of the western mountains of the Indian Penin- sula, and its cultivation in the neighbourhood is probably not earlier than the Christian era. It was introduced into Jamaica by Admiral Rodney in 1782, and thence into San Domingo.^ It has also been introduced into Brazil, Mauritius, the Seychelles, and Rodriguez Island,^ Date-Palm — Fhwnix dactylifera, Linnseus. The date-palm has existed from prehistoric times in the warm dry zone, which extends from Senegal to the basin of the Indus, principally between parallels 15 and 30. It is seen here and there further to the north, by reason of exceptional circumstances and of the aim which is proposed in its cultivation. For beyond the limit within which the fruit ripens every year, there is a zone in which they ripen ill or seldom, and a further region within which the tree can live, but without fruitinor or even flowering. These limits have been traced by de Martins, Carl Ritter, and myself^ It is needless to repro- duce them here, the aim of the present work being to study questions of origin. As regards the date-palm, we can hardly rely on the more or less proved existence of really wild indigenous individuals. Dates are easily transported ; the stones germinate when sown in damp soil near the source of a river, and even in the fissures of rocks. The inhabitants of oases have planted or sown date-palms in favourable localities where the species perhaps existed before man, and when the traveller comes across isolated trees, at a distance from dwellings, he cannot know that they did not spring from stones thrown away by caravans. Botanists admit a variety, sylvcstris, that is to say wild, with small and sour fruit ; but it is perhaps the result of recent naturalization in an unfavourable soil. His- torical and philological data are of more value here, though doubtless from the antiquity of cultivation they can only establish probabilities. ' Tussac, Flore des Antilles, pi. 4. * Baker, Ft. of Mav.rif., p. 282. ' Martius, Gen. et Spec. Palmarum, in folio, vol. iii. p. 257 ; C Riitcr, Frdkunde, xiii. p. 760 ; Alph. de Candolle, Geog. Bot. Rais.y p. 343. 302 ORIGIN OF CULTIVATED PLANTS. From Egyptian and Assyrian remains, as well as from tradition and the most ancient writings, we find that the date-palm grew in abundance in the region lying between the Euphrates and the Nile. Egyptian monuments con- tain fruits and drawings of the tree.^ Herodotus, in a more recent age (fifth century before Christ), mentions the wood of the date-palms of Babylonia, and still later Strabo used similar expressions about those of Arabia, whence it seems that the species was commoner than it is now, and more in the condition of a natural forest tree. On the other hand, Carl Ritter makes the ingfenious observation that the earliest Hebrew books do not speak of the date-palm as producing a fruit valued as a food for man. David, about one thousand years before Christ, and about seven centuries aiter Moses, does not mention the date palm in his list of trees to be planted in his gardens. It is true that except at Jericho dates seldom ripen in Palestine Later, Herodotus says of the Baby- lonian date-pahns that only the greater part produced good fruit which was used for food This seems to indi- cate the beginning of a cultivation perfected by the selection of varieties and of the transport of male flowers into the middle of the branches of female trees, but it perhaps signifies also that Herodotus was ignorant of the existence of the male plant. To the west of Egypt the date-palm had probably existed for centuries or for thousands of years when Herodotus mentioned them. He speaks of Libya. There is no historical record with respect to the oases in the Sahara, but Pliny ^ mentions the date-palm in the Canaries. The names of the species bear witness to its great antiquity both in Asia and in Africa, seeing they are nume- rous and very difi^erent. The Hebrews called the date- palm tamar, and the ancient Egyptians beq.^ The com- plete difference between these words, both very ancient, shows that these peoples found the species indigenous and perhaps already named in Western A.^ia and in * Unger, Fflanzen d. Alt. JE^jypt., p. 38. 2 Plinj, Hist., lib. vi. cap. 37. ' linger, xihi snpra. PLANTS CULTIVATED FOR THEIR FRUITS. 303 Egypt. The number of Persian, Arabic, and Berber names is incredible.^ Some are derived from the Hebrew word, others from unknown sources. They often apply to different states of the fruit, or to different cultivated varieties, which aoain shows ancient cultivation in different countries. Webb and Berthelot have not dis- covered a name for the date-palm in the language of the Guanchos, and this is much to be regretted. The Greek name, fhaanix, refers simply to Phoenicia and the Phoenicians, possessors of the date-palm.^ The names dactyliis and date are derivations of dachel in a Hebrew dialect.^ No Sanskrit name is known, whence it may be inferred that the plantations of the date-palm in Western India are not very ancient. The Indian climate does not suit the species.* The Hindustani name khunna is borrowed from the Persian. Further to the East the date-palm remained long unknown. The Chinese received it from Persia, in the third century of our era, and its cultivation was resumed at different times, but they have now abandoned it.^ As a rule, beyond the arid region which lies between the Euphrates and the south of the Atlas and the Canaries, the date-palm has not succeeded in similar latitudes, or at least it has not become an important culture. It might be grown with success in Australia and at the Cape, but the Europeans who have colonized these regions are not satisfied, like the Arabs, with figs and dates for their staple food. I think, in fine, that in times anterior to the earliest Egyptian dynasties the date-palm already existed, wild or sown here and there by wandering tribes, in a narrow zone extending from the Eu[>hrates to the Canaries, and that its cultivation began later as far as the north-west of India on the one hand and the Ca|)e de Verde Islands ^ on the other, so that the natural area * See C. Eitter, ubi supra, * Hehn, CuUurpf.anzen, edit. 3, p. 234. * C. Ritter, ibid., p. 828. * According to Roxburgh, Royle, etc. * Bretschneider, Stridy avd Valuer etc., p. 31. * According to Schmidt, Fl. d. Cap.-Verd. IsL, p. 1G8, the dato- pjilm is rare in these islands, and is certainly not wild. Webb and Berthelot, on the contrary, assert that in some of the Canaries it ia apparently indigenous {Hist. Nat. des CanaHes^ Botaniquej iii. p. 28S). 80Jj ORIGIN OF CULTIVATED PLANTS. has remained very nearly the same for about five thou- sand years. What it was previously, palneontological discoveries may one day reveal. Banana — 3Iasa sapientum and M. paradisiaca, Linn?eus ; M. sapientu7n, Brown. The banana or bananas were generally considered to be natives of Southern Asia, and to have been carried into America by Europeans, till Humboldt threw doubts upon their purely Asiatic origin. In his work on New Spain ^ he quoted early authors who assert that the banana was cultivated in America before the conquest. He admits, on Oviedo's authority,^ its introduction by Father Thomas of Berlangas from the Canaries into San Domingo in 15 IG, whence it was introduced into other islands and the mainland.^ He recoofnizes the absence of any mention of the banana in the accounts of Columbus, Alonzo Negro, Pinzon, Vespuzzi, and Cortez. The silence of Hernandez, who lived half a century aftei- Oviedo, astonishes him and appears to him a remarkable carelessness; *'for,"he says,* "it is a constant tradition in Mexico and on the whole of the mainland that the 2)latano avion, and the doininico were cultivated lono- before the Spanish conquest." The author who has most carefully noted the different epochs at which American agriculture has been enriched by foreign pro- ducts, the Peruvian Garcilasso de la Vega,^ says dis- tinctly that at the time of the Incas, maize, quinoa, the potato, and, in the warm and temperate regions, bananas formed the staple food of the natives. He describes the Miisa of the valleys in the Andes ; he even distinguishes the rarer species, with a small fruit and a sweet aromatic flavour, the doininico, from the connnon banana or arton. * Ilumbolclt, Nouvelle Esx)ajne, 1st edit., ii. p. 3G0. 2 Oviedo, Hist. Nat., 1556, p. 112. Oviedo's first work is of 1526. He is the earliest naturalist quoted by Drjander (Bihl. Eatilcs) lor America. ' I have also seen this passage in the translation of Oviedo by Eamusio, vol. iii. p. 115. * Humboldt, Nouvelle Espagne, 2nd edit., p. ?>So. ® Garcilasso de la Vega, Commentarios Eeales, i. p. 2S2. PLANTS CULTIVATED FOR THEIR FRUITS. 305 Father Acosta^ asserts also, although less positively, that the Miisa was cultivated by the Americans before the arrival of the Spaniards. Lastly, Humboldt adds from his own observation, " On the banks of the Orinoco, of the Cassiquaire or of the Beni, between the mountains of Esmeralda and the banks of the river Carony, in the midst of the thickest forests, almost everywhere that Indian tribes are found who have had no relations with European settlements, we meet with plantations of Manioc and bananas." Humboldt suggests the hypothesis that several species or constant varieties of the Banana have been confounded, some of which are indigenous to the new world. Desvaux studied the specific question, and in a really remarkable work, published in 1814,^ he gives it as his opinion that all the bananas cultivated for their fruits are of the same species. In this species he distinguishes forty -four varieties, which he arranges in two groups ; the large-fruited bananas (seven to fifteen inches long), and the small-fruited bananas (one to six inches], commonly called fig bananas. R. Brown, in 1818, in his work on the Plants of the Congo, p. 51, maintains also that no structural difference in the bananas cultivated in Asia and those in America prevents us from considering them as belonging to the same species. He adopts the name Mihsa sajnentuni, which appears to me preferable to that of M. jparadisiaca adopted by Desvaux, because the varieties with small fertile fruit appear to be nearer the condition of the wild Musce found in Asia. Brown remarks on the question of origin that all the other species of the genus Miisa belong to the old world ; that no one pretends to have found in America, in a wild state, varieties with fertile fruit, as has happened in Asia; lastly, that Piso and Marcgraf considered that the banana was introduced into Brazil from Conofo. In spite of the force of these three arguments, Humboldt, in his second edition of his essay upon New Spain (ii. p. 897), does not entirely renounce his opinion. He * Acosta, Hist. Nat. Be Indian, 1608, p. 250. * Desvaux, Journ. Bot., iv. p. 5. 306 OPJaiN" OF CULTIVATED PLANTS. says that the traveller Caldcleugh ^ found among the Puris the tradition that a small species of banana was cultivated on the borders of the Prato long before they had any communications with the Portuguese. He adds that words which are not borrowed ones are found in American lano-uasfes to distinofuish the fruit of the Miisa: for instance, 'parwrii in Tamanac, etc., arata in Maypur. I have also read in Stevenson's travels^ that beds of the leaves of the two bananas commonly cultivated in America have been found in the huacas or Peruvian tombs anterior to the conquest; but as this traveller also says that he saw beans ^ in these huacas, a plant which undoubtedly belongs to the old world, his asser- tions are not very trustworthy. Boussingault ^ thought that the ^jlafano arton at least was of American origin, but he gives no proof. Meyen, who had also been in America, adds no argument to those which were already known ;^ nor does the geographer E-itter,^ who simply reproduces the facts about America, given by Humboldt. On the other hand, the botanists who have more recently visited America have no hesitation as to the Asiatic origin. I may name Seemann for the Isthmus of Panama, Ernst for Venezuela, and Sagot for Guiana.'^ The two first insist upon the absence of names for the banana in the languages of Peru and Mexico. Piso knew no Brazilian name. Martins ^ has since indicated, in the Tupi language of Brazil, the names imcoha or hacohcL This same word hacove is used, according to Sagot, by the French in Guiana. It is perhaps derived from the name hala, or ixdan, of Malabar, from an intro- duction by the Portuguese, subsequent to Piso's voyage. The antiquity and wild character of the banana in Asia are incontestable facts. There are several Sanskrit ' Caldcleugh, Trav. in S. Amer., 1825, i. p. 23. * Stevenson, Trav. in 8. Amer., i. p. 328. 3 Ibid., p. 363. * Boussingault, C. r. Acad. Sc. Paris, May 9, 1836. * Mcyen, Tfianzen Geog., 1836, p. 383. « Ritter, Erdk., iv. p. 8/0. ^ Seemann, Bot. of the Herald, p. 213 ; Ernst, in Seemann's Journ. o/BoL, 1867, p. 289; Sagot, Journ. de la Soc. d'Hort. de Fr., 1872, p. 226. ' Martins, Eth. Sprachenkunde Amer., p. 123. PLANTS CULTIVATED FOR THEIR FRUITS. 307 names.^ The Greeks, Latins, and Arabs have mentioned it as a remarkable Indian fruit tree. Pliny ^ speaks of it distinctly. He says that the Greeks of the expedi- tion of Alexander saw it in India, and he quotes the name pala which still persists in Malabar. Sages re- posed beneath its shade and ate of its fruit. Hence the botanical name Musa scqnentum. Mwsa is from the Arabic ttiouz or viauiuz, which we find as early as the thirteenth century in Ebn Baithar. The speciiic name IJavadisiaca comes from the ridiculous hypothesis which made the banana figure in the story of Eve and of Paradise. It is a curious fact that the Hebrews and the ancient Egyptians^ did not know this Indian plant. It is a sign that it did not exist in India from a very remote epoch, but was first a native of the Malay Archipelago. There is an immense number of varieties of the banana in the south of Asia, both on the islands and on the continent ; the cultivation of these varieties dates in India, in China, and in the archipelago, from an epoch impossible to realize ; it even spread formerly into the islands of the Pacific * and to the west coast of Africa ; ^ lastly, the varieties bore distinct names in the most separate Asiatic languages, such as Chinese, Sanskrit, and Malay. All this indicates great antiquity of culture, consequently a primitive existence in Asia, and a difiu- sion contemporary with or even anterior to that of the human races. The banana is said to have been found wild in several places. This is the more worthy of attention since the cultivated varieties seldom produce seed, and are multiplied by division, so that the species can hardly have become naturalized from cultivation by sowing itself lioxburgh had seen it in the forests of Chittagong,^ in * Roxbureh and Wallich, Ft. Iml., ii. p. 485 ; Pidtlington, Index. ^ Pliny, Hist., lib. xii. cap. 6. ' Unger, nhi supra, and Wilkinson, ii, p. 403, do not mention it. The banana is now cultivated in Egypt. •* Foister, Plant. Esc, p. 28. * Clusius, Exot., p. 229; Brown, Bot. Congo, p. 51. ® Roxburgh, Coram., tab. 275 ; Fl. Ind. 303 ORIGIN OF CULTIVATED PLANTS. the form of Miisa sainentum. Kiimphius^ describes a wild variety with small fruits in the Philippine Isles. Loureiro^ probably speaks of the same form by the name M seminifera agrestis, -which he contrasts with 31. semiiiifera doniestica, which is w^ild in Cochin-Cliina. Blanco also mentions a wild banana in the Philippines/ but his description is vague. Finlayson^ found the banana wild in abundance in the little island of Pulo Ubi at the southern extremity of Siam. Thwaites ^ saw the variety M. sapientinn in the rocky forests of the centre of Ceylon, and does not hesitate to pronounce it the original stock of the cultiv^ated bananas. Sir Joseph Hooker and Thomson'' found it wild at Khasia. The facts are quite different in America. The wild banana has been seen nowhere except in Barbados,^ but here it is a tree of w^hich the fruit does not ripen, and w^hich is, consequently, in all probability the result of cultivated varieties of which the seed is not al)undant. Sloane's icild plantain ^ appears to be a ])lant very different to the musa. The varieties which are supposed to be possibly indigenous in America are only two, and as a rule far fewer varieties are grown than in Asia. The culture of the banana may be said to be recent in the greater part of i\m erica, for it dates from but little more than three centuries. Piso ^^ says positively that it was imported into Brazil, and has no Brazilian name. He does not say whence it came. We have seen that, according to Oviedo, the species w^as brought to San Domingo from the Canaries. This fact and the silence of Hernandez, generally so accurate about the useful plants, wild or cultivated, in Mexico, convince me that at the time of the discovery of America the banana did not exist in the whole of the eastern part of the continent. J Eumphius, Amh., v. p. 139. * Loureiro, Fl. Coch., p. 791. 3 Loureiro, Fl. Coch., p. 791. ■* Blanco, Flora, 1st edit., p. 2-17. Fiiilayson, Journey to Siam, 182G, p. 86, according to Eitter, Erdk., iv. 878. P-Thwaites, Emim. PI. Cey., p. 321. ' Aitchison, Catal. of Punjab, p. 1-17. ^ Hughes, Barb., p. 182 ; Majcock, Fl. Barb., p 396. ® Sloane, Jamai a, ii. p. 148. " Piso, edit. 16'i8, Hist. Nat, p. 7«5. PLANTS CULTIVATED FOR THEIR FRUITS. 309 Did it exist, then, in the western part on the shores of the Pacific ? This seems very unlikely when we reflect that communication was easy between the two coasts towards the isthmus of Panama, and that before the arrival of the Europeans the natives had been active in diffusing throughout America useful plants like the manioc, maize, and the potato. The banana, which they have prized so highly for three centuries, which is so easily multiplied by suckers, and whose appearance must strike the least observant, would not have been forgotten in a few villages in the depths of the forest or upon the littoral. I admit that the opinion of Garcilasso, descendant of the Incas, an author who lived from 1530 to 1568, has a certain importance when he says that the natives knew the banana before the conquest. However, the expressions of another writer, extremely worthy of attention, Joseph Acosta, who had been in Peru, and whom Humboldt quotes in support of Garcilasso, incline me to adopt the contrary opinion.^ He says,^ " The reason the Spaniards called it plane (for the natives had no such name) was that, as in the case of their trees, they found some resemblance between them." He goes on to show how different was the plane (Plafanus) of the ancients. He describes the banana very well, and adds that the tree is very common in the Indies (i.e. America), " although they (the Indians) say that its origin is Ethiopia. . . . There is a small white species of plantain (banana), very delicate, which is called in Espagnolle ^ dominico. There are others coarser and larger, and of a red colour. There are none in Peru, but they are imported thither from the Indies,^ as ' Huttiboldt quotes the Spanish edition of 1608. The first edition is of 1591. I have onlj been able to consult the French translation of Eegnault, published in 1598, and which is apparently accurate. ^ Acosta, trans., lib. iv. cap. 21. ' That is probably Hispaniola or San Domingo ; for if he had meant the Spanish language, it would have been translated by castillan and without the capital letter. * This is probably a misprint for Andes, for the word Indes hns no sense. The work says (p. 166) that pine-apples do not grow in Peru, but that they are brought thither from the Andes, and (p. 173) that the cacao comes from the Andes. It seems to have meant hot regions. The wrod 310 OEIGIN OF CULTIVATED PLANTS. ' into Mexico from Cuernavaca and the other valleys. On the continent and in some of the islands there are great plantations of them which form dense thickets." Surely it is not thus that the author would express himself were he writing of a fruit tree of American origin. He would quote American names and customs ; above all, he would not say that the natives regarded it as a plant of foreign origin. Its diffusion in the warm regions of Mexico may well have taken place between the epoch of the conquest and the time when Acosta wrote, since Hernandez, whose conscientious researches go back to the earliest times of the Spanish dominion in Mexico (though published later in Rome), says not a word of the banana.^ Prescott the historian saw ancient books and manuscripts which assert that the inhabitants of Tumbez brouofht bananas to Pizarro when he disembarked upon the Peruvian coast, and he believes that its leaves were found in the huacas, but he does not give his proofs.^ As regards the argument of the modern native plantations in regions of America, remote from European settlements, I find it hard to believe that tribes have remained absolutely isolated, and have not received so useful a tree from colonized districts. Briefly, then, it appears to me most probable that the species was early introduced by the Spanish and Portu- guese into San Domingo and Brazil, and I confess that this implies that Garcilasso was in error with regard to Peruvian traditions. If, however, later research sliould prove that the banana existed in some parts of America before the advent of the Europeans, I should be inclined to attribute it to a chance introduction, not very ancient, the effect of some unknown communication with the islands of the Pacific, or with the coast of Guinea, rather than to believe in the primitive and simultaneous existence Andes "has since been applied to the chain of mountains by a strange and unfortunate transfer*. * I have read through the entire work, to make sure of this fact. ^ Prescott, Conquest of Peru. The author has consulted valuable records, among others a manuscript of Montesinos of 1527; but he does not quote his authorities for each fact, and contents himself with vague and general indications, which are very insutficient. PLANTS CULTIVATED FOR THEIR FRUITS* 311 of the species in both hemispheres. The whole of geo- graphical botany renders the latter hypothesis improbable, I might almost sa}^ impossible, to admit, especially in a genus which is not divided between the two worlds. In conclusion, I would call attention to the remarkable way in which the distribution of varieties favours the opinion of a single species — an opinion adopted, purely from the botanical point of view, by Roxburgh, Desvaux, and R. Brown, If there were two or three species, one would probably be represented by the varieties suspected to be of American origin, the other would belong, for instance, to the Malay Archipelago or to China, and the third to India. On the contrary all the varieties are geographically intermixed, and the two which are most widely diffused in America differ sensibly the one from the other, and each is confounded with or approaches very nearly to Asiatic varieties. Pine- Apple — Ananassa sativa, Lindley; Bromelia Ananas, Linnseus. In spite of the doubts of a few writers, the pine- apple must be an American plant, early introduced by Europeans into Asia and Africa Nana was the Brazilian name,^ which the Portue^uese turned into ananas. The Spanish called it pinas, because the shape i^sembles the fruit of a species of pine.^ AH early writers on America mention it.^ Hernandez says that the pine-apple grows in the warm regions of Haiti and Mexico. He mentions a Mexican name, matzatli A pine-apple was brought to Charles V., who mistrusted it, and would not taste it. The works of the Greeks, Romans, and Arabs make no allusion to this species, which was evidently introduced into the old world after the discovery of America. Rheede * in the seventeenth century was persuaded of this ; but Rumphius ^ disputed it later, because he said ^ Marcf^raf, Brasil., p. 33. ^ Oviedo, Eaniusio's trans., iii. p. 113 ; Jos. Acosta, Hist. Nat. des. Jndes, French trans., p. 166. ' Thevet, Piso, etc. ; Hernandez, Thes., p. 341. * Rheede, Hort. Ma^.ah., xi. p. 6. * Rumphins, Amhcin, v. p. 228. 312 ORIGIN OF CULTIVATED PLANTS. the pine-apple was cultivated in liis time in every part of India, and was found wild in Celebes and elsewhere. He notices, however, the absence of an Asiatic name. That given by Rheede for Malabar is evidently taken from a comparison with the jack-fruit, and is in no sense original. It is doubtless a mistake on the part of Piddington to attribute a Sanskrit name to the pine-apple, as the name anariish seems to be a corruption of ananas. Roxburgh knew of none, and Wilson's dictionary does not mention the word anariish. Royle ^ says that the pine-apple was introduced into Bengal in 1594". Kirch er ^ says that the Chinese cultivated it in the seventeenth century, but it was believed to have been brought to them from Peru. Clusius^ in 1599 had seen leaves of the pine-apple brought from the coast of Guinea. This may be explained by an introduction there subsequent to the discovery of America. Robert Brown speaks of the pine-apple among the plants cultivated in Congo; but he considers the species to be an American one. Although the cultivated pine-apple bears few seeds or none at all, it occasionally becomes naturalized in hot countries. Examples are quoted in Mauritius, the Seychelles, and Rodriguez Island,* in India,^ in the Malay Archipelago, and in some parts of America, where it was pi'obably not indigenous — the West Indies, for instance. It has been found wild in the warm regions of Mexico (if we may trust the phrase used by Hernandez), in the province of Veraguas ^ near Panama, in the upper Orinoco valley,'^ in Guiana ^ and the province of Bahia.^ ^ Eoyle, III., p. 376. ' Kircher, Chine Illustrde, trans, of 1G70, p. 253. ' Clusiu.s, Exotic, cap. 4i. * Baker, Fl. of Maurit. ^ Royle, iihi supra. " Seemann, Br>t. of the Herald, p. 215. ' Humboldt, Nouv. Esp., 2ud edit., ii. p. 478. * Gardeners' Chronicle, 1881, vol. i. p. 657. " Mai-tius, letter to A. de Candolle, GSojr, Bot. Bais , p. 927. CHAPTER Y. PLANTS CULTIVATED FOR THEIR SEEDS. Article I. — Seeds used for Food. Cacao — Theohroma Cacao, Linnreus. The genus Theohroina, of the order Bi/ttneriacece, allied to the Malvacece, consists of fifteen to eighteen species, all belonging to tropical America, principally in the hotter parts of Brazil, Guiana, and Central America. The common cacao, Theohroma Cacao, is a small tree wild in the forests of the Amazon and Orinoco basins ^ and of their tributaries up to four hundred feet of alti- tude. It is also said to grow wild in Trinidad, which lies near the mouth of the Orinoco.^ I find no proof that it is indigenous in Guiana, although it seems probable. Many early writers indicate that it was both wild and cultivated at the time of the discovery of America from Panama to Guatemala and Campeachy ; but from the numerous quotations collected by Sloane,^ it is to be feared that its wild character was not sufficiently verified. Modern botanists are not very explicit on this head, and in general they only mention the cacao as cultivated in these regions and in the West India Islands. G. Ber- noulli,* who had resided in Guatemala, only says, " wild * Humboldt, Voy., ii. p. 511 ; Kunth, in Humboldt and Bonpland, Nova Genera, v. p. 31G ; Martius, Ueher den Cacao, in Biichner, Rcpert. Pharm. * Schach, in Grisebach, Flora of Brit. W. Ind. Is., p. 91. ' Sloane, Jamaica, ii. p. 15. 4 G. Bernoulli, Uebersicht der Arten von Theohroma, p. 5. 314! ORIGIN OF CULTIVATED PLANTS. and cultivated throngliout tropical America;" anclHemsley,^ in his review of the plants of Mexico and Central America, made in 1879 from the rich materials of the Kewherbarium, gives no locality where the species is indigenous. It was perhaps introduced into Central America and into the warm regions of Mexico by the Indians before the dis- covery of America. Cultivation may have naturalized it here and there, as is said to be the case in Jamaica.^ In support of this hypothesis, it must be observed that Triana^ indicates the cacao as only cultivated in the warm regions of New Granada, a country situated be- tween Panama and the Orinoco valley. However this may be, the species was grown in Central America and Yucatan at the time of the dis- covery of America. The seeds were sent into the high- lands of Mexico, and were even used as money, so highly were they valued. The custom of drinking chocolate was general. The name of this excellent drink is Mexi- can. The Spaniards carried the cacao from Acapulco to the Philippine Isles in lG74and 1680,^ where it succeeded wonderfully. It is also cultivated in the Sunda Isles. I imagine it would succeed on the Guinea and Zanzibar coasts, but it is of no use to attempt to grow it in countries which are not very hot and very damp. Another species, Theohroma hicolor, Humboldt and Bonpland, is found growing with the common cacao in American plantations. It is not so much prized. On the other hand, it does not require so high a temperature, and can live at an altitude of nearly three thousand feet in the valley of the Magdalena. It abounds in a wild state in New Granada.^ Bernoulli asserts that it is only cultivated in Guatemala, though the inhabitants call it mountain cacao. Litchi — XejyJielium Litchi, Cambessides. The seed of this species and of the two following is * Hemsley, Biologia Centrali Americana, part ii. p. 133. ^ Grisebach, ^lbi supra. ' Triana and Planchon, Prodr. Fl. Koro Granatensis, p. 208. * Blanco, Fl. de Filipinas, edit. 2, p. 420. * Kunth, in Humboldt aud Bonpland, uhi supra ; Triana, vbi tsvpra. PLANTS CULTIVATED FOR THEIR SEEDS. 815 covered with a fleshy excrescence, very sweet and scented, which is eaten with tea. Like most of the SapindacecfJ, the nepheliums are trees. This one has been cultivated in the south of China, India, and the Malay Archipelago from a date of which we cannot be certain. Chinese authors living at Pekin only knew the Litchi late in the third century of our era.^ Its introduction into Bengal took place at the end of the eighteenth century.^ Every one admits that the species is a native of the south of China, and, Blume^ adds, of Cochin-China and the Philippine Isles, but it does not seem that any botanist has found it in a truly wild state. This is probably because the southern part of China towards Siam has been little visited. In Cochin- China and in Burmah and at Chittagong the Litchi is only cultivated.^ Longan — Nej^heliitm longana, Cambessides. This second species, very often cultivated in Southern Asia, like the Litchi, is wild in British India, from Ceylon and Concan as far as the mountains to the east of Bengal, and in Pegu.^ The Chinese introduced it into the Malay Archipelago some centuries ago. Hambutan — ])fej)heliitm lajpidciceii'in, Linnaeus. It is said to be wild in the Indian Archipelago, where it must have been long cultivated, to judge from the number of its varieties. A Malay name, given by Blume, signifies wild tree. Loureiro says it is wild in Cochin- China and Java. Yet I find no confirmation for Cochin- China in modern works, nor even for the islands. The new flora of British India ^ indicates it at Singapore and Malacca w^ithout aflirming that it is indigenous, on which head the labels in herbaria commonly tell ns nothing. Certainly the species is not wild on the continent of Asia, in spite of the vague expressions of Blume and » Bretsclmeidev, letter of Aug. 23, 1881. * Roxbiu-gh, Fl. Indica, ii. p. 269. ' Blume, Bnmphia, iii. p. 106. * Loureiro, Flora Coch., p. 233 ; Kurz, Forest Fl. of Brit. Burmah, p. 293. * Roxburgh, Fl. Ind., ii. p. 271 ; Thwaites, Enum. ZeyL, p. 58 ; Hiern, in Fl. of Brit. Ind., i. p. 688. « Hiern, in Fl. of Brit. Ind., i. p. 687. 810 OHIGIN OF CULTIVATED PLANTS. Miquel,^ but it is more probably a native of the Malay- Archipelago. In spite of the reputation of the nepheliums, of which the fruit can be exported, it does not appear that these trees have been introduced into the tropical colonies of Africa and America except into a few gardens as curiosities. Pistachio Nut — Pistacia vera, Linnseus. The pistachio, a shrub belonging to the order Ana- cardiacece, grows naturally in Syria. Boissier ^ found it to the north of Damascus in Anti-Lebanon, and he saw specimens of it brought from Mesopotamia, but he could not be sure that they were found wild. There is the same doubt about branches gathered in Arabia, which have been mentioned by some writers. Pliny and Galen^ knew that the species was a Syrian one. The former tells us that the plant was introduced into Italy by Vitellius at the end of the reign of Tiberius, and thence into Spain by Flavins Pompeius There is no reason to believe that the cultivation of the pistachio was ancient even in its primitive country, but it is practised in our own day in the East, as well as in Sicily and Tunis. In the south of France and Spain it is of little importance. Broad Bean — Faba vulgaris, Moench ; Vicia fdba, Linnaeus. Linnaeus, in his best descriptive work, Hortus cliff or- tianiis, admits that the origin of this species is obscure, like that of most plants of ancient cultivation. Later, in his Species, which is more often quoted, he says, with- out giving any proof, that the bean "inhabits Egypt." Lerche, a Russian traveller at the end of the last centurv, found it wild in the Mun^an desert of the Mazanderan, to the south of the Caspian Sea."* Travellers * Blume, Rumphia, iii. p. 103 ; Miquel, Fl. Indo-Batava, i. p. 554. * Bossier, Fl. Orient., ii. p. 5. ^ Pliny, Hist. Nat., lib. xiii. cap. 15; lib. xv. cap. 22; Galen, De Ali- Tuentis, lib. ii. cap. 30. * Lerche, Nora Acta Acad. Cesareo. Leopold, vol. v., appendix, p. 203, published in 1773. Maxiinowicz, in a letter of Feb. 2-i, 1882, tells me that Lerche's specimen exists in the herbarium of the Imperial Garden PLANTS CULTIVATED FOR THEIR SEEDS. 317 who have collected in this region have sometimes come across it/ but they do not mention it in their writings,^ excepting Ledebour,^ and the quotation on which he relies is not correct. Eosc ^ says that Olivier found the bean wild in Persia ; I do not find this confirmed in Olivier's Voyage, and as a rule Bosc seems to have been too ready to believe that Olivier found a good many of our cultivated plants in the interior of Persia. He says it of buckwheat and of oats, which Olivier does not mention. The only indication besides that of Lerche which I find in floras is a very difterent locality. Munby mentions the bean as wild in Algeria, at Oran. He adds that it is rare. No other author, to my knowledge, has spoken of it in northern Africa. Cosson, who knows the flora of Algeria better than any one, assures me he has not seen or received any specimen of the wild bean from the north of Africa. I have ascertained that there is no specimen in Munby 's ^ herbarium, now at Kew. As the Arabs grow the bean on a large scale, it may perhaps be met with accidentally outside cultivated plots. It must not be forgotten, however, that Pliny (lib. xviii. c. 12) speaks of a wild bean in Mauritania, but he adds that it is hard and cannot be cooked, which throws doubt upon the species. Botanists who have written upon Egypt and Cyrenaica, especially the more recent,^ give the bean as cultivated. This plant alone constitutes the genus Faha. We cannot, therefore, call in the aid of any botanical analogy at St. Petersbnrgh. It is in flower, and resembles the cultivated bean in all points excepting height, which is about half a foot. The label mentions the locality and its wild character without other remarks. * There are Transcaucasian specimens in the same herbarium, but taller, and they are not said to be wild. ^ Marschall Bieberstein, Flora Caucai^o-Taurica ; C. A. Meyer, Fe?*- zeicliniss ; Hohenacker, Eniim. Plant. Talysch ; Boisdier, Fl. Orient., p. 578, Buhse and Boissier, Plant. Tra)iscaucasice. * Ledebour, Ft. Ross., i. p. 664, quotes de Candolle, Prodromus, ii. p. 354 ; now Seringe wrote the article Faha in Prodi'omus, in which the south of the Caspian is indicated, probably on Lerche's authority. * Diet. d'Agric.^ v. p. 512. * Munby, Catal. Plant, in Alger, sponte nascent., edit. 2, p. 12. * Schweinf orth and Ascherson, Aufzdhlung, p. 256 j Eohlfs, Kvfra. 318 ORIGIN OF CULTIVATED PLANTS. to discover ifcs origin. We must have recourse to the history of its cultivation and to the names of the species to find out the country in which it was originally indigenous. We must first eliminate an error which came from a wrong interpretation of Chinese works. Stanislas Julien believed that the bean was one of the five plants which the Emperor Chin-nong commanded, 4600 years ago, to be sown every year with great solemnity.^ Now, accord- ing to Dr. Bretschneider,^ who is surrounded at Pekin with every possible resource for arriving at the truth, the seed similar to a bean which the emperors sow in the enjoined ceremony is that of Dollchos soja, and the bean was only introduced into China from Western Asia a century before the Christian era, at the time of Chang- kien s embassy. Thus falls an assertion which it is hard to reconcile with other facts, for instance with the absence of an ancient cultivation of the bean in India, and of a Sanskrit name, or even of any modern Indian name. The ancient Greeks were acquainted with the bean, which they called huamos, and sometimes kuamos ellenikos, to distinguish it from that of Egypt, which was the seed of a totally different aquatic species, Nelum- hiii7n. The Iliad ^ already mentions the bean as a culti- vated plant, and Virchow found some beans in the excavations at Troy."* The Latins called it faba. We find nothing in the works of Theophrastus, Dioscorides, Pliny, etc., which leads us to believe the plant indigenous in Greece or Italy. It was early known, because it was an ancient Roman rite to put beans in the sacrifices to the goddess Carna, whence the name Faharice Calendce.^ The Fabii perhaps took their name from faha, and the twelfth chapter of the eighteenth book of Pliny shows, without the possibility of a doubt, the antiquity and importance of the bean in Italy. * Loiscleur Deslongchamps, Consid. sur les Cereales, part i. p. 29. 2 Bretschneider, Study and Valuer etc., pp. 7, 15. s Iliad, 13, V. 589. * Wittraack, Sitz. hericht Vereins, Brandenburg, 1879, . * Novitius Dictionnarium, at the word Faba. PLANTS CULTIVATED FOR THEIR SEEDS. 319 The word faha recurs in several of the Aryan lan- guages of Europe, but with modifications which philolo- gists alone can recognize. We must not forget, however, Adolphe Pictet's very just remark,^ that in the cases of the seeds of cereals and leguminous plants the names of one species are often transferred to another, or that cer- tain names were sometimes specific and sometimes generic. Several seeds of like form were called kuamos by the Greeks; several different kinds of haricot bean {Pha~ seolus,Dolichos) bear the same name in Sanskrit, and /a5a- in ancient Slav, hohu in ancient Prussian, hobo in Armo- rican, few, etc., may very well have been used for peas, haricot beans, etc. In our own day the phrase co&QQ-hean is used in the trade. It has been rightly supposed that when Pliny speaks of faharice islands, where beans were found in abundance, he alludes to a species of wild pea called botanically Pisuni maritimnum. The ancient inhabitants of Switzerland and of Italy in the age of bronze cultivated a small-fruited variety of Faha vulgaris.'^ Heer calls it Celtica nana, because it is only six to nine millimetres long, whereas our modern field bean is ten to twelve millimetres. He has compared the specimens from Montelier on Lake Morat, and St. Peter's Islands on Lake Bienne, with others of the same epoch from Parma. Mortellet found, in the contem- porary lake-dwellings on the Lake Bourget, the same small bean, which is, he says, very like a variety culti- vated in Spain at the present day.^ The bean was cultivated by the ancient Egyptians.* It is true that hitherto no beans have been found in the sarcophagi, or drawings of the plant seen on the monu- ments. The reason is said to be that the plant was reckoned unclean.^ Herodotus ^ says, " The Egyptians ^ Orirjines Tndo-Europcc7iy\es, edit. 2, vol. i. p. 35^. * Heer, PJlanzen der Ffahlhauten, p. 22, figs. 44-1-7. ^ Perrin, ^tude Prehistorique sur la Savoie, p. 2. * Delile, Plant. Cult, en £gypte, p. 12 ; Reynier, ^conomxe des Egyp. Uens et Carthaginois, p. 340; Unger, Pjlan. d. Alt. ^gyp., p. Glj Wilkin- son, Man. and Cus. of Anc. Egyptians, p. 402. * Reynier, uhi supra, tries to discover the reason of thia. * Herodotus, Histoire, Larclier's trans., vol. ii. p. 32. 320 ORIGIN OF CULTIVATED PLANTS. never sow the bean in their land, and if it grows they do not eat it either cooked or raw. The priests cannot even endure the sight of it; they imagine that this vegetable is unclean." The bean existed then in Egypt, and probably in cultivated places, for the soil which would suit it was as a rule under cultivation. Perhaps the poor population and that of certain districts did not share the prejudices of the priests ; we know that the superstitions varied with the nor}ies. Plutarch and Diodorus Siculus mention the cultivation of the bean in Egypt, but they wrote five hundred years later than Herodotus. The word 2^ol occurs twice in the Old Testament ; ^ it has been translated bean because of the traditions pre- served by the Talmud, and of the Arabic name foul, fol, or fid, which is that of the bean. The first of the two verses shows that the Hebrews were acquainted with the bean one thousand ^^ears before Christ. Lastly, I shall mention a sign of the ancient existence of the bean in the north of Africa. This is the Berber name ihiou, in the plural iahoiven, used by the Kabyles of the province of Algiers.^ It has no resemblance to the Semitic name, and dates perhaps from a remote antiquity. The Berbers formerly inhabited Mauritania, where Pliny asserts that the species was wild. It is not known whether the Guanchos (the Berber people of the Canaries) knew the bean. I doubt whether the Iberians had it, for their supposed descendants, the Basques, use the name haha,^ answering to the Roman faha. We judge from these facts that the bean was culti- vated in Europe in prehistoric terms. It was introduced into Europe probably by the western Aryans at the time of their earliest migrations (Pelasgians, Kelts, Slavs). It was taken to China later, a century before the Christian era, and still later into Japan, and cjuite recently into India. Its wild habitat was probably twofold some thousands of years ago, one of the centres being to the south of the ]^ 2 Sam. xvii. 28 ; Ezek. iv. 9. " Diet. Franrais-Berhere, published by the French government. ' Note communicated to M. Clos by M. d'Abadie. PLANTS CULTIVATED FOR THEIR SEEDS. 321 Caspian, the other in the north of Africa. This kind of area, which I have called disjunctive, and to which I formerly paid a good deal of attention,^ is rare in dicoty- ledons, but there are examples in those very countries of which I have just spoken.^ It is probable that the area of the bean has long been in process of diminution and of extinction. The nature of the plant is in favour of this hypothesis, for its seed has no means of dispersing itself, and rodents or other animals can easily make prey of it. Its area in Western Asia was probably less limited at one time, and that in Africa in Pliny's day was more or less extensive. The struggle for existence which was going against this plant, as against maize, would have gradually isolated it and caused it to disappear, if man had not saved it by cultivation. The plant which most nearly resembles the bean is Vicia narhonensis. Authors who do not admit the genus Faha, of which the characters are not very distinct from those of Vicia, place these two species in the same section. Now, Vicia narhonensis is wild in the Mediterranean basin and in the East as far as the Caucasus, in the north of Persia, and in Mesopotamia.^ Its area is con- tinuous, but this renders the hypothesis I mentioned above probable by analogy. Lentil — Erviivi lens, Linnaeus ; Lens esculenta, Moench. The plants which most nearly resemble the lentil are classed by authors now in the genus Ervum, now in a distinct genus Lens, and sometimes in the genus Cicer ; but the species of these ill-defined groups all belong to the Mediterranean basin or to Western Asia. This throws some light on the origin of the cultivated plant. Unfortunately, the lentil is no longer to be found in a wild state, at least with certainty. The floras of the south of Europe, of Northern Africa, of the East, and of India always mention it as cultivated, or as growing in fields after or with other cultivated species. A botanist * * A. de Candolle, G6ogr. Bof. Rais., chap. x. ' Rhododendron pontimm now exists only in Asia Minor and in the S( nth of the Spanish peninsula. ' Boissier, Fl. Orient., ii. p. 577. * C. A. Meyer, Vcrzeichniss Fl. Caucas., p. 147. 322 ORIGIN OF CULTIVATED PLANTS. saw it in tlie provinces to the south of the Caucasus, "cultivated and nearly wild here and there round vil- lages." Another ^ indicates it vaguely in the south of Russia, hut more recent floras fail to confirm this. The history and names of this plant may give clearer indications of its orioin. It has been cultivated in the East, in the Mediterranean basin and even in Switzerland, from prehistoric time. According to Herodotos, Theo- phrastus, etc., the ancient Egyptians used it largely. If their monuments give no proof of this, it was probably because the lentil was, like the bean, considered common and coarse. The Old Testament mentions it three times, by the name adaschiim or adascJmn, which must cer- tainly mean lentil, for the Arabic name is ads,^ or adas.^ The red colour of Esau's famous mess of pottage has not been understood by most authors. Reynier,^ who had lived in Egypt, confirms the explanation given formerly by Josephus; the lentils were red because they were hulled. It is still the practice in Egypt, says Reynier, to remove the husk or outer skin from the lentil, and in this case they are a pale red. The Berbers have the Semitic name ades for the lentil.^ The Greeks cultivated the species — -fahos or falcai. Aristophanes mentions it as an article of food of the poor.^ The Latins called it le'iis, a name whose origin is unknown, which is evidently allied to the ancient Slav leshay Illyrian lechja, Lithuanian lenszic? The differ- ence between the Greek and Latin names shows that the species perhaps existed in Greece and Italy before it was cultivated. Another proof of ancient existence in Europe is the discovery of lentils in the lake-dwellings of St. Peter's Island, Lake of Bienne,^ which are of the age of * Georg]*, in Ledebour, Fl. Boss. ^ Forskal, Fl. jEgypt. ; Delile, Plant. Cult, en ^gypte, p. 13. 3 Ebn Baithar, i"i. p. 134. * 'Re y mer, £conomie ptihlique et rurale des Arahes et desJuifs, Geneve, 1820, p.'429. * Did. Frav^.-Berhere, in 8vo, 1844. * Hehn, Culturpflanzen, etc., edit. 3, vol. ii. p. 188. ^ Ad. Pictet, Origines Indo-Europeennes, edit. 2, vol. i. p. 36 i; Hehn, uhi sxipra. * Heer, Pjianzen der Pfahlhaufen, p. 23, fig. 49. PLANTS CULTIVATED FOR THEIR SEEDS. 323 bronze. The species may have been introduced from Italy. According to Theophrastus,^ the inhabitants of Bac- triana (the modern Bokkara) did not know the fahos of the Greeks. Adolphe Pictet quotes a Persian name, Tnangu or Tiiargw, but he does not say whether it is an ancient name, existing, for instance, in the Zend Avesta. He admits several Sanskrit names for the lentil, masuya, renuka, riiangalya, etc., while Anglo-Indian botanists, Koxburgh and Piddington, knew none.^ As these authors mention an analogous name in Hindustani and Bengali, mussour, we may suppose that masiira signifies lentil, while mangw in Persian recalls the other name mangalya. As Roxburgh and Piddington give no name in other Indian languages, it may be supposed that the lentil was not known in this country before the invasion of the Sanskrit-speaking race. Ancient Chinese works do not mention the species; at least. Dr. Bretschneider says nothing of them in his work published in 1870, nor in the more detailed letters which he has since written to me. The lentil appears to have existed in western tem- perate Asia, in Greece, and in Italy, where its cultivation was first undertaken in very early prehistoric time, when it was introduced into Egypt. Its cultivation appears to have been extended at a less remote epoch, but still hardly in historic time, both east and west, that is into Europe and India. Chick-Pea — Cicer arietinurti, Linnreus. Fifteen species of the genus Cicer are known, all ot Western Asia or Greece, except one, which is Abyssinian. It seems, therefore, most probable that the cultivated species comes from the tract of land lying between Greece and the Himalayas, vaguely termed the East. The species has not been found undoubtedly wild. All the floras of the south of Europe, of Egypt, and of Western Asia as far as the Caucasus and India, give it as a cultivated species, or growing in fields and cultivated ' Theoplirastn«, Hist., lib. iv. cap. 5. 2 Roxburgh, Fl. Jnd., edit- 1832, ygl, jii, p, 324} Piddington, Index. 15 824? ORIGIN O^ CULTIVATED PLANTS. grounds. It has sometimes^ been indicated in the Crimea, and to the north, and especially to the south of the Caucasus, as nearly wild ; but well-informed modern authors do not think so.^ This quasi- wildness can only point to its origin in Armenia and the neighbouiing countries. The cultivation and the names of the species may perhaps throw some light on the question. The Greeks cultivated this species of pea as early as Homer's time, under the name of erehinthos,^ and also of krios,'^ from the resemblance of the pea to the head of a ram. The Latins called it cicer, which is the origin of all the modern names in the south of Europe. The name exists also among the Albanians, descendants of the Pelasofians, under the form kikere.^ The existence of such widely difierent names shows that the plant was very early known, and perhaps indigenous, in the south- east of Europe. The chick-pea has not been found in the lake-dwell- ings of Switzerland, Savo}^, and Italy. In the first- named locality its absence is not singular ; the climate is not hot enough. A common name among the peoples of the south of the Caucasus and of the Caspian Sea is, in G eor gisin, nachiida ; in Turkish and Armenian, 7iac/m(s, nachunt ; in Persian, nochot.^ Philologists can tell if this is a very ancient name, and if it has any connection with the Sanskrit chennuka. The chick-pea is so frequently cultivated in Egypt from the earliest times of the Christian era,'^ that it is supposed to have been also known to the ancient Egyptians. There is no proof to be found in the draw- ings or stores of grain in their monuments, but it may be supposed that this pea, like the bean and the lentil, was * Ledebour, Fl. Boss., i. p. 6G0, according to Pallas, Falk, and Kocli. * Boissier, Fl. Orient., ii. p. 5G0; Steven, Verzeic}inii>s des Tauriachen Hahlinseln, p. 134. 3 Iliad, bk. 13, verse 589 ; Theophrastus, Hist., lib. viii. c. 3. * Dioscorides, lib. ii. c. 126. * Heldreich, Kutzpjlaiizen Griechenlavds, p. 71. « Nemnich, Polyjlott. Lex., i. p. 1037 ; Bunge, in Goehels Reise, ii. p. 328. ' Clement d' Alexandria, Strom., lib. i., quoted from Eeynier, £con. des Egiip. et Carthag.f p. 343. PLANTS CULTIVATED FOR THEIR SEEDS. 325 considered common or unclean. Reynier^ thought that the ketscch, mentioned by Isaiah in the Old Testament, was perhaps the chick-pea; but this name is generally attributed, though without certainty, to JSigella sativa or Vicia sativa.^ As the Arabs have a totally different name for the chick-pea, omnos, homos, which recurs in the Kabyl language as hamviez,^ it is not likely that the ketsech of the Jews was the same plant. These de- tails lead me to suspect that the species was unknown to the ancient Egyptians and to the Hebrews. It was perhaps introduced among them from Greece or Italy towards the beginning of our era. It is of more ancient introduction into India, for there is a Sanskrit name, and several others, analogous or different, in modern Indian languages.* Bretschneider does not mention the species in China. I do not know of any proof of antiquity of culture in Spain, yet the Castilian name garhanzo, used also by the Basques under the form garlantziia, and by the French as garvance, being neither Latin nor Arabic, may date from an epoch anterior to the Roman conquest. Botanical, historical, and philological data agree in indicatincf a habitation anterior to cultivation in the countries to the south of the Caucasus and to the north of Persia. The western Aryans (Pelasgians, Hellenes) perhaps introduced the plant into Southern Europe, where, however, there is some probability that it was also indigenous. The western Aryans carried it into India. Its area perhaps extended from Persia to Greece, and the species now exists only in cultivated ground, Avhere we do not know whether it springs from a stock originally wild or from cultivated plants. Lupin — Livpinus albus, Linnams. The ancient Greeks and Romans cultivated this leguminous plant to bury it as a green manure, and also ' Eeynier, ^con. des Arahes et Juifs, p. 430. ^ Rosenmiiller, Bihl. Alterth., i. p. 100 ; Hamilton, Bot. de la Bible, p. 180. ' Eauwolf, i^. Orient., No. 220; Forskal, Fl. JEgypf., p. 81 j Diet. Frang.-Berhere. * Roxburgh, Fl. Ind., iii. p. 324 ; PicldiDgton, Index. S26 ORIGIi^ OF CULTIVATED PLANTS. for the sake of the seeds, which are a good fodder for cattle, and which are also used by man. The expressions of Theophrastus, Dioscorides, Cato, Varro, Pliny, etc., quoted by modern writers, refer to the culture or to tlie medical properties of the seeds, and do not show whether the species was the white lupin, L, albus, or the blue- flowered lupin, L. hirsutus, which grows wild in the south of Europe. Fraas says ^ that the latter is grown in the ]\[orea at the present day ; but Heldreich says ^ that L. albus grows in Attica. As this is tlie species which has been long cultivated in Italy, it is probable that it is the lupin of the ancients. It was much grown in the eighteenth century, especially in Italy,^ and de I'Ecluse settles the question of the species, as he calls it Lupimts sativus albo floret The antiquity of its cultivation in Spain is shown by the existence of four different common names, according to the province ; but the plant is only found cultivated or nearly wild in fields and sandy places.^ The species is indicated by Bertoloni in Italy, on the hills of Sarzana. Yet Caruel does not believe it to be wild here, any more than in other pai'ts of the peninsula.^ Gussone^ is very positive for Sicily — "on barren and sandy hills, and in meadows (in herhidis)" Lastly, Grisebach^ found it in Turkey in Europe, near Ruskoi, and d'Urville ^ saw it in abundance, in a wood near Constantinople. Castagne confirms this in a manu- script catalogue in my possession. Boissier does not men- tion any locality in the East ; the species does not exist in India, but Russian botanists have found it to the south of the Caucasus, thousfh we do not know witli certainty if it was really wild.^*^ Other localities will perhaps be found between Sicily, Macedonia, and the Caucasus. * See Froas, Fl. Clapis., p. 51 ; Lenz., Bot. der Altei}, p. 73. 2 Hddreicii, Nutzpjlanzen Griechenlands, p. 69. 3 Olivier do Serres, Theatre de VAgric, edit. 1529, p. 88. * Clnsius, Hist. Plant, ii. p. 228. * Willko-mm and Lange, FJ. Hisp., iii, p. 4G6. ' Caruel, FL Toscana, p. 136. ^ Gnssone, Fl. Sicxdco Syn., edit. 2, Vol. ii. p. ^jG. « Cri3cbach, Spicil. FL Ilumal., p. 11. » D'ljiville, Fnum.f p. f6. *" Ledebour, FL Boss., i. p. 510. PLANTS CULTIVATED FOR THEIR SEEDS. S27 Egyptian Lupin — Lupiniis termis, Forskal. This species of lupin, so nearly allied to L. alhus that it has sometimes been proposed to unite them/ is largely cultivated in Egypt and even in Crete. The most obvious difference is that the upper part of the flowers of L. termis is blue. The stem is taller than that of L. alhus. The seeds are used like those of the common lupin, after they have been steeped to get rid of their bitterness. L. terviis is wild in sandy soil and mountainous dis- tricts, in Sicily, Sardinia, and Corsica ; ^ in Syria and Egypt, according to Boissier ; ^ but Schweinf urth and As- cherson * say that it is only cultivated in Egypt. Hart- mann saw it wild in Upper Egypt.^ linger ^ mentions it among the cultivated specimens of the ancient Egyp- tians, but he gives neither specimen nor drawing. Wil- kinson "^ says only that it has been found in the tombs. No lupin is grown in India, nor is there any Sanskrit name ; its seeds are sold in bazaars under the name tourinus (Royle, III., p. 194). The Arabic name, tennis or terraiis, is also that of the Greek lupin, termos. It may be inferred that the Greeks had it from the Egyptians. As the species was knoAvn to the ancient Egyptians, it seems strange that it has no Hebrew name ; ^ but it may have been introduced into Egypt after the departure of the Israelites. Field-Pea — Fiswin arvense, Linnseus. This pea is grown on a large scale for the seed, and also sometimes for fodder. Although its appearance and botanical characters allow of its being easily distinguished from the garden-pea, Greek and Koman authors con- founded them, or are not explicit about them. Their writings do not prove that it was cultivated in their time. It has not been found in the lake-dweUings of » Caruel, Fl. Tosc, p. 136. 2 Gussone, Fl. Sic. Syn., ii. p. 267; Moris, Fl. SardoOy i. p. 596. 3 Boissier, Fl. Orient., ii. p. 29. * Avfzdhlung, etc., p. 257. * Schweinfurth, Plantce Nilot. a Hartman Coll., p. 6. ® Unger, Pflanzen d. Alt. JEgyp., p. 65. ^ Wilkinson, Manners and C ustoms of the Ancient Fgyjptians,u.Y>.403. * Rosenmiiller, Bibl. Alterth., vol. i. 328 ORIGIN OF CULTIVATED PLANTS. Switzerland, France, and Italy. Bobbio has a legend (a.d. 930), in which it is said that the Italian peasants called a certain seed herhUia, whence it has been sup- posed to be the modern ruhiglia or the Pisuvi sativum of botanists.^ The species is cultivated in the East, and as far as the north of India.^ It is of recent cultivation in the latter country, for there is no Sanskrit name, and Piddington gives only one name in one of the modern lano'uaoes. Whatever may be the date of the introduction of its culture, the species is undoubtedly wild in Italy, not only in hedges and near cultivated ground, but also in forests and wild mountainous districts.^ I find no positive indication in the floras that it grows in like manner in Spain, Algeria, Greece, and the East. The plant is said to be indigenous in the south of Russia, but some- times its wild character is doubtful, and sometimes the species itself is not certain, from a confusion with Pisuni sativum and P. elatius. Of all Anglo-Indian botanists, only Royle admits it to be indigenous in the north of India. Garden-Pea — Pisum sativutn, Linngeus. The pea of our kitchen gardens is more delicate than the field-pea, and satfers from frost and drought. Its natural area, previous to cultivation, was probably more to the south and more restricted. It has not hitherto been found wild, either in Europe or in the west of Asia, whence it is supposed to have come. Bieberstein's indica- tion of the species in the Crimea is not correct, according to Steven, who was a resident in the country.* Perhaps botanists have overlooked its habitation ; perhaps the plant has disappeared from its original dwelling ; perhaps also it is a mere modification, efiected by culture, of Pisunn arvense, Alefeld held the latter opinion,^ but he * Mnratori, Antich. Ital.^ i. p. 347; Diss., 24, quoted by Targioni, Cenni Sfon'ct, p. 31. 2 Boissier, FL Orient, ii. p. 623 ; Eoyle, III. Himal, p. 200. ' Bei'toloni, FL Ital., vii. p. 419; Caruel, Ft. Tosc, p. 184; Gussonc, Fl. Sic. Si/nopsis, ii. p. 279 ; Moris, FL Sardoa, i. p. 577. •* Steven, Verzeichniiois de Congo, pois pigeon. It is remarkable that, though the species is difiused in three continents, the varieties are not numerous. Two are cited, based only upon the yellow or reddish colour of the flower, which were formerly regarded as distinct species; but a more attentive examination has resulted in their beinof classed as one, in accordance with Linnaeus' opinion.*^ The small number of variations obtained even in the organ for which the species is cultivated is a sign of no ver}'' ancient culture. Its habitation previous to culture is uncertain. The best botanists have sometimes supposed it to be a native of India, sometimes of tropical * Soja angustifoUa, Miqnel ; see Hooker, Fl. Brit. Ind., ii. p. 184. 2 Kumphius, Amh., vol. v. p. 3SS. * Tussac, i^Zore des Antilles, \o\. iv. p. 94, pi. 32; Grisebach, 17. o/ Brit. W. Indies, i. p. 191. * See Wight and Arnott, Prod. FI. Penins. Ind., p. 256 ; Klotzsch, in Peters, Reise nach 3Iozarnhique, i. p. 36. The yellow variety is figured in Tussac, that with the red flowers in the Botanical Register, 1845, pi. 31. PLANTS CULTIVATED FOR THEIR SEEDS. 333 Africa. Bentham, who has made a careful study of the leguminous plants, believed in 18G1 in the African origin ; in 1865 he inclined rather to Asia.^ The problem is, therefore, an interesting one. There is no question of an American origin. The cajan was introduced into the West Indies from the coast of Africa by the slave trade, as the common names quoted above show,^ and the unanimous opinion of authors or American floras. It has also been taken to Brazil, Guiana, and into all the warm parts of the American continent. The facility with which the species is naturalized, would alone prevent attaching great importance to the statements of collectors, who have found it more or less wild in Asia or in Africa; and besides, these assertions are not precise, but are usually doubtful. Most writers on the flora of continental India have only seen the plant cultivated,^ and none, to my knowledge, affirms that it exists wild. For the island of Ceylon Thwaites says,* " It is said not to be really wild, and the country names seem to confirm this." Sir Joseph Hooker, in his Flora of British India, says, "Wild (?) and cultivated to the height of six thousand feet in the Himalayas." Loureiro ^ gives it as cultivated and non-cultivated in China and Cochin-China. Chinese authors do not appear to have spoken of it, for the species is not named by Bretschneider in his work On the Study, etc. In the Sunda Isles it is mentioned as cultivated, and that rarely, at Amboyna at the end of the eighteenth century, according to Rum- phius.^ Forster had not seen it in the Pacific Isles at the time of Cook's voyages, but Seemann says that it has been recently introduced by missionaries into the Fiji Isles.'' All this argues no very ancient extension of cul- tivation to the east and south of the continent of Asia. Besides the quotation from Loureiro, I find the species * Bentham, Flora HonglcongensiSf p. 89 ; Flora Bradl., vol. xy. p. 190; Bentham and Hooker, i. p. 541. * Tussac, Flore des Antilles ; .Tacquin, Ohs., p. 1. ' Rheede, Roxburgh, Kurz, Burni. FL, etc. * Thwaites, Enum, PL Ceylan. * Loureiio, FL CocJ i *., p. 5G5. * Rurnphius, Amb., vol. v. t. 135. " Seemann, FL Vitiensis, p. 74. 334i ORIGIN OF CULTIVATED PLANTS. indicated on the mountain of Magelang, Java ; ^ but, sup- posing this to be a true and ancient wild growth in both cases, it would be very extraordinary not to find the species in many other Asiatic localities. The abundance of Indian and Malay names ^ shows a somewhat ancient cultivation. Piddington even gives a Sanskrit name, arhuku, which was not known to Kox- burgh, but he gives no proof in support of his assertion. The name may have been merely supposed from the Hindu and Bengali names urur and orol. No Semitic name is known. In Africa the cajan is often found from Zanzibar to the coast of Guinea.^ Authors say it is cultivated, or else make no statement on this head, which would seem to show that the specimens are sometimes wild. In Egypt this cultivation is quite modern, of the nineteenth century.^ Briefly, then, I doubt that the species is really wild in Asia, and that it has been grown there for more than three thousand years. If more ancient peoples had known it, it would have come to the knowledge of the Arabs and Egyptians before our time. In tropical Africa, on the contrary, it is possible that it has existed wild or culti- vated for a very long time, and that it was introduced into Asia by ancient travellers trading between Zanzibar and India or Ceylon. The genus Cajanus has only one species, so that no analogy of geographical distribution leads us to believe it to be rather of Asiatic than African origin, or vice versa. Carob Tree^ — Ceratonia siliqua, Linnseus. The seeds and pods of the carob are highly prized in the hotter parts of the Mediterranean basin, as food for animals and even for man. De Gasparin ^ has given in- * JungTmhn, Plantce Jungh., fasc. i. p. 241. ' Piddington, Index ; Eheede, Malah., vi. p. 23, etc. ' Pickering, Chron. Arrang. of Plants, p. 442; Peters, Reise, p. 36; E. Brown, Bot. of Congo, p. 53 ; Oliver, Fl. of Trop. Afr., ii. p. 216. * Bulletin de la Societe d' Acclimation, 1871, p. 663. ^ The species is given here in order not to separate it from the other leguminous plants cultivated for the seeds alone. * De Gasparin, Cours. d'Agric, iv. p. 3'i8. PLANTS CULTIVATED FOR THEIR SEEDS- 335 teresting details about the raising, uses, and habitation of the species as a cultivated tree. He notes that it does not pass the northern limit beyond which the orange cannot be grown without shelter. This fine evergreen tree does not thrive either in very hot countries, especially where there is much humidity. It likes tlie neighbour- hood of the sea and rocky places. Its original country, according to Gasparin, is " probably the centre of Africa. Denham and Clapperton found it in Burnou." This proof seems to me insufficient, for in all the Nile Valley and in Abyssinia the carob is not wild nor even culti- vated.^ R. Brown does not mention it in his account of Denham and Clapper ton's journey. Travellers have seen it in the forests of Cyrenaica between the high-lands and the littoral ; but the able botanists who have drawn up the catalogue of the plants of this country are careful to say,^ "perhaps indigenous." Most botanists merely mention the species in the centre and south of the Medi- terranean basin, from Spain and Marocco to Syria and Anatolia, without inquiring closely whether it is indi- genous or cultivated, and without entering upon the question of its true country previous to cultivation. Usually they indicate the carob tree, as " cultivated and subspontaneous, or nearly wild." However, it is stated to be wild in Greece by Heldreich, in Sicily by Gussone and Bianca, in Algeria by Manby ; ^ and these authors have each lived long enough in the country for which each is quoted to form an enlightened opinion. Bianca remarks, however, that the carob tree is not always healthy and productive in those restricted localities where it exists in Sicily, in the small adjacent islands, and on the coast of Italy. He puts forward the opinion, moreover, based upon the similarity of the Italian name carriiho with the Arabic word, that the species was * Schweinfarthand Ascherson,^w/3a/iZMn(7, p. 255 ; Richard, Tentamen Fl. Ahyss. 2 Ascherson, etc., in Rohls, Kufra, 1 vol. in 8vo, 1881, p. 519. ' Heldreich, Nutzpjlanzen Griecheiilands, p. 73 ; Die Pdanzen der Attischen Ebene, p. 477 ; Gussone, Syn. Fl. Sic, p. 646 ; Bianca, II Carrtiho, in the Giornale d' Agricoltura Italiana, 1881 ; Muubj, Catal. PI. in Alg. Spont., p. 13. 8oG ORIGIN OF CULTIVATED PLANTS. anciently introduced into the south of Europe, the species being of Syrian or north African origin. He maintains as probahle the theory of Ha3fer and Bonne,^ that the lotus of th3 lotophagi was the carob tree, of which the flower is sweet and the fruit has a taste of honey, which agrees with the expressions of Homer. The lotus-eaters dwelt in Cyrenaica, so that the carob must have been abundant in their country. If we admit this hypothesis we must suppose that Pliny and Herodotus did not know Homer's plant, ibr the one describes the lotos as bearing a fruit like a mastic berry {Pistacia lentiscus), the other as a deciduous tree.^ An hypothesis regarding a doubtful plant formerly mentioned by a poet can hardly serv^e as the basis of an argument upon facts of natural history. After all, Homer's lotus plant perhaps existed only in the fabled garden of Hesperides. I return to more serious argu- ments, on which Bianca has said a few words. The carob has two names in ancient lancfuaofes — the one Greek, keraunia or kerateia;^ the other Arabic, ddrnnh or chai-ub. The first alludes to the form of the pod, which is like a slightly curved horn ; the other means merely pod, for we find in Ebn Baithar's ^ work that four other leguminous plants bear the same name, with a quali- fying epithet. The Latins had no special name ; they used the Greek word, or the expression siliqiva, siliqua grceca (Greek pod).^ This dearth of names is the sign of a once restricted area, and of a culture which probably does not date from prehistoric time. The Greek name is still retained in Greece. The Arab name persists among the Kabyles, who call the fiTiit IJiarrouh, the tree takhar- roiit,^ and the Spaniards algarroho. Curiously enough, * Hoefer, Hist. Bot. Miner, et Geol., 1 vol. in 12mo, p. 20; Bonne, Le Carouhier, on VArhre des Lotopharjes, Algiers, 18G9 (quoted by Uoefer). See above, the article on the jujube tree. ^ Pliny, Hist., lib. i. cap. 30. ' Theophrastus, Hist. Plant., lib. i, cap. 11; Dioscoritles, lib. i. cap. 155; Fraas, Syn. Fl. Class., p. 65. * Ebn Baithar, German trans., i. p. 354 ; Forskal, Fl. JE^ypt., p. 77. " Columna, quoted by Lenz, Bot. der Alten, p. 73 ; Pliny, Hid., lib. xiii. cap. 8. * Did. Fran^'.-Berbere, at the Avord Carouhe. PLANTS CULTIVATED FOR THEIR SEEDS. 337 the Italians also took the Arab name curraho, caruhio, whence the French carouhier. It seems that it must have been introduced after the Roman epoch by the Arabs of the Middle Ages, when there was another name for it. These details are all in favour of Bianca's theory of a more southern origin than Sicily. Pliny says the species belonged to Syria, Ionia, Cnidos, and Rhodes, but he does not sav whether it was wild or cultivated in these places. Piiny also says thafc the carob tree did not exist in Egypt. Yet it has been recocfnized in monuments belonoino; to a much earlier epoch than that of Pliny, and Egyptologists even attribute two Egyptian names to it, kontrates or jiri} Lepsius gives a drawing of a pod which appears to him to be certainly a carob, and the botanist Kotschy made certain by microscopic investigation that a stick taken from a sarcophagus was made from the wood of the carob tree.^ There is no known Hebrew name for the species, which is not mentioned in the Old Testament. The New Testament speaks of it by the Greek name in the ])arable of the prodigal son. It is a tradition of the Christians in the East that St. John Baptist fed upon the fruit of the carob in the desert, and hence came the names given to it in the Middle Ages — bread of St. John, and Johannis brcdhaum. Evidently this tree became important at the beginning of the Christian era, and it spread, especiall}^ througli the agency of the Arabs, towards the West. If it had previously existed in Algeria, among the Berbers, and in Spain, older names would have persisted, and the species would probably have been introduced into the Canaries by the Pha\nicians. The information gained on the subject may be summed up as follows : — Tlie carob grew wild in the Levant, probably on the southern coast of Anatolia and in Syria, perhaps also in * Lexicon Oxon., qunterl by Pickering', Chron. Hist, of Plants, p. 141. ' The drawing is reproduced in Unger's Vjianzen des ^Iten yTjijiiptcns, fig. 22. The observation which he quotes from Kotschy needs coufirma- «iou by a special anatomist. 338 ORIGIN OF CULTIVATED PLANTS. Cyrenaica. Its cultivation began within historic time. The Greeks diffused it in Greece and Italy ; but it was afterwards more highly esteemed by the Arabs, who propagated it as far as Marocco and Spain. In all these countries the tree has become naturalized here and there in a less productiv^e form, which it is needful to graft to obtain good fruit. The carob has not been found in the tufa and quater- nary deposits of Southern Europe. It is the only one of its kind in the genus Ceratonia, which is somewhat exceptional among the Lcgiiminoscc, especially in Europe. Nothing shows that it existed in the ancient tertiary or quaternary flora of the south-west of Europe. Common Haricot Kidney Bean — Phaseolns vulgaris, Savi. When, in 1855, I wished to investigate the origin of the genera Phaseolus and Doliclios,^ the distinction of species was so little defined, and the floras of tropical countries so rare, that I was obliged to leave several questions on one side. Now, thanks to the works of Bentham and Georg von Martens,^ completing the previous labours of Savi,^ the Legumince of hot countries are better known ; lastly, the seeds discovered quite recently in the Peruvian tombs of Ancon, examined by Wittmack, have completely modified the question of origin. I will speak first of the common haricot bean, after- wards of some other species, without, however, enume- rating all those which are cultivated, for several of these are still ill defined. Botanists held for a long time that the common haricot was of Indian origin. No one had found it wild, nor has it yet been found, but it was supposed to be of Indian origin, although the species was also cultivated in Afiica and America, in temperate and hot regions, at least in those where the heat and humidity are not excessive. I called attention to the fact that there is > A. de Caudolle, Gt'offr. Bof. Rais., p. 961. ' Bentham, in Ann. Wieyier Museum, vol. ii. ; Martens, Die Garten- hohnen, in 4to, Stuttgart, 1860, edit. 2, 1869. ' Savi, Ossei'v. sopra Phaseolus e Dolichos, 1, 2, 3. PLANTS CULTIVATED FOR THEIR SEEDS. 839 no Sanskrit name, and that sixteenth-century gardeners often called the species Turkish hean. Convinced, more- over, that the Greeks cultivated this plant under the names fasiolos and dolichos, I suggested that it came originally from Western Asia, and not from India. Georg von Martens adopted this hypothesis. However, the meaning of the words dolichos of Theophrastus, fasiolos of Dioscorides, faseolus and johaseolus of the Romans,^ is far from being sufficiently defined to allow them to be attributed with certainty to Phaseolus vulgaris. Several cultivated Leguniinosce are supported by the trellises mentioned by authors, and have pods and seeds of a similar kind. The best argu- ment for translating these names by Phaseolus vulgaris is that the modern Greeks and Italians have names derived from fasiolus for the common haricot. In modern Greek it is fasoulia, in Albanian (Pelasgic ?) fasule, in Italian fagiolo. It is possible, however, that the name has been transferred from a species of pea or vetch, or from a haricot formerly cultivated, to our modern haricot. It is rather bold to determine a species of Phaseolus from one or two epithets in an ancient author, when we see how difficult is the distinction of species to modern botanists with the plants under their eyes. Nevertheless, the dolichos of Theophrastus has been definitely referred to the scarlet runner, and the fasiolos to the dwarf haricot of our gardens, which are the two principal modern varieties of the common haricot, with an immense number of sub-varieties in the form of the pods and seed. I can only say it may be so. If the common haricot was formerly known in Greece, it was not one of the earliest introductions, for the faseolos did not exist at Rome in Cato's time, and it is only at the beginning of the empire that Latin authors speak of it. Yirchow brought from the excavations at Troy the seeds of several legumintie, which Wittmack ^ * Theophrastus, Hist., lib. viii. cap. 3; Dioscorides, lib. ii. cap. 130; Pliny, Hist., lib. xviii. cap. 7, 12, interpreted by Fraas, Syn. Fl. Class.^ p. 52 ; Leuz, Bot. der Alien, p. 731 ; Martens, Die Gartenhohnen, p. 1. ' Wittmack, Bot. Vereins Brandenburg, Dec. 19, 1879. 840 ORIGIN OF CULTIVATED PLANTS. has ascerfcained to belong to the following species : broad bean (Faha vulgaris), gSbYden'-pesi (Fisum sativum), ervilla (Ervum ervilia), and perhaps the flat-podded vetchling (Lathyrus Cicera), but no haricot. Nor has the species been found in the lake-dwellings of Switzerland, Savoy, Austria, and Italy. There are no proofs or signs of its existence in ancient Egypt. No Hebrew name is known answering to the Fhaseolus or Dolichos of botanists. A less ancient name, for it is Arabic, loubia, exists in Egypt for Dolichos luhia, and in Hindustani as loha for Fhaseolus vulgaris} As regards the latter species, Piddington only gives two names in modern languages, and those both Hindustani, loha and hakla. This, together with the absence of a Sanskrit name, points to a recent introduction into Southern Asia. Chinese authors do not mention F. vulgaris,^ which is a further indication of a recent introduction into India, and also into Bactriana, whence the Chinese have imported plants from the second century of our era. All these circumstances incline me to doubt whether the species Avas known in Asia before the Christian era. The argument based upon the modern Greek and Italian names for the haricot, derived from fasiolos, needs some support. It may be said in its favour that it was used in the Middle Ages, probably for the common haricot. In the list of veiietables which Charlemame commanded to be sown in his farms, we find fasioliun,^ without ex- planation. Albertus Magnus describes under the name faseolus a leguminous plant which appears to be our dwarf haricot.* I notice, on the other hand, that writers * Delile, Plantes Cultivees en £[jypfe, p. 14; Piddington, Index. * Bretschneider does not mention anv, either in his pamphlet On the Study and Value of Chinese Botanical Works, or in his private letters to me. ' E. Meyer, Geschichte der Botanique, iii. p. 404. * " Faseolus est species leguminis et yrani, quod est in quant itate paruni 7}iinus quani Faba, et in figura est columnare sicut faha, herhaque ejus minor est aliquantulum quam herha Fahrp. Et sunt faseoli tnultorum coloruni, sed quodlihet granoimm habet maculani nigram ni loco cotyledonis" (Jessen, Alberti Magni, De Vegetabilihus, edit, critica, p. 515). PLANTS CULTIVATED FOR THEIR SEEDS. 341 in the fifteenth century, such as Pierre Crescenzio ^ and Macer Floridus,^ mention no faseolus or similar name. On the other hand, after the discovery of America, from the sixteenth century all authors publish descriptions and drawings of Fhaseolus vulgaris, with a number of varieties It is doubtful that its cultivation is ancient in tropical Africa. It is indicated there less often than that of other species of the Dolichos and Phaseolus genera. It had not occurred to any one to seek the origin of the haricot in America till, quite recently, some remark- able discoveries of fruits and seeds were made in Peru- vian tombs at Ancon, near Lima, Rochebrune^ published a list of the species of different families from the collection made by Cossac and Savatier. Among the number are three kinds of haricot, none of which, says the author, is Fhaseolus vulgaris; but Wittmack,* wdio studied the leguminse brought from these same tombs by Reiss and Stubel, says he made out several varieties of the common haricot amonohla in Russian,^ arc names which seem to come from very different roots. In the south of Europe the names are rarer. There is a Spanish one, however, of Asturia, escaiidia,^ but I know of none in Basque. History, and especially philology, point to an origin in eastern temperate Europe and the neighbouring countries of Asia. We have to discover whether the plant has been found wild. Olivier,^ in a passage already quoted, says that he several times found it in Mesopotamia, in particular upon the right bank of the Euphrates, north of Anah, in places unfit for cultivation. Another botanist, Andre Michaux, saw it in 1783, near Hamadan, a town in the temperate region of Persia. Dureau de la Malle says that he sent some grains of it to Bosc, who sowed them at Paris and obtained the common spelt ; but this seems to me doubtful, for Lamarck, in 178G,^ and Bosc ^himself, in the JJidionnaire cV Agriculture, article L'jMautre (spelt), published in 1809, says not a word of this. The herbariums of the Paris Museum contain no specimens of the cereals mentioned by Olivier. There is, as we have seen, much uncertainty as to the origin of the species as a wild plant. This leads me to attribute more importance to the hypothesis that spelt is derived by cultivation from the common wheat, or from an intermediate form at some not very early prehistoric time. The experiments of H. Yilmorin^ support this theory, for cross fertilizations of the spelt * Ad. Pictet, Orig. Indo-Europ., edit. 2, vol. i. p. 318, * Ad. Pictet, ibid. ; Nemnich, Lexicon. * Willkomm and Lange, Prodr. Fl. Hisp., i. p. 107. 4 Olivier, Voyage, 1807, vol. iii. p. 460. ^ Lamarck, Diet. Encycl., ii. p. 560. * H. Yilmorin, Bull. Soc. Bot. de France, ISSl, p. SoS, PLANTS CULTIVATED FOR THEIR SEEDS. 365 by the downy white wheat, and vice versd, yield " hybrids whose fertility is complete, with a mixture of the characters of both parents, those of the spelt pre- ponderating." 2, Starch Wheat — Triticwiii dicoccum, EchYunk ; Triti- cwin amyleuvi, Seringe. This form (Eininer, or Aemer in German), cultivated for starch chiefly in Switzerland, resists a hard winter It contains two grains in each little ear, like the true spelt. Heer ^ attributes to a variety of T. dicoccum an ear found in a bad state of preservation in the lake-dwellings of Wangen, Switzerland. Messicommer has since found some at Robenhausen. It has never been found wild ; and the rarity of common names is remarkable. These two circumstances, and the slio^ht value of the botanical characters which serve to distinguish it from Tr. spelta, lead to the con- clusion that it is an ancient cultivated variety of the latter. 3. Oiie-grainedWheat — Triticuni monococcum, Jj'.nTineus. The one-grained wheat, or little spelt, Einkorii in German, is distinguished from the two preceding by a single seed in the little ear, and by other characters which lead the majority of botanists to consider it as a really distinct species. The experiments of H. Vilmorin con- firm this opinion so far, for he has not yet succeeded in crossing T. monococciivi with other spelts or wheats. This may be due, as he says himself, to some detail in the manner of operating. He intends to renew his attempts, and may perhaps succeed. [In the Bulletin de la Socie'te Botaniqiie de France, 1883, p. 62, Mr. Vilmorin sa3's that he has not met with better success in the third and fourth years in his attempts at crossing T. "iuonococcuin with other species. He intends to make the experiment with T, hoeoticum, Boissier, wild in Servia, of whicli I sent him some seeds gathered by Pancic. As this species is supposed to be the original stock of T. ononococciini, the experiment is an interesting one. — Author's Note, ' Heer, PJlanz. der. Pfahlh., p. 5, fig. 23, and p. 15. 366 ORIGIN OF CULTIVATED PLANTS. 1884. J In the mean time let us see whether this form of spelt has been long in cultivation, and if it has any- where been found growing wild. The one-grained wheat thrives in the poorest and most stony soil. It is not very productive, but yields excellent meal. It is sown es2)ecially in mountainous districts, in Spain, France, and the east of Europe, but I do not find it mentioned in Barbary, Egypt, the East, or in India or China. From some expressions it has been believed to be the tiphai of Theophrastus.^ It is easier to invoke Dioscorides,^ for he distinguishes two kinds of zeia, one with two seeds, another with onlv one. The latter would be the one-grained wheat. Nothing proves that it was commonly cultivated by the Greeks and Romans. Their modern descendants do not sow it.^ There are no Sans- krit, Persian, or Arabic names. I suggested formerly that the Hebrew word hiisse'ineth might apply to this species, but this hypothesis now seems to me difficult to maintain. Marschall Bieberstein* mentions Triticuni mono- coccuon, or a variety of it, growing wild in the Crimea and the eastern Caucasus, but no botanist has confirmed this assertion. Steven,^ who lived in the Crimea, declares that he never saw the species except cultivated by the Tartars. On the other hand, the plant which Balansa gathered in a wild state near Mount Sipyliis, in Anatolia, is T. ononococcum, according to J. Gay,^ wlio takes with this form Tritlcuin hceotlcum, Boissier, which grows wild in the plains of Boeotia "^ and in Servia.^ * Fraas, Syn. Fl. Class., p. 307. ^ Diuscorides, Mat. Med., 2, c. iii. 155. ' Heldreich, Nutz. Griech. * Bieberstein, Fl. Tauro- Caucasaica, voL i. p. 85. * Steven, Verzeichniss Taur. Halhins. PJlan., p. 354-. « Bull. Soc. Bot. Fran., 1860, p. 30. ' Boissier, Diagnoses, 1st series, vol. ii. fasc. 13, p. 69. * Balansa, 1854, No. 137 in Boissier's Herbarium, in wliicli there is also a specimen found in the fields in Servia, and a variety with brown beards sent by Pancic, growing in Servian meadows. The same botanist (of Belgrade) has just sent me wild specimens from Servia, which I cannot distinguish from T. iiionococcum, which he assures me is not cultivated in Servia. Bentham writes to me that T. hceoticum. PLANTS CULTIVATED FOR THEIR SEEDS. 367 Admitting these facts, T. monococcum is a native of Servia, Greece, and Asia Minor, and as the attempts to cross it with other spelts or wheats have not been successful, it is rightly termed a species in the Linneean sense. The separation of wheat with free grains from spelt must have taken place before all history, perhaps before the beginning of agriculture. Wheat must have appeared first in Asia, and then spelt, probably in Eastern Europe and Anatolia. Lastly, among spelts T. mionococcum seems to be the most ancient form, from which the others have gradually developed in several thousand years of cultivation and selection. Two-rowed Barley — Hordeum distichon, Linnaeus. Barley is among the most ancient of cultivated plants. As all its forms resemble each other in nature and uses, we must not expect to find in ancient authors and in common names that precision which would enable us to recognize the species admitted by botanists. In many cases the name barley has been taken in a vague or generic sense. This is a difficulty which we must take into account. For instance, the expression of the Old Testament, of Berosus, of Moses of Chorene, Pausanias, Marco Polo, and more recently of Olivier, indicating "wild and cultivated barley" in a given country, prove nothing, because we do not know to which species they refer. There is the same obscurity in China. Dr. Bretschneider says^ that, according to a work published in the year A.D. 100, the Chinese cultivated barley, but he does not specify the kind. At the extreme west of the old world the Guanchos also cultivated a barley, of which we know the name but not the species. The common variety of the two-rowed barley, in which the husk remains attached to the ripened grain, has been found wild in Western Asia, in Arabia Petrea,^ of which he saw several specimens, is, he thinks, the same as T. monococcum. * Bretschneider, On the Study, etc., p. 8. ' A specimen determined by Reuter in Boissier's Herbarium. 368 ORIGIN OF CULTIVATED PLANTS. near Mount Sinai,^ in the ruins of Persepolis,^ near the Caspian Sea,^ between Lenkoran and Baku, in the desert of Chirvan and Awhasia, to the south of the Ca ueasus,^ and in Turcomania.^ No author mentions it in Greece, Egypt, or to the east of Persia. Willdenow ^ indicates it at Samara, in the south-east of Russia; but more recent authors do not confirm this. Its modern area is, therefore, from the Red Sea to the Caucasus and the Caspian Sea. Hence this barley should be one of the forms cultivated by Semitic and Turanian peoples. Yet it has not been found in Egyptian monuments. It seems that the Aryans must have known it, but I find no proof in vernacular names or in history. Theophrastus '^ speaks of the two-rowed barley. The lake-dwellers of Eastern Switzerland cultivated it before they possessed metals,® but the six-rowed barley was more common amonof them. The variety in which the grain is bare at maturity {H. distichon niidum, Linna3us), which in France has all sorts of absurd names, orge d cafe, orge dio Perovu (coffee barley, Peruvian barley), has never been found wild. The fan-shaped barley {Hordeum Zeocriton, Linnaeus) seems to me to be a cultivated form of the two-rowed barley. It is not known in a wild state, nor has it been found in Egyptian monuments, nor the lake-dwellings of Switzerland, Savoy, and Italy. Common Barley — Hordewm vulgare, Linnreus. The common barley with four rows of grain is mentioned by Theophrastus,^ but it seems to have been * Figari and de Notaris, Agrosfologice jEfiypt. Fragm., p. 18. - A very starved plant gathered by Kotschy, No. 290, of ■which I possess a specimen. Boissier terms it II. distichon, varictas. ' C. A. Meyer, Verzeicluiiss, p. 2G, from specimens seen also by Ledebour, Fl. Ross., iv. p. 327. * Ledebour, ihid. 5 Kegel, Descr. Plant, Nov., 18S1, fasc. 8, p. 37. « Willdenow, Sp. Plant., i. p. 473. ^ Theophrastus, Hist. Plant., lib. viii. cap. 4. ' Heer, Pjianzen der Pfahlhauten, p. 13 ; Messicommer, Flora Bot. Zeifung, 18f59, p. 320. * Theophrastus, Hist., lib. viii. cap. 4. PLANTS CULTIVATED FOR THEIR SEEDa 3G9 less cultivated in antiquity than that with two rows, and considerably less than that with six rows. It has not been found in Egyptian monuments, nor in the lake- dwellings of Switzerland, Savoy, and Italy. Willdenow^ says that it grows in Sicily and in the south-east of Russia, at Samara, but the modern floras of these two countries do not confirm this. We do not know what species of barley it was that Olivier saw growing wild in Mesopotamia ; consequently the common barley has not yet been found certainly wild. The multitude of common names which are attributed to it prove nothing as to its origin, for in most cases it is impossible to know if they are names of barley in general, or of a particular kind of barley cultivated in a given country. Six-rowed Barley — Hordeion hexastichon, Linnaeus. This was the species most commonly cultivated in antiquity. Not only is it mentioned by Greek authors, but it has also been found in the earliest Egyptian monu- ments,^ and in the remains of the lake -dwellings of Switzerland (age of stone), of Italy, and of Savoy (age of bronze).^ Heer has even distinguished two varieties of the species formerly cultivated in Switzerland. One of them answers to the six-rowed barley represented on the medals of Metapontis, a town in the south of Italy, six centuries before Christ. According to Roxburgh,^ it was the only kind of barley grown in India at the end of the last century. He attributes to it the Sanskrit name yuva, which has become juha in Bengali. Adolphe Pictet^ has care- fully studied the names in Sanskrit and other Indo- European languages which answer to the generic name ' Willdenow, Species Plant, i. p. 472. ' Un^er, Fjlanzen des Alten Egyptens, p. 33 ; Ein Ziegel der Dashur Pyramide, p. 109. ' Heer, Pflanzen der Pfahlbauten, p. 5, ficfs. 2 and 3 ; p. 13, fig:. 9 ; Flora Bot. Zeitung, 1869, p. 320; de Mortillefc, according to Pern'n, Etxtdes prehistoriques siir la Savoie, p. 23 ; Sordelli, Sulle piante della torhiera di Lagozza, p. 33. « Roxburgh, Fl. Ind., edit. 1832, vol. i. p. 358. ' Ad. Pictet, Origines Indo-Europ., edit. 2, vol. i. p. 333. 370 . ORIGIN OF CULTIVATED PLANTS. barley, but he has not been able to go into the details of each species. The six-rowed barley has not been seen in the con- ditions of a wild plant, of which the species has been determined by a botanist. I have not found it in Bois- sier's herbarium, which is so rich in Eastern plants. It is possible that the wild barleys mentioned by ancient authors and by Olivier were Hordeuon hexastichon, but there is no proof of this. On Barleys in general, "We have seen that the only form which is now found w^ild is the simplest, the least productive, Hordeum dis- tichon, which was, like H. hexastichon, cultivated in prehistoric time. Perhaps //. vulgare has not been so lonor in cultivation as the two others. Two hypotheses may be drawn from these facts; 1. That the barleys with four and six rows were, in prehis- toric agriculture anterior to that of the ancient Ei^^yptians who built the monuments, derived from // distichon. 2. The barle3^s Avith six and four ranks were species formerly wild, extinct since the historical epoch. It would be strange in this case that no trace of them has remained in the floras of the vast region comprised be- tween India, the Black Sea, and Abyssinia, where we are nearly sure of their cultivation, at least of that of the six-ranked barley. Rye — Secale cereale, Linn?eus. Rye has not been very long in cultivation, unless, perhaps, in Russia and Thrace. It has not been found in Egyptian monuments, and has no name in Semitic languages, even in the modern ones, nor in Sanskrit and the modern Indian languages derived from Sanskrit. These facts agree with the circumstance tliat rye thrives better in northern than in southern countries, where it is not usually cultivated in modern times. Dr. Bret- schneider^ thinks it is unknown to Chinese aii-riculture. He doubts the contrary asseition of a modern writer, * Bretschneider, On Study and Value, etc., pp 18, 44. PLANTS CULTIVATED FOR THEIR SEEDS. S71 and remarks that the name of a cereal mentioned in the memoirs of the Emperor Kanghi, which may be sup- posed to be this species, signifies Russian wheat. Now rye, he says, is much cultivated in Siberia. There is no mention of it in Japanese floras. The ancient Greeks did not know it. The first author who mentions it in the Roman empire is Pliny,^ who speaks of the secale cultivated at Turin at the foot of the Alps, under the name of Asia. Galen,^ born in a.d. 131, had seen it cultivated in Thrace and Macedonia under the name hriza. Its cultivation does not seem ancient, at least in Italy, for no trace of rye has been found in the remains of the lake-dwellings of the north of that country, or of Switzerland and Savoy, even of the age of bronze. Jetteles found remains of rye near Olmutz, together with • instruments of bronze, and Heer,^ who saw the specimens, mentions others of the Roman epoch in Switzerland. Failing archaeological proofs, European languages show an early knowledge of rye in German, Keltic, and Sla- vonic countries. The principal names, according to Adolphe Pictet,^ belong to the peoples of the north of Europe : Anglo-Saxon, ryge, rig ; Scandinavian, rllgr ; Old High German, roggo ; Ancient Slav, ruji, roji ; Polish, rez ; Illyrian, raz, etc. The origin of this name must date, he says, from an epoch previous to the sepa- ration of the Teutons from the Lithuano-Slavs. The word secale of the Latins recurs in a similar form amonof the Bretons, segal, and the Basques, cehela, zehhalea ; but it is not known whether the Latins borrowed it from the Gauls and Iberians, or whether, conversely, the latter took the name from the Romans. This second hypo- thesis appears to be the more probable of the two, since the Cisalpine Gauls of Pliny's time had quite a different name. I also find mentioned a Tartar name, aresch,^ and an Ossete name, syl, sil,^ which points to an ancient cultivation to the east of Europe. I'liny, Hist., lib. xviii. c. 16. * Galeu, De Alimentis, lib. xiii., quoted by Lenz, Bot. de Alt'en, p. 259. ' Heer, Die Pflanzen der Pfahlbauten, p. IG. * Ad. Pictet, Origines Indo-Europ., edit. 2, vol. i. p. 341. * Nemnicli, Lexicon Naturgesch. ? Ad. Pictet, uhi supra, 17 :J72 OEIGIN OF CULTIVATED PLANTS. Thus historical and philological data show that the species probably had its origin in the countries north of the Danube, and that its cultivation is hardly earlier than the Christian era in the Roman empire, but perhaps more ancient in Russia and Tartary. The indication of wild rye given by several authors should scarcely ever be accepted, for it has often hap- pened that Secale cereale has been confounded with perennial species, or with others of which the ear is easily broken, which modern botanists have rightly dis- tinguished.^ Many mistakes which thus arose have been cleared up by an examination of original specimens. Others may be suspected. Thus I do not know what to think of the assertions of L. Ross, who said he had found rye growing wild in several parts of Anatolia,^ and of the Russian traveller Ssaewerzoff, who said he saw it in Turkestan.^ The latter fact is probable enough, but it is not said that any botanist verified the species. Kunth * had previously mentioned it in " the desert between the Black Sea and the Caspian," but he docs not say on what authority of traveller or of specimens. Boissier's herbarium has shown me no wild Secale cereale, but it has persuaded me that another species of rye might easily be mistaken for this one, and that asser- tions require to be carefully verified. Failing satisfactory proofs of wild plants, I formerly urged, in my GeograpJde Botanique Maisonne'e, an argu- ment of some value. Secale cereale sows itself from cultivation, and becomes almost w41d in parts of the Austrian empire/ which is seldom seen elsewhere.^ Thus * Secale fragile, Bieberstein ; S. anatoUcum, Boissier ; S. montanum, Gussone ; S. villosum, Linnaeus. I explained in my Geogr. Botanique, p. 936, the errors which result from this confusion, when rye was said to be wild in Sicily, Crete, and sometimes in llussia. « Flora, Bot. Zeitung, 185«, p. 520. * Flora, Bot. Zeitung, 1869, p. 93. * Kunth, Fnum.f i. p. 449. * Sadler, Fl. Pesth., i. p. 80; Host, Fl. Austr., i. p. 177; Baump^arten, Fl. Transylv., p. 225; Neilreich, Fl. Wien., p. 58; Vivian i, fZ. Daltnat., i. p. 97 ; Farkas, Fl. Croat., p. 1288. * Strobl saw it, however, in the woods on the slopes of Etna, a result of its introduction into cultivation in the eighteenth century {(Ester. Bot. Zeit., 1881, p. 159). PLANTS CULTIVATED FOR THEIR SEEDS. 378 in the east of Europe, where history points to an ancient cultivation, rye finds at the present day the most favour- able conditions for livinsf without the aid of man. It can hardly be doubted, from these facts, that its original area was in the region comprised between the Austrian Alps and the north of the Caspian Sea. This seems the more probable that the five or six known species of the genus Secale inhabit western temperate Asia or the south-east of Europe. Admittinof this orimn, the Arvan natives would not have known the species, as philology already shows us ; but in their migrations westward they must have met with it nnder different names, which they transported here and there. Common Oats and Eastern Oats — Avena sativa, Lin- nseus ; Avena orient alls, Schreber. The ancient Egyptians and the Hebrews did not cultivate oats, but they are now grown in Egypt.^ There is no Sanskrit name, nor any in modern Indian languages. They are only now and. then planted by the English in India for their horses.^ The earliest mention of oats in China is in an historical work on the period 618 to 907 A.D. ; it refers to the variety known to botanists as Avena sativa nuda.^ The ancient Greeks knew the genus very well ; they called it hromos,^ as the Latins called it avena ; but these names were commonly applied to species which are not cultivated, and which are weeds mixed with cereals. There is no proof that they culti- vated the common oats. Pliny's remark^ that the Germans lived on oatmeal, implies that the species was not cultivated by the Romans. The cultivation of oats was, therefore, practised an- ciently to the north of Italy and of Greece. It was diffused later and partially in the south of the Roman empire. It is possible that it was more ancient in Asia Minor, for Galen ^ says that oats were abundant in * Schweinfurth and Ascherson, Beitrage zur Fl. J^thiop , p. 298. » Eoyle, III., p. 419. " Bretschneider, On Study and Value, etc., pp. 18, 44. * Fraas, Syn. Fl. Class., p. 303 ; Leiiz, Bot. der Alien, p. 243. * Pliuy, Hist., lib. xviii. cap. 17. ® Galen, De Alimentis, lib. i. cap. 12. 374) ORIGIN OF CULTIVATED PLANTS. Mysia, above Pergamus ; that they were given to horses, and that men used them for food in years of scarcity. A colony of Gauls had formerly penetrated into Asia Minor. Oats have been found among the remains of the Swiss lake-dwellings of the age of bronze/ and in Germany, near Wittenburg, in several tombs of tho first centuries of the Christian era, or a little earlier.^ Hitherto none hav^e been found in the lake-dwellings of the north of Italy, which confirms the belief that oats were not cultivated in Italy in the time of the Roman republic. The vernacular names also prove an ancient existence north and west of the Alps, and on the borders of Europe towards Tartary and the Caucasus. The most widely diffused of these names is indicated by the Latin avena, Ancient Slav ovisu, ovesu, ovsa, Russian ovesii, Lithuanian awiza, Lettonian aiisas, Ostias ahis.^ The English word oats comes, according to A. Pictet, from the Anglo-Saxon ata or ate. The Basque name, olba or o/oa,* argues a very ancient Iberian cultivation. The Keltic names are quite diff'erent:^ Irish coirce, cuirce, corca, Armorican kerch. Tartar sidii, Georgian kari, Hungarian zah, Croat zoh, Esthonian kaer, and othere are mentioned by Nemnich^ as applying to the generic name oats, but it is not likely that names so varied do not belong to a cultivated species. It is strange that there should be an independent Berber name zekkoum^ as there is nothing to show that the species was anciently cultivated in Africa. All these facts show how erroneous is the opinion which reigned in the last century,^ that oats were brought originally from the island of Juan Fernandez, a belief which came apparently from an assertion of the navigator Anson.^ It is evidently not in the Austral * Heer, Pflanzen der Pfahlhauten, p. 6, fig. 24. ' Leuz, Bot. der Alien, p. 245. ' Ad. Pictet, Orig. Indo.-Europ., edit. 2, vol. i. p. 350. * Notes communicated by ISL Clos. * Ad. Pictet, \ili supra. " Nemnich, Polyglott. Lexicon, p. 548. ' Diet. Fr.-Berhere, published by the French Government. * Linnaeus, Species, p. 118; Lamarck, Diet. Enc., i. p. 4>31, ^ Phillips, Cult. Veget., ii. p. 4. PLANTS CULTIVATED FOR THEIR SEEDS. 375 hemisphere that we must seek for the home of the species, but in those countries of the northern hemisphere where it was anciently cultivated. Oats sow themselves on rubbish-heaps, by the way- side, and near cultivated ground more easily than other cereals, and sometimes persist in such a way as to appear wild. This has been observed in widely separate places, as Algeria and Japan, Paris and the north of China.^ Instances of this nature render us sceptical as to the wild nature of the oats which Bove said he found in the desert of Sinai. It has also been said ^ that the traveller Olivier saw oats wild in Persia, but he does not mention the fact in his work. Besides, several annual species nearly resembling oats may deceive the traveller. I cannot discover either in books or herbaria the existence of really wild oats either in Europe or Asia, and Bentham has assured me that there are no such specimens in the herbarium at Kew ; but certainly the half- wild or naturalized condition is more frequent in the Austrian states from Dalmatia to Transylvania^ than elsewhere. This is an indication of origin which may be added to the historical and philological arguments in favour of eastern temperate Europe. Avena strigosa, Schreber, appears to be a variety of the common oats, judging from the experiments in culti- vation mentioned by Bentham, who adds, it is true, that these need confirmation.* There is a good drawing of the variety in Host, Icones Graminum Austriacoriun, ii. pi. 56, which may be compared with A. sativa, pi. 59. For the rest, Avena strigosa has not been found wild. It exists in Europe in deserted fields, which confirms the hypothesis that it is a form derived by cultivation. Avena orientcdis, Schreber, of which the spikelets * Munby, Cafal. Alger., edit. 2, p. 38 ; Franchet .ind Savatier, Enum. Pi, Jap., ii. p. 175 ; Cosson, Fl. Paris, ii. p. Go7 ; Buiige, Enum. Chin., p. 71, for tlie variety nuda. * Lamarck, Diet. Encycl., i. p. 331. ' Viviani, Fl. Dalmat., i. p. G9 ; Host, Fl. Anstr., i. p. 13S ; Ncilreich, Fl. Wien., p. 85; Baumgarten, Enum. Transylv., iii. p. 259 j Farkas, FL Croatica, p. 1277. * BQiithnra., Uandboolc of Br Hi ^h Flora, edit. 4, p. 54)4. o76 ORIGIN OF CULTIVATED PLANTS. lean all to one side, has also been grown in Europe from the end of the eighteenth century. It is not known in a Avild state. Often mixed with common oats, it is not to be distinguished fiom them at a glance. The names it bears in Germany, Turkish or Hungarian oats, points to a modern introduction from the East. Host ogives a fjood drawing of it (Gram. Austr., i. pi. 44). As all the varieties of oats are cultivated, and none have been discovered in a truly wild state, it is very probable that they are all derived from a single j^re- historic form, a native of eastern temperate Europe and of Tartary. Common Millet — Panicum oiiiliaceuin, Linnreus. The cultivation of this plant is prehistoric in the south of Europe, in Egypt, and in Asia. The Greeks knew it by the name hcychros, and the Latins by that of miliuDi} The Swiss lake-dwellers of the aire of stone made great use of millet,^ and it has also been found in the remains of the lake-dwellings of Varese in Italy.^ As we do not elsewhere find specimens of these early times, it is impossible to know wliat was the j)anwuin or the sorghutn mentioned by Latin authors which was used as food by the inhabitants of Gaul, Panonia, and other countries. Un2:er ^ counts P. miliaceiiin amonjx the species of ancient Egypt, but it does not appear that he had positive proof of this, for he has mentioned no monu- ment, drawing, or seed found in the tombs. Nor is there any material proof of ancient cultivation in Mesopotamia India, and China. For the last-named country it is a question whether the slai, one of the five cereals sown by the emperors in the great yearly ceremony, is Panicxun 'iuiliaceum, an allied species, or sorghum ; but it appears that the sense of the word shu has chano-ed, and that formerly it was perhaps sorghum which was sov/n.^ * The passages from Theophrastus, Cato, and others, are translated in Lenz, Botanik der Alten, p. 232. '■* Heer, Pjianzen der Pfahlbauten, p. 17. 3 Regazzoni, Riv. Arch. Prov. di Como, 18S0, fasc. 7. * Unger, Pflmzen des Allen JEn^iptens, p. 34. * Bretschueider, Study and Value of Chinese Botanical Works, pp. 7, 8, 45. PLANTS CULTIVATED FOR THEIR SEEDS. o77 Anglo-Indian botanists^ attribute two Sanskrit names to the modern species, llnu and vreehib-heda, althouoh the modern Hindu and Bencfali name cheena and the Telinga name iuo7Y/a are quite different. If the Sanskrit names are genuine, they indicate an ancient cultivation in India. No Hebrew nor Berber name is known, ^ but there are Arab names, doJdtn, used in Egypt, and Jwsjccjb in Arabia.^ There are various European names. Besides the Greek and Latin words, there is an ancient Slav name, proso,'^ retained in Russia and Poland, an old Gemian word hirsi, and a Lithuanian name sora^ The absence of Keltic names is remarkable. It appears that the species was cultivated especially in Eastern Europe, and spread westward towards the end of the Gallic dominion. With regard to its wild existence, Linnseus says ^ that it inhabits India, and most authors repeat this ; but Anglo-Indian botanists '^ always give it as cultivated. It is not found in Japanese floras. In the north of China de Bunge only saw it cultivated,^ and Maximowicz near the Ussuri, on the borders of fields and in places near Chinese dwellings.^ Ledebour says ^"^ it is nearly wild in Altaic Siberia and Central Russia, and wild south of the Caucasus and in the country of Talysch. He quotes Hohenacker for the last-named locality, who, however, says only " nearly wild." ^^ In the Crimea, where it furnishes bread for the Tartars, it is found here and there nearly wild/^ which is also the case in the south of France, in Italy, and in Austria.^^ It is not wild in > Roxburgh, Fl. Ind., edit. 1832, p. 310 ; Piddington, I?} Jp-t., ' Rosenmiiller, Bibl. Alterth. ; Diet. Franq.'Berhere. * Delile, FL JSgypt., p. 3 ; Forskal, Fl. Arab., civ. * Ad. Pictet, Origines Indo-Europdennes, edit. 2, vol. i. p. 351, * Ibid. ^ Linnoeus, Spec. Plant., i. p. 86. ^ Roxburgh, i'L Ind., edit. 1832, p. 310 j Aitchison, Cat. of Punjab PI., p. 159. * Bunge, Fnum.y No. 400. • Maximowicz, Primitice Amur., p. 3C0. " Ledebour, Fl. Ross., iv. p. 469. " Hohenacker, Plant. Talysch., p. 13. '* Steven, Verzeich. Halb. Taur., p. 371. " Mutel, Fl. Franc^., iv. p. 20 ; Parlatore, Fl. ItaL, i. p. 122 ; Yiviani, Fl. Damat., i. p. 60 j Neilreich, Fl. Nicd. QHsterr., p. 32. 378 OFJGIN OF CULTIVATED PLANTS. Greece,^ and no one has found it in Persia or in Syria. Forskal and Delile indicated it in Egypt, but Ascherson does not admit tliis;^ and Forskal gives it in Arabia.'^ The species may have become naturalized in these regions, as the result of frequent cultivation from the time of the ancient Egyptians. However, its wild nature is so doubtfid elsewhere, that its Egypto- Arabian origin is very probable. Italian Millet — Panicmyi Italicum, Linn?eus; Setaria Italica, Beauvois, The cultivation of this species was very common in the temperate parts of the old world in prehistoric times. Its seeds served as food for man, thouo^h now they are chiefly given to birds. In China it is one of the five plants which the emperor sow^s each year in a public ceremony, according to the command issued bv Chin-nonc: 2700 B.C.* The common name is siao im (little seed), the more ancient name being kvb ; but the latter seems to be applied also to a very ditierent species.^ Pickering says he recognized it in two ancient Egyptian diawings, and that it is now cultivated in Egypt ^ under the name doJihn ; but that is the name of Panicuvi iniliaceitin. It is, therefore, very doubtful that the ancient Egyptians cultivated it. It has been found among the remains of the Swiss lake-dw^ell- ings of the stone epoch, and therefore a fortiori among the lake-dwellers of the subsequent epoch in Savoy.'' The ancient Greeks and Latins did not mention it, or at least it lias not been possible to certify it from wdiat they say of several panicums and millets. In our own day the species is rarely cultivated in the south of Europe, not at all in Greece,** for instance, and I do not ^ Heldreich, Niitz. Griechenl., p. 3 ; Pflinz. Aftisch. Ehene., p. 516. ' M. Ascherson informs me in a letter that in his Aufzdhlung the word cult, has been omitted by mistake after Fanicv.ni miliaceum. ^ Forskal, FL Arab., p. civ, * Bretschneider, Stxidy and Value, etc., pp. 7, 8 * Bretschneider, ibid. * According to Unger, PJlanz. d. Alt. j^gypt., p. 34. ' Heer, Pfanzen d. Pfnhlbaut.^ p. 5, fig. 7 ; p. 17, figs. 28, 29 ; Perrin, JStudes Prehifforiques sur la /S'avoi'e, p. 22. * Heldreich, NutZ2jl. Griech. PLANTS CULTIVATED FOR THEIR SEEDS. 370 find it indicated in Egypt, but it is common in Southern Asia.^ The Sanskrit names kiingil and iJriyvMgiil, of which the first is retained in Bengali,^ are attributed to this species. Piddingtori mentions several other names in Indian languages in his Index. Ainslie^ gives a Per- sian name, arzun, and an Arabic name ; but the latter is commonly attributed to Fanicum oniUaceuni. There is no Hebrew name, and the plant is not mentioned in botanical works upon Egypt and Arabia. The European names have no historical value. They are not original, and commonly refer to the transmission of the species or to its cultivation in a given country. The specitic name, italicuni, is aii absurd example, the plant being rarely cultivated and never wild in Italy. Rumphius says it is wild in the Sunda Isles, but not very positively.'^ Linnseus j)robably started from this basis to exaggerate and even promulgate an error, saying, " inhabits the Indies." ^ It certainly does not come from the West Indies ; and further, Roxburgh asserts that he never saw it wild in India. The Gramin^e have not yet appeared in Sir Joseph Hooker's flora; but Aitchi- son^ gives the species as only cultivated in the north- west of India. The Australian plant which Robert Brown said belonged to this species belongs to another."^ P. italicuni appears to be wild in Japan, at least in the form called gennanica by ditterent autliors,^ and the Chinese consider tlie five cereals of the annual ceremony to be natives of their country. Yet Bunge, in the north of China, and Maximowicz in the basin of the river Amur, only saw the species cultivated on a large scale, in the form of the gennanica variety.^ In * Roxburgh, FI.. Ind., edit. 1832, vol. i. p. 302; Rumpliius, A7nhoin.,y. p. 202, t. 75. 2 Roxburgh, ibid. ' Ainslie, 3Tat. Med. Ind., i. p. 226. * " Obcuin-it in Baleya," etc. (Rumphius, v. p. 20:^). * "Habitat in Indiis" (Linnaeus, Species, i. p. 83). * Aitchison, Catal. of Punjab PL, p. 102. ^ Bent ham, Flora Austral., vii. p. 493. •* Franchet and Savatier, Enuni. Japon., ii. p. 262. * Bunge, Enum., No. 399; Maximowicz, Primitice Amur., p. 330. 380 ORIGIN OF CULTIVATED PLANTS. Persia,^ the Caucasus Mountains, and Europe, I only find in floras the plant indicated as cultivated, or escaped sometimes from cultivation on rubbisli-heaps, waysides, waste ground, etc.^ The sum of the historical, philological, and botanical data make me think that the species existed before all cultivation, thousands of years ago in China, Japan, and in the Indian Archipelago. Its cultivation must have early spread towards the West, since we know of Sanskrit names, but it does not seem to have been known in Syria, Arabia, and Greece, and it is probably through Russia and Austria that it early arrived among the lake-dwellers of the stone age in Switzerland. Common Sorghum — Holcus sorghum, Linnaeus ; An- dropogon sorghum, Brotero ; Sorghum vulgare, Persoon. Botanists are not as^reed as to the distinction of several of the species of sorghum, and even as to the genera into which this group of the Graminre should be divided. A good monograph on the sorghums is needed, as in the case of the panicums. In the mean time I will give some information on the principal species, because of their immense importance as food for man, rearing of poultry, and as fodder for cattle. We may take as a typical species the sorghum culti- vated in Europe, as it is figured by Host in his Gramince Austriacce (iv. pi. 2). It is one of the plants most com- monly cultivated by the modern Egyptians, under the name of dourra, and also in equatorial Africa, India, and China.^ It is so productive in hot countries that it is a staple food of immense populations in the old world. Linnaeus and all authors, even our contemporaries, say that it is of Indian origin ; but in the first edition of Roxburgh's flora, published in 1820, this botanist, who should have been consulted, asserts that he had only seen it cultivated. He makes the same remark for the allied forms (Jjicolor, saccharatus, etc.), which are often regarded ^ Buhse, Aufzdhlung, p. 232. ' See Pavlatore, Fl. Ital., i. p. 113 ; Mutel, Fl. Franc., iv. p. 20, etc. ' Delile, Plantes Cult, en Eg\ipte, p. 7 ; Roxburjrh, Fl. Inch, edit. 1832, vol. i, p. 269; Aitchison, Catal. of Punjab PL, p. 175; Bretschneider, Study and Value, etc., p. 9. PLANTS CULTIVATED FOR THEIR SEEDS. 381 as mere varieties. Aitchison also bad only seen the sor- ghum cultivated. The absence of a Sanskrit name also renders the Indian origin very doubtful. Bretschneider, on the other hand, says the sorghum is indigenous in China, although he s&ys that ancient Chinese authors have not spoken of it. It is true that he quotes a name, common at Pekin, kao-liang (tall millet), which also applies to Holcus saccharatus, and to which it is better suited. The sori^hum has not been found amono: the remains of the lake-dwellings of Switzerland and Italy. The Greeks never spoke of it. Pliny's phrase ^ about a miliuim introduced into Italy from India in his time has been supposed to refer to the sorghum ; but it was a taller plant, perhaps Holcus saccharatus. The sorghum has not been found in a natural state in the tombs of ancient Egypt. Dr. Hannerd thought he recognized it in some crushed seeds brought by Kosellini from Thebes ; ^ but Mr. Birch, the keeper of Egyptian antiquities in the British Museum, has more recently declared that the species has not been found in the ancient tombs.^ Pickering says he recog- nized its leaves mixed with those of the papyrus. He says he also saw paintings of it ; and Leipsius has copies of drawings which he, as well as linger and Wilkinson, takes to be the doiirra of modern cultivation.^ The height and the form of the ear are undoubtedly those of the sorghum. It is possible that this species is the clochan, once mentioned in the Old Testament ^ as a cereal from which bread was made ; j^et the modern Arabic word dohhn refers to the sweet sorghum. Common names tell us nothing, either from their lack of meaning, or because in many cases the same name has been applied to the different kinds of panicum and sorghum. I can find none which is certain in the ancient languages of India or Western Asia, which * Pliny, Hist, lib. xviii. c. 7» * Quoted by Unger, Die Pflanzen des Alien Egypfens, p. 34. ' S. Birch, in Wilkinson, Man. and Gust, of Anc. Egyptians, 1878, vol. ii. p. 427. * Lepsius* drawings are reproduced by Unger and by Wilkinson. « Ezek. iv. 9. 382 ORIGIN OF CULTIVATED PLANTS. argues an introduction of but few centuries before the Christian era. No botanist mentions the dourra as wild in Egypt or in Arabia. An analogous form is wild in equatorial Africa, but R. Brown has not been able to identify it/ and the flora of tropical Africa in course of publication at Kew has not yet reached the order Graminse. There remains, therefore, the single assertion of Dr. Bretsch- neider, that the tall sorghum is indigenous in China. If it is really the species in question, it spread westward very late. But it was known to the ancient Egyptians, and how could they have received it from China while it remained unknown to the intermediate peoples ? It is easier to understand that it is indigenous in tropical Africa, and was introduced into Egypt in prehistoric time, afterwards into India, and finally into China, where its cultivation does not seem to be very ancient, for the first work which mentions it belongs to the fourth cen- tury of our era. In support of the theor}^ of African origin, I may quote the observation of Schmidt,^ that the species abounds in the island of San Antonio, in the Cape Verde group, in rocky places. He believes it to be " completely natural- ized," which perhaps conceals a true origin. Sweet Sorghum — Holcus saccharatiis, Linna3us ; A'ii- dropogon saccharatus, Boxburgh ; SorgJaim sacchara- tmn, Persoon. This species, taller than the common sorghum and with a loose panicle,^ is cultivated in tropical countries for the seed — which, however, is not so good as that of the common soro-hum — and in less hot countries as fodder, or even for the su^ar wliich the stem contains in con- siderable quantities. The Chinese extract a spirit from it, but not sugar. The opinion of botanists and of the public in general is that it comes from India ; but Boxburgh says that it is only cultivated in that country. It is the same in * Brown, Bof. of Congo, p. SM. ' Schmidt, Beitrdge zur Flora Capverdischen Inseln, p. 158. ' See Host, Gramince Austriacoe, vol. iv. pi. 4. PLANTS CULTIVATED FOR THEIR SEEDS. 883 the Siinda Isles, where the hattari is certainly this species. It is the kao-liang, or great millet of the Chinese. It is not said to be indigenous in China, nor is it men- tioned by Chinese authors who lived before the Christian era.^ From these facts, and the absence of any Sanskrit name, the Asiatic origin seems to me a delusion. The plant is now cultivated in Egypt less than the common sorghum, and in Arabia under the name dohhna or dokhn.^ No botanist has seen it wild in these countries. There is no proof that the ancient Egyptians cultivated it. Herodotus ^ spoke of a " tree-millet " in the plains of Assyria. It might be the species in question, but it is not possible to prove it. The Greeks and Komans were not acquainted with it, not at least before the Roman empire, but it is possible that this was the millet, seven feet high, which Pliny mentions ^ as having been introduced from India in his lifetime. We must probably seek its origin in tropical Africa, where the species is generally cultivated. Sir William Hooker ^ mentions specimens from the banks of the river Nun, which were perhaps wild The approaching pub- lication of the Graminse in the flora of tropical Africa will probably throw some light on this question. The spread of its cultivation from the interior of Africa to Egypt after the Pharaohs, to Arabia, the Indian Archi- pelago, and, after the epoch of Sanskrit, to India, lastly to China, towards the beginning of our era, tallies with historical data, and is not ditiicult to admit. The inverse hypothesis of a transmission from east to west presents a number of objections. Several varieties of sorghum are cultivated in Asia and in Africa; for instance, cernuus with drooping * Roxburgh, Fl. Ind., edit. 2, vol. i. p. 271; Rumphins, Amhoin., v. p. 19 1, pi. 75, fig. 1; Miqnel, FL Indo-Batava, iii. p. 503; Bretschiieidei", Study and Value, etc., pp. 9, 46 ; Loureiro, FL Cochin., ii. p. 792. ' Forskal, Delile, Schweinfurth, and Ascherson, uhi sujora. * Herodotus, lib. i. cap. 193. * Plinv, Hist., lib. xviii. cap. 7. This may also be the variety oi species known as hicolor. * W. Hooker, Niger Flora. 384- OPwIGIN OF CULTIVATED PLANTS. panicles, mentioned by Roxburgh, and which Prosper Alpin had seen in Egypt; hicolor, which in height re- sembles the saccharatits ; and niger and rubens, which also seem to be varieties of cultivation. None of these has been found wild, and it is probable that a monograph would connect them with one or other of the above- mentioned species. Coracan — Eleusine coracana, Gsertner This annual grass, which resembles the millets, is cul- tivated especially in India and the Malay Archipelago. It is also grown in Egypt ^ and in Abyssinia;^ but the silence of many botanists, w^ho have mentioned the plants of the interior and west of Africa, shows that its cultiva- tion is not widely spread on that continent. In Japan ^ it sometimes escapes from cultiv^ation. The seeds will ripen in the south of Europe, but the plant is valueless tliere except as fodder^ No author mentions havin^f found it in a wild state in Asia or in Africa. Roxburgh,^ who is attentive to such matters, after speaking of its cultivation, adds, " I never saw it wild." He distino^uishes under the name Eleusine stricta a form even more commonly cultivated in India, which appears to be simply a variety of E, coracana, and which also he has not found uncultivated. We shall discover its country by other means. In the first place, the species of the genus Eleusine are more numerous in the south of Asia than in other tropical regions. Besides the cultivated plant, Royle ^ mentions other species, of which the poorer natives of India gather the seeds in the plains. According to Piddington's Index, there is a Sanskrit name, rajika, and several other names in the modern lanofuajxes of India. That of coracana comes from an old name used in Ceylon, kouraldian? In the Malay Archipelago the names appear less numerous and less original. ' Schweinfurth and Ascherson, Aufzdhlung, p. 299. ' Bon Jardiiiier, 1880, p. 585. ' Franchet and Savatier, Enum. Plant. Japon., ii. p. 172. * Bon Jardinier, ibid. * Roxburgh, Fl. Indica, edit. 2, vol. i. p. 313. • B-oyle, III. Him. Plants. ^ Thwaites, Enum. PL Zeylan.., p. 371. PLANTS CULTIVATED FOR THEIR SEEDS. 885 In Egypt the cultivation of this species is perhaps not very ancient. The monuments of antiquity bear no trace of it. Groeco-Koman authors who knew the country did not speak of it, nor later Prosper Alpin, Forskal, and Delile. We must refer to a modern work, that of Schweinfurth and Ascherson, to find mention of the species, and I cannot even discover an Arab name.-^ Thus botany, history, and philology point to an Indian origin. The flora of British India, in which the Graminge have not yet appeared, will perhaps tell us the plant has been found wild in recent explorations. A nearly allied species is grown in AhyHsmia,, EI eitsine Tocussa, Fresenius,^ a plant very little known, which is perhaps a native of Africa. Eice — Oryza saliva, Linnaeus. In the ceremony instituted by the Chinese Emperor Chin-nong, 2800 years B.C., rice plays the principal part. The reigning emperor must himself sow it, whereas the four other species are or may be sown by the princes of his family^ The five species are considered by the Chinese as indigenous, and it must be admitted that this is probably the ease with rice, which is in general use, and has been so for a long time, in a country intersected by canals and rivers, and hence peculiarly favourable to aquatic plants. Botanists have not sufficiently studied Chinese plants for us to know whetlier rice is often found outside cultivated ground ; but Loureiro * had seen it in marshes in Cochin-China. Rumphius and modern writers upon the Malay Archipelago give it only as a cultivated plant. The multitude of names and varieties points to a very ancient cultivation. In British India it dates at least from the Aryan invasion, for rice has Sanskrit names, vrihi, * Several synonyms and the Arabic name in Linnains, Delile, etc., apply to Dactyloctenium cegyptiucum, Willdeuovv, or Eleusine cejyjjtiaca of some authors, which is not cultivated. 2 Freseuius, Catal. Sem. Rorti. Franco/., 1834, Beitr. z. Fl. Abyss., p. 141. ' Stanislas Julien, in Loiseleur, Consid. sur les Cdrcales, part i. p. 29 ; Bretschneider, Study and Value of Chinese Botanical Works, pp. 8 and 9. * Loureiro, Fl. Cochin., i. p. 207. 386 ORIGIN OF CULTIVATED PLANTS. arunya} whence come, probably, several names in modern Indian languages, and oriiza or ovxizon of tlie ancient Greeks, roiiz or arous of the Arabs. Theophrastus ^ mentioned rice as cultivated in India. The Greeks became acquainted with it through Alexander's expedi- tion. "According to Aristobulus/' says Strabo,^ "rice grows in Bactriana, Babjdonia, Susida ; " and he adds, " we may also add in Lower Syria." Further on he notes that the Indians use it for food, and extract a spirit from it. These assertions, doubtful perhaps for Bactriana, show that this cultivation was firmly established, at least, from the time of Alexander (400 B.C.), in the Euphrates valley, and from the beginning of our era in the hot and irrigated districts of Syria. The Old Testament does not mention rice, but a careful and judicious writer, Reynier,^ has remarked several passages in the Talmud which relate to its cultis^ation. These facts lead us to suppose that the Indians employed rice after the Chinese, and that it spread still later towards the Euphrates — earlier, however, than the Aryan invasion into India. A thousand years elapsed between the existence of this cultivation in Babylonia and its transportation into Syria, whence its introduction into Egypt after an interval of probably two or three centuries. There is no trace of rice among the grains or paintings of ancient Egypt.^ Strabo, who had visited this country as well as Syria, does not say that rice was cultivated in Egypt in his time, but that the Garam antes ^ grew it, and this people is believed to have inhabited an oasis to the south of Cartlmge. It is possible that they received it from Syria. At all events, Egypt could not lung fail * PiddingtoTi, Indev ; Hehn, Culturpflanzen^ edit. 3, p. 437. 2 Theophrastus, Hist., lib. iv. cap. 4, 10. ^ Strabo, Geographic, Tardieu's translation, lib. xy. cap. 1, § 18; lib. XV. cap. 1, § 53. * Eeynier, ^Vonomie des Arahes et des Juifs (1820), p. 450 ; Ecmomie Puhlique et Rurale des £gyptiens et des Carthaijinois (181:3), p. 324. * Unger mentions none ; Birch, in 1878, furnishes a note to Wilkin- son's Manners and Customs of the Ancient Eijyptiuns, ii. p. 402, " TJiere is no proof of the cultivation of rice, of which no grains have been found." ® Beynier, ihid. PLANTS CULTIVATED FOR THEIR SEEDS. 887 to possess a crop so well suited to its peculiar conditions of irrigation. The Arabs introduced the species into Spain, as we see from the Spanish name arroz. Rice was first cultivated in Italy in 14G8, near Pisa.-^ It is of recent introduction into Louisiana. When I said that the cultivation of rice in India was probably more recent than in China, I did not mean that the plant Avas not wild there. It belongs to a family of which the species cover wide areas, and, besides, aquatic plants have commonly more extensive habitations than others. Rice existed, perhaps, before all cultivation in Southern Asia from China to Bengal, as is shown by the variety of names in the monosyllabic languages of the races between India and China.^ It has been found outside cultivation in several Indian localities, according to Roxburgh.^ He says that wild rice, called newaree by the Telingas, grows in abundance on the shores of lakes in the country of the Circars. Its grain is prized by rich Hindus, but it is not planted because it is not very productive. Roxburgh has no doubt that this is the original plant. Thomson * found wild rice at Moradabad, in the province of Delhi. Historical reasons support the idea that these specimens are indigenous. Otherwise they might be supposed to be the result of the habitual cultivation of the species, all the more that there are examples of the facility with which rice sows itself and becomes naturalized in warm, damp climates.^ In any case historical evidence and botanical probability tend to the belief that rice existed in India before cultivation.^ Maize — Zea Tiiays, Linnseus. " Maize is of American origin, and has only been intro- duced into the old world since the discovery of the new. * Targioni, Cenni Storict, 2 Crawfurd, in Journal of Botany, 18GG, p. 324. 3 Roxburgh, Fl. Ind., edit. 1832, vol. ii. p. 2U0, * Aitcliinson, Catal. Pu,njab., p. 157. * Nees, in Martius, FL Brasil., in Svo, ii. p. 518 j Baker, Fl. of Mauri' ills, p. 458. •* Von Mueller wntes to me that rico is certainly wild in tropical Australia, It may have been accidentally sown, and have become naturalized. — Authgh's note, 188 i. 388 ORIGIN OF CULTIVATED PLANTS. I consider these two assertions as positive, in spite of the contrary opinion of some authors, and the doubts of the celebrated agriculturist Bonafous, to whom we are indebted for the most complete treatise upon maize." ^ I used these words in 1855, after having already contested the opinion of Bonafous at the time of the publication of his work.^ The proofs of an American origin have been since reinforced. Yet attempts have been made to prove the contrary, and as the French name, hie de Turquie, gives currency to an error, it is as well to resume the discussion with new data. No one denies that maize was unknown in Europe at the time of the Roman empire, but it has been said that it was brougrht from the East in the Middle Acjes. The principal argument is based upon a charter of the thir- teenth century, published by Molinari,^ according to which two crusaders, companions in arms of Boniface III., Marquis of Monferrat, gave in 1204 to the town of Incisa a piece of the true cross . . . and a purse containing a kind of seed of a golden colour and partly white, unknown in the country and brought from Anatolia, where it was called inelkja, etc. Tlie historian of the crusades, Michaux, and later Daru and Sismondi, said a great deal about this charter; but the botanist Delile, as well as Targioni- tozzetti and Bonafous himself, thought that the seed in question might belong to some sorghum and not to maize. These old discussions have been rendered absurd by the Comte de Riant's discovery * that the charter of Incisa is the fabrication of a modern impostor. I quote this instance to show how scholars who are not naturalists may make mistakes in the interpretation of the names of plants, and also how dangerous it is to rely upon an isolated proof in historical questions. The names hU de Tiirquie, Turkish wheat (Indian * Bonafous, Hist. Nat. Ajric. et £conomique du Mais, 1 vol. in folio, Paris and Turin, 1836. ^ A. de Candolle, Bihliothequo Universelle de Qeneve, Aug. 1836, GSogr. Bot. Rais., p. 942. ' Molinari, Storia d'Incisa, Asti, 1810. * Riant, La Chai-te d' Incisa, Svo pamphlet, 1877, reprinted from the Revue des Questions Historiques. PLANTS CULTIVATED FOE THEIR SEEDS. 880 corn), given to maize in almost all modern European lan- guages no more prove an Eastern origin than the charter of Incisa. These names are as erroneous as that of coq d'Inde, in English turkey, given to an American bird. Maize is called in Lorraine and in the Vossres Roman coi'n ; in Tuscany, Sicilian corn ; in Sicily, Indian corn ; in the Pyrenees, Spanish corn ; in Provence, Barbary or Guinea corn. The Turks call it Egyptian corn, and the Egyp- tians, Syrian dourra. This last case proves at least that it is neither Egyptian nor Syrian. The widespread name pf Turkish wheat dates from the sixteenth century. It sprang from an error as to the origin of the plant, which was fostered perhaps by the tufts which terminate the ears of maize, which were compared to the beard of the Turks, or by the vigour of the plant, which may have given rise to an expression similar to the French fort comme un turc. The first botanist who uses the name, Turkish wheat, is Ruellius, in 1536.^ Bock or Tragus,^ in 1552, after giving a drawing of the species which he calls FrumentuTti turcicuni, Welachkorn, in Germany, having learnt by merchants that it came from India, conceived the unfortunate idea that it was a certain typha of Bac- triana, to which ancient authors alluded in vague terms. Dodoens in 1583, Camerarius in 1588, and Matthiole^ rec- tified these errors, and positively asserted the American origin. They adopted the name mays, which they knew to be American. We have seen (p. 863) that the zea of the Greeks was a spelt. Certainly the ancients did not know maize. The first travellers* who described the productions of the new world were surprised at it, a clear proof that they had not known it in Europe. Hernandez,^ who left Europe in 1571, according to some authorities, in 1598 according to others,*^ did not know that from the * Ruellius, De Natura Stirpium, p. 428, " Hanc quoniam nostrorum ODtate e Grsccia vel Asia veuerit Turcicum fruinentum noininaut." Fuch- sius, p. 824, repeats this phrase in 1543. * Tragus, Stirpium, etc., edit. 1552, p. 650. ' Dodoens, Pcmptades, p. 509; Camerarius, JJor^, p. 94 ; Matthiole, deit. 1570, p. 305. * P. Martyr, Ercilla, Jean de Lcry, etc., 1510-1578. * Hernandez, Thes. Mexic. , p. 242. ** Lasegue, Muti6e Delessert, p. 4G7. S90 ORIGIN OF CULTIVATED PLANTS. year 1500 maize had been sent to Seville for cultivation. This iact, attested by Fee, who has seen the munici[)al records/ clearly shows the American origin, which caused Hernandez to think the name of Turkish wheat a very bad one. It may perhaps be urged that maize, new to Europe in the sixteenth century, existed in some parts of Asia or Africa before the discovery of America. Let us see what truth there may be in this. The famous orientalist D'Herbelot ^ had accumulated several errors pointed out by Bonafous and by me, on the subject of a passage in the Persian historian Mirkoud of the fifteenth century, about a cereal which Rous, son of Japhet, sowed upon the shores of the Caspian Sea, and which he takes to be the Indian corn of our day. It is hardly worth considering these assertions of a scholar to whom it had never occurred to consult the works of the botanists of his own day, or earlier. AVhat is more im- portant is the total silence on the subject of maize of the travellers who visited Asia and Africa before the discovery of America; also the absence of Hebrew and Sanskrit names for this plant; and lastly, that Egyptian monu- ments present no specimen or drawing of it.^ Rifaud, it is true, found an ear of maize in a sarcophagus at Thebes, but it is believed to have been the trick of an Arab impostor. If maize had existed in ancient Egypt, it would be seen in all monuments, and would have been connected with religious ideas like all other remarkable plants. A species so easy of cultivation would have spread into all neighbouring countries. Its cultivation would not have been abandoned ; and we find, on the contrary, that Prosper Alpin, visiting Egypt in 1592, does not speak of it, and that Forskal,^ at the end of the eighteenth century, men- tioned maize as still but little grown in Egypt, where it had no name distinct from the sorghums. Ebn Baithar, * Fee, Souvenirs de la Guerre cVE.^pagne, p, 128. ' Biblioiheque Orientale, Paris, 16U7, at the word Rous. ' Kunth, Ann. Sc. Nat., ser. 1, vol. viii. p. 418; Raspail, ibid. ; TJnger, rjlanzen des Alien JE'jyiytens ; A. Braun, Pjlanzenreste jfj'jypt. Mus. in Berlin; Wilkiuson, Manners and Customs of Ancient Egyptians. * Forskal, p. liii. PLANTS CULTIVATED FOR THEIR SEEDS. 891 an Arab physician of the thirteenth century, who had travelled through the countries lying between Spain and Persia, indicates no plant which can be supposed to be maize. J, Crawfurd,^ having seen maize generally cultivated in the Malay Archipelago under a name jaricng, which appears to be indigenous, believed that the species was a native of these islands. But then how is it Rumphius makes no mention of it. The silence of this author points to an introduction later than the seventeenth century. Maize was so little diffused on the continent of India in the last century, that Roxburgh ^ wrote in his flora, which was published long after it was drawn up, " Cultivated in different parts of India in gardens, and only as an ornament, but nowhere on the continent of India as an object of cultivation on a large scale." We have seen that there is no Sanskrit name. Maize is frequently cultivated in China in modern times, and particularly round Pekin for several genera- tions,^ although most travellers of the last century make no mention of it. Dr. Bretschneider, in his work pub- lished in 1870, does not hesitate to say that maize is not indiofenous in China; but some words in his letter of 1881 make me think that he now attributes some impor- tance to an ancient Chinese author, of whom Bonafous and afterwards Hance and Ma}' ers have said a great deal. This is a work by Li-chi-tchin, entitled Phen-thsao-kang- Tiiow, or Pen-tsao-kung-'inu, a species of treatise on natural history, which Bretschneider ^ says was written at the end of the sixteenth century. Bonafous says it was concluded in 1578, and the edition which he had seen in the Huzard library was of 1637. It contains a drawing of maize with the Chinese character. This plate is copied in Bonafous' work, at the beginning of the chapter on the original country of the maize. It is clear that it repre- * CrawfardjiJisfon/ oftJie Indian Archipelago, Edinbargb, 1820, voL i.; Journal of Botany, 18GG, p. 326. * Roxburgh, Flora Indica, edit. 1832, vol. iii. p. 5G3. ' Bretschueider, Study and Value, etc., pp. 7, 18. < Ibid. 392 ORIGIN OF CULTIVATED PLANTS. sents the plant. Dr. Hance^ appears to have based his arguments upon the researches of ]\rayers, who says that early Chinese authors assert that maize was imported from Sifan (Lower Mongolia, to the west of China) long before the end of the fifteenth century, at an unknown date. The article contains a copy of the drawing in the Pen-tsao-Jcung-onu, to which he assigns the date 1597. The importation through Mongolia is improbable to such a degree that it is hardly worth speaking of it, and as for the principal assertion of the Chinese author, the dates are uncertain and late. Tl)e work was finished in 1578 according to Bonafous, in 1597 according to Mayers. If this be tine, and especially if the second of these dates is the true one, it may be admitted that maize was brought to China after the discovery of America. The Portuguese came to Java in 149G,^ that is to say four years after the discovery of America, and to China in 1510.^ Magellan's voyage from South America to tlie Philippine Islands took place in 1520. During the fifty-eight or seventy-seven years between 1516 and the dates assigned to the Chinese work, seeds of maize may have been taken to China by navigators from America or from Europe. Dr. Bret- schneider wrote to me recently that the Chinese did not know the new world earlier than the Europeans, and that the lands to the east of their country, to which there are some allusions in their ancient writings, are the islands of Japan. He had already quoted the opinion of a Chinese savant, that the introduction of maize in the neighbourhood of Pekin dates from the last years of the Ming dynast3\ which ended in 161'4. This date agrees with the other facts. The introduction into Japan was probably of later date, since K?empfer makes no mention of the species.* From all these facts, we conclude that maize is not a native of the old world. It became rapidly diffused in it * The article is in the Fharwacevtical Journal of 1870; I onlj know it from a short extract in Seemaun's Journal of Botany, 1871, p. C2. * Rumphius, Amhoin., vol. v. p. 525. ' Malte-Brun, Geographie, i. p. 493. * A plant engi-aved on an ancient weapon which Sicbold had taken for maize is a sorghum, according to Eein, quoted by Wittmack, Uehjr Antiken Ma'is. PLANTS CULTIVATED FOR THEIR SEEDS. 393 after the discovery of America, and this very rapidity completes the proof that, had it existed anywhere in Asia or Africa, it would have played an important part in agriculture for thousands of years. We shall see that the facts are quite contrary to these in America. At the time of the discovery of the new continent, maize was one of the staples of its agriculture, from the La Plata valley to the United States. It had names in all the languages.^ The natives planted it round their temporary dwellings where they did not form a fixed population. The burial-mounds of the natives of North America who preceded those of our day, the tombs of the Incas, the catacombs of Peru, contain ears or grains of maize, just as the monuments of ancient Egypt contain grains of barley and wheat and millet-seed. In Mexico, a goddess who bore a name derived from that of maize {Cinteutl, from Cintli) answered to the Ceres of the Greeks, for the first-fruits of the maize harvest were offered to her, as the first-fruits of our cereals to the Greek goddess. At Cusco the virgins of the sun offered sacrifices of bread made from Indian corn. Nothing is better calculated to show the antiquity and generality of the cultivation of a plant than this intimate connection with the religious rites of the ancient inhabitants. We must not, however, attribute to these indications the same importance in America as in the old world. The civilization of the Peruvians under the Incas, and that of the Toltecs and Aztecs in Mexico, has not the extra- ordinary antiquity of the civilizations of China, Chaldea, and Egypt. It dates at earliest from the beginning of the Christian era; but the cultivation of maize is more ancient than the monuments, to judge from the numerous varieties of the species found in them, and their dispersal into remote regions. A yet more remarkable proof of antiquity has been discovered by Dai'win. He found ears of Indian corn, and eighteen species of shells of our epoch, buried in the soil of the shore in Peru, now at least eighty-five feet ' See Martius, Beitrdge zur Ethnograpliie AmeriJcas, p. 127. 394 ORIGIN OF CULTIVATED PLANTS. above the level of the sea.^ This maize was perhaps not cultivated, but in this case it would be yet more interesting, as an indication of the origin of the species. Although America has been explored by a great number of botanists, none have found maize in the conditions of a wild plant. Auo'uste de Saint-Hilaire ^ thous^ht he recoomized the wild type in a singular variety, of which ea:h grain is enclosed within its sheath or bract. It is known at Buenos-Ayres under the name pinsigallo. It is Zea Mays tunicata of Saint-Hilaire, of which Bonafous gives an illustration, pi. 5, his, under the name Zea cryptosperina, Lindley^ also gives a description and a drawing from seeds brought, it is said, from the Rocky Mountains, but this is not confirmed by recent Califomian Horas. A young Guarany, born in Paraguay on its frontiers, had recognized this maize, and told Saint-Hilaire that it grew in the damp forests of his country. This is very in- sufficient proof that it is indigenous. No traveller to my knowledge has seen this plant wild in Paraguay or Brazil. But it is an interesting fact that it has been cultivated in Europe, and that it often passes into the ordinary state of maize. Lindley observed it when it had been only two or three years in cultivation, and Professor Radic obtained from one sowing 225 ears of the form tunicata, and 105 of the common form with naked grains.* Evidently this form, which might be believed a true species, but whose country is, however, doubtful, is hardly even a race. It is one of the innumerable varieties, more or less hereditary, of which botanists who are con- sidered authorities make only a single species, because of their want of stability and the transitions which they frequently present. On the condition of Zea Mays, and its habitation in America before it was cultivated, we have nothing but con- * Darwin, Yar. nf Plants and Anim. under Domest., i. p. 320. ^ A. de Saint-Hilaire, Ann. Sc. Kat., xvi. p. 143. ' Lindley, Journ. of the Hortic. Soc, i. p. 114. * I quote these facts from Wittmack, Ueher Antilcen Ma'is aus Nord und Slid Amerika, p. 87, in Berlin Anthropol. Ges., Nov. 10, 1879. PLANTS CULTIVATED FOR THEIR SEEDS. 895 jectural knowledge. I will state what I take to be the sum of this, because it leads to certain probable indications. I remark first that maize is a plant singularly un- provided with means of dispersion and protection. The grains are hard to detach from the ear, which is itself enveloped. They have no tuft or wing to catch the wind, and when the ear is not gathered by man the grains fall still fixed in the receptacle, and then rodents and other animals must destroy them in quantities, and all the more that they are not sufficiently hard to pass intact through the digestive organs. Probably so unprotected a species was becoming more and more rare in some limited region, and was on the point of becoming extinct, when a wandering tribe of savages, having perceived its nutritious qualities, saved it from destruction by culti- vating it. I am the more disposed to believe that its natural area was small that the species is unique ; that is to say, that it constitutes what is called a single-typed genus. The genera w^hich contain few species, and especially the monotypes, have as a rule more restricted areas than others. Palaeontology will perhaps one day show whether there ever existed in America several species of Zea, or similar Graminae, of which maize is the last survivor. Now, the genus Zea is not only a monotype, but stands almost alone in its family. A single genus, Euchlcena of Schrader, may be compared with it, of which there is one species in Mexico and another in Guatemala ; but it is a quite distinct genus, and there are no inter- mediate forms between it and Zea. Wittmack has made some curious researches in order to discover which variety of maize probably represents the form belonging to the epoch anterior to cultivation. For this purpose he has compared ears and grains taken from the mounds of North America with those from Peru. If these monuments offered only one form of maize, the result would be important, but several different varieties have been found in the mounds and in Peru. This is not very surprising ; these monuments are not very ancient. The cemetery of Ancon in Peru, whence Wittmack obtained bis best specimens, is nearly contemporary with 18 396 ORIGIN OF CULTIVATED PLANTS. the discovery of America.^ Now, at that epoch the number of varieties was already considerable, which proves a much more ancient cultivation. Experiments in sowing varieties of maize in unculti- vated ground several years in succession would perhaps show a reversion to some common form which miofht then be considered as the original stock, but nothing of this kind has been attempted. The varieties have only been observed to lack stability in spite of their great diversity. As to the habitation of the unknown primitive form, the following considerations may enable us to guess it. Settled populations can only have been formed where nutritious sjoecies existed naturally in soil easy of cultivation. The potato, the sweet potato, and maize doubtless fulfilled these conditions in America, and as the great populations of this part of the world existed first in the high grounds of Chili and Mexico, it is there probably that wild maize existed We must not look for it in the low-lying regions such as Paraguay and the banks of the Amazon, or the hot districts of Guiana, Panama, and Mexico, since their inhabitants were formerly less nume- rous. Besides, forests are unfavourable to annuals, and maize does not thrive in the warm damp climates where manioc is grown.^ On the other hand, its transmission from one tribe to another is easier to comprehend if we suppose the point of departure in the centre, than if we place it at one of the limits of the area over which the species was cultivated at the time of the Incas and the Toltecs, or rather of the Mayas, Nahuas, and Chibchas, who preceded these. The migrations of peoples have not always followed a fixed course from north to south, or from south to north. They have taken different directions according to the epoch and the country.^ The * Rochebrune, Reclierches Ethnographiques surles Sipultures reruviennes d'Ancon, from an extract by Wittmack in Uhlworm, Bot. Central-Blatt., 1880, p. 1C33, where it may be seen that the burial-gruund was used before and after the discovery of America. ' Sagot, Cidt. des Cereales de la Guyane Franq. (Joum. de la Soc. Centr. d'Hortic. de France, 1872, p. 94). 3 De Naidaillac, in his' work entitled Les Premiers Hommes et les PLANTS CULTIVATED FOR THEIR SEEDS. 897 ancient Peruvians scarcely knew the Mexicans, and vice versa, as the total difference of their beliefs and customs shows. As they both early cultivated maize, we must suppose an intermediate point of departure. New Granada seems to me to fidfil these conditions. The nation called Chibcha which occupied the table-land of Bogota at the time of the Spanish conquest, and con- sidered itself aboriginal, was an agricultural people. It enjoyed a certain degree of civilization, as the monu- ments recently investigated show. Perhaps this tribe first possessed and cultivated maize. It marched with Peru, then but little civilized, on the one hand, and with the Mayas on the other, who occupied Central America and Yucatan. These were often at war with the Nahuas, predecessors of the Toltecs and the Aztecs in Mexico. There is a tradition that Nahualt, chief of the Nahuas, taught the cultivation of maize.^ I dare not hope that maize will be found wild, although its habitation before it w^as cultivated was probably so small that botanists have perhaps not yet come across it. The species is so distinct from all others, and so striking, that natives or unscientific colonists would have noticed and spoken of it. The certainty as to its origin will probably come rather from archseological discoveries. If a great number of monuments in all parts of America are studied, if the hieroglyphical inscriptions of some of these are deciphered, and if dates of migrations and economical events are discovered, our hypothesis will be justified, modified, or rejected. Article II. — Seeds used for Different Purposes. Poppy — Pajpaver somniferurti^ Linnaeus. The poppy is usually cultivated for the oil contained in the seed, and sometimes, especially in Asia, for the sap. Temps PreMstoriqxies, gives briefly the sum of our knowledge of these migrations of the ancient peoples of America in general. See especially vol. ii. chap. 0. * De Naidaillac, ii. p. G9, who quotes Bancroft, The NaHve Races of the Pacific States. 898 ORIGIN OF CULTIVATED PLANTS. extracted by making incisions in the capsules, and from which opium is obtained. The variety which has been cultivated for centuries escapes readily from cultivation, or becomes almost naturalized in certain localities of the south of Europe.^ It cannot be said to exist in a really wild state, but botanists are agreed in regarding it as a modification of the poppy called Papaver setigerum, which is wild on the shores of the Mediterranean, notably in Spain, Algeria, Corsica, Sicily, Greece, and the island of Cyprus. It has not been met with in Eastern Asia,^ consequently this is really the original of the cultivated form. Its cultivation must have begun in Europe or in the north of Africa. In support of this theory we find that the Swiss lake- dwellers of the stone age cultivated a poppy which is nearer to P. setlgeruin than to P. somniferum. Heer^ has not been able to find any of the leaves, but the capsule is surmounted by eight stigmas, as in P. setigerum, and not by ten or twelve, as in the cultivated poppy. This latter form, unknown in nature, seems therefore to have been developed within historic times. P. sctigerwiu is still cultivated in the north of France, together with P. soinniferwin, for the sake of its oil.* The ancient Greeks were well acquainted with the cultivated poppy. Homer, Theophrastus, and Dioscorides mention it. They were aware of the somniferous pro- perties of the sap, and Dioscorides ^ mentions the variety with white seeds. The Romans cultivated the poppy before the republic, as we see by the anecdote of Tarquin and the poppy-heads. They mixed its seeds with their flour in makins: bread. The Egyptians of Pliny's time ^ used the juice of the poppy as a medicament, but w^e have no proof that this ' Willkomm and Lange, Prodr. Fl. Hisp., iii. p. 872. ' Boissier, Fl. Orient.; TcliihatchefE, Aaie Mineure; Ledebour, Fl. Ross., and others. * Heer, Pjianzen der Pfalilhanfen, p. 32, figs. 65, 66. * De Lanessan, in his translation from Fliickiger and Hanburj, His- toire des Drogues d'Origine Vegdtale, i. p. 129. * Dioscorides, Hist. Plant., lib. iv. c. 65. « Pliny, Hist. Plant., lib. xx. c. 18. PLANTS CULTIVATED FOR THEIR SEEDS. 890 plant was cultivated in Egypt in more ancient times.^ In the Middle Ages ^ and in our own day it is one of the principal objects of cultivation in that country, especially for the manufacture of opium. Hebrew writings do not mention the species. On the other hand, there are one or two Sanskrit names. Piddington gives chosa, and Adolphe Pictet Jchasl'Jiasa, which recurs, he says, in the Persian chaskchdsh, the Armenian chasJichash'^ and in Arabic. Another Persian name is Jcouknar.^ These names, and others I could quote, very different from the maihon (M>'/kwv) of the Greeks, are an indication of an ancient cultivation in Europe and Western Asia. If the species was first cultivated in prehistoric time in Greece, as appears probable, it may have spread eastward before the Aryan invasion of India, but it is strange that there should be no proof of its extension into Palestine and Egypt before the Roman epoch. It is also possible that in Europe the variety called Fapaver setigeruin, emj^loyed by the Swiss lake-dwellers, was first cultivated, and that the variety now grown came from Asia Minor, where the species has been cultivated for at least three thousand years. This theory is supported by the existence of the Greek name raaikon, in Dorian Tnakon, in several Slav languages, and in those of the peoples to the south of the Caucasus, under the form mack.^ The cultivation of the poppy in India has been recently extended, because of the importation of opium into China ; but the Chinese will soon cease to vex the English by buying this poison of them, for they are be- ginning eagerly to produce it themselves. The poppy is now grown over more than half of their territory.^ The species is never wild in the east of Asia, and even as reo:ards China its cultivation is recent."^ * linger, Die Pflanze als Errerungs und Betaiihungsmittel, p. 47 j Die Pflanzen des Alien jEgyptens,!.^). 50. * Ebn Baithar, German trans., i. p. 64. ' Ad. Pictet, Origines Indo-Europeennes, edit. 3, vol. i. p. 3GS. * Ainslie, Mat. Med. Indica, i. p. 326. * Nemnich, Polygl, Lexicon, p. 848. * Martin, in Bull. Soc. d'Acdimafation, 1872, p. 200. ' Sir J. Hooker, Flora of Brit. Ind., i. p. 117 j Bretschneidcr, Study and Value, etc., 47. 400 ORIGIN OF CULTIVATED PLANTS. The name opium given to the drug extracted from the juice of the capsule is derived from the Greek. Dios- corides wrote oj^os (Ottoc) The Arabs converted it into ojiun} and spread ifc eastwards even to China. Fluckiger and Hanbury ^ give a detailed and interest- ing account of the extraction, trade, and use of opium in all countries, particularly in China. Yet I imagine my readers may like to read the following extracts from Dr. Bretschneider's letters, dated from Pekin, Aug. 23, 1881, Jan. 28, and June 18, 1882. They give the most certain information which can be derived from accurately translated Chinese works. *' The author of the Pent-sao-kang-moii, who wrote in 1552 and 1578, gives some details concerning the a-fou- yong (that is afioun, opiun), a foreign drug produced by a species of ying-soib with red flowers in the country of Tien-fang (Arabia), and recently used as a medicament in China. In the time of the preceding dynasty there had been much talk of the a-foii-yong. The Chinese autlior gives some details relative to the extraction of opium in his native country, but he does not say that it is also pro- duced in China, nor does he allude to the practice of smoking it. In the Descriptive Dictionary of the Indian Islands, by CraAvfurd, p. 312, I find the following pas- sage : * The earliest account we have of the use of opium, not only from the Archipelago, but also from India and China, is by the faithful, intelligent Barbosa.^ He rates it among the articles brought by the Moorish and Gentile merchants of Western India, to exchange for the cargoes of Chinese junks.' " "It is difficult to fix the exact date at which the Chinese began to smoke opium and to cultivate the poppy which produces it. As I have said, there is much confusion on tliis head, and not only European authors, but also the modern Chinese, apply the name ying-sou to P. somniferwiTi as well as to P. rhccas. P. somni- ferum is now extensively cultivated in all the provinces 1 Ebn Baithar, i. p. 64. 2 Fliickigor aud Hanbury, Pharmacogrnphia, p. 4.0. * Barbosa's work "vvas publishod in 1516, PLANTS CULTIVATED FOR THEIR SEEDS. 401 of the Chinese empire, and also in Mantchuria and Mon- golia. Williamson {Journeys in North China, Mant- churia, Mongolia, 1868, ii. p. 55) saw it cultivated every- whei*e in Mantchuria. He was told that the cultivation of the poppy was twice as profitable as that of cereals. Potanin, a Russian traveller, wdio visited Northern Mon- golia in 1876, saw immense plantations of the poppy in the valley of Kiran (between lat. 47° and 48°), This alarms the Chinese government, and still more the Eng- lish, who dread the competition of native opium." " You are probably aware that opium is eaten, not smoked, in India and Persia, The practice of smoking this drug appears to be a Chinese invention, and modern. Nothing proves that the Chinese smoked opium before the middle of the last century. The Jesuit missionaries to China in the seventeenth and eiMiteenth centuries do not mention it; Father d'Incarville alone sa3^s in 1750 that the sale of opium is forbidden because it was used by suicides. Two edicts forbidding the smoking of opium date from before 1730, and another in 1796 speaks of the progress made by the vice in question. Don Sinibaldo di Mas, who in 1858 published a very good book on China, where he had lived many years as Spanish ambassador, says that the Chinese took the practice from the people of Assam, where the custom had long existed." So bad a habit, like the use of tobacco or absinth, is sure to spread. It is becoming gradually introduced into the countries which have frequent relations with China. It is to be hoped that it will not attack so large a proportion of the peoples of other countries as in Amoy, where the proportion of opium-smokers are as fifteen to twenty of the adult population.^ Amotto, or Anatto — Bisca orellana, Linnaeus. The dye, called rocovu in French, amotto in English, is extracted from the pulp which encases the seed. The inhabitants of the West India Islands, of the Isthmus of Darien, and of Brazil, used it at the time of the discovery of America to stain their bodies red, and the Mexicans * Hughes, Trade Report, quoted by Fliickiger and Uanburj. 402 ORIGIN OF CULTIVATED PLANTS. in painting.^ The arnotto, a small tree of the order Bixacece, grows wild in the West Indies,^ and over a great part of the continent of America between the tropics. Herbaria and floras abound in indications of locality, but do not generally specify whether the species is cultivated, wild, or naturalized. I note, however, that it is said to be indigenous by Seemann on the north- west coast of Mexico and Panama, by Triana in New Granada, by Me^^er in Dutch Guiana, and by Piso and Claussen in Brazil.^ Witli such a vast area, it is not surprising that the species has many names in American languages ; that of the Brazilians, urucu, is the origin of rocou. It w^as not very necessary to plant this tree in order to obtain its product ; nevertheless Piso relates that the Brazilians, in the sixteenth century, were not content with the wild plant, and in Jamaica, in the seventeenth century, the plantations of Bixa were common. It was one of the first species transported from America to the south of Asia and to Africa. It has become so entirely naturalized, that Koxburgh ^ believed it to be indigenous in India. Cotton — Gossyinum herhaceum, Linnoeus. When, in 1855, I sought the origin of the cultivated cottons,^ there was still great uncertainty as to the dis- tinction of the species. Since then two excellent works have appeared in Italy, upon which we can rely ; one by Parlatore,^ formerly director of the botanical gardens at Florence, the other by Todaro,"^ of Palermo. These two * Sloane, Jamaica, ii. p. 53. ^ Sloaue, ihid. ; Clos, Ann. Sc. Kat., 4th series, vol. viii. p. 2G0 ; Grisebach, J'L of Brit. W. Ind. Is., p. 20. ^ Seemann, Bot. of Herald., pp. 79, 2fi8 ; Triana and Planchon, Prodr. Fl. Novo-Grajiat., p. 9-1; Meyer, Essequcbo, p. 202; Pido, Hist. Nat, Brasil, edit. 1648, p. 65 ; Claussen, in Clos, tthi svpra. * Eoxburgh, Fl. Ind., ii. p. 581; Oliver, Ft. Trap. Africa, i. p. 114. * Gcogr. Bot. Rais., p. 971. ® Parlatore, Le Specie dei Cotoni, text in 4to, plates in folio, Florence, 1866. ' Todaro, Relazione della CoUura dci Cotoni in I>alia, serjnita da una lyTonographia del Genere Gossifpium, text large 8vo, plates in folio, Rome and Palermo, 1877-78 ; a -work preceded by several others of lesa im- portance, which were known to Parlatore. PLANTS CULTIVATED FOR THEIR SEEDS. 403 works are illustrated with magnificent coloured plates. Nothing better can be desired for the cultivated cottons. On the other hand, our knowledge of the true species, I mean of those which exist naturally in a wild state, has not increased as much as it might. However, the definition of species seems fairly accurate in the works of Dr. Masters,^ whom I shall therefore follow. This author asfrees with Parlatore in admittino^ seven well- known species and two doubtful, while Todaro counts fifty-four, of which only two are doubtful, reckoning as species forms with some distinguishing character, but which originated and are preserved by cultivation. The common names of the cottons give no assistance ; they are even calculated to lead us completely astray as to the origin of the species. A cotton called Siamese comes from America ; another is called Brazilian or Ava cotton, according to the fancy or the error of cultivators. We will first consider Gossypium herhacewin, an ancient species in Asiatic plantations, and now the com- monest in Europe and in the United States. In the hot countries whence it came, its stem lasts several years, but out of the tropics it becomes annual from the effect of the winter's cold. The flower is generally yellow, with a red centre; the cotton yellow or white, according to the variety. Parlatore examined in herbaria several wild specimens, and cultivated others derived from wild plants of the Indian Peninsula. He also admits it to be indigenous in Burmah and in the Indian Archipelago, from the specimens of collectors, who have not perhaps been sufiiciently careful to verify its wild character. Masters regards as undoubtedly wild in Sindh a form which he calls Gossypiuiin Stocksii, which he says is probably the wild condition of Gossypiuni hevhaceum, and of other cottons cultivated in India for a lone: time. Torlaro, who is not g'iven to unitini^ manv forms in a single species, nevertheless admits the identity of this variety with the common G. Iterhaceum. The yellow colour of the cotton is then the natural condition of the ' Masters, in Oliver, Fl. Trop. Afr., i. p. 210; and in Sir J. Hooker, Fl. Brit. Ind., i. p. 31G. 404! ORIGIN OF CULTIVATED PLANTS. species. The seed has not the short down which exists between the Ioniser hairs in the cultivated G. herbaceuiii. Cultivation has probably extended the area of the species beyond the limits of the primitive habitation This is, I imaoine, the case in the Sunda Islands and the Malay Peninsula, where certain individuals appear more or less wild. Kurz/ in his Burmese flora, mentions G. hevhacewm, with yellow or white cotton, as cultivated and also as wild in desert places and waste ground. The herbaceous cotton is called hapase in Bengali, Jcapas in Hindustani, which shows that the Sanskrit word karpassi undoubtedly refers to this species.^ It was early cultivated in Bactriana, where the Greeks had noticed it at the time of the expedition of Alexander. Theophrastus speaks of it ^ in such a manner as to leave no doubt. The tree-cotton of the Isle of Tylos, in the Persian Gulf, of wliich he makes mention further on,* was probably also G. herbaceum; for Tylos is not far from India, and in such a hot climate the herbaceous cotton becomes a shrub. The introduction of a cotton plant into China took place only in the nintli or tenth century of our era, which shows that probably the area of G. herhaceum was originally limited to the south and east of India. The knowledge and perhaps the cultiva- tion of the Asiatic cotton was propagated in the Gryeco- Boman world after the expedition of Alexander, but before the first centuries of the Christian era.^ If the hyssos of the Greeks was the cotton plant, as most scholars think, it was cultivated at Elis, according to Pausanias and Pliny ;^ but Curtius and C. Pwitter' con- sider the word hyssos as a general term for threads, and that it was probably applied in this case to fine linen. It is evident that the cotton was never, or very rarely, cultivated by the ancients. It is so useful that it would have become common if it had been introduced * Kiirz, Forest Flora of British Bin-mah, i. p. 129. * Piddiiif^ton, Index. ' Theophrastus, Hi^t. Plant., lib. iv. cap. 5. * Ibid:, lib. iv. cap. 9. * ]3rct.schneider, Sttnly and VaJue, etc., p. 7. * Pausanias, lib. v., cap. 5 ; lib. vi. cajJ. 26 ; Pliny, lib. xix. cap. 1. See Brandes, BaiomcoUe, p. 90. ' C. Ritter, Die Geographisclie Verhreitung der Baumwolle, p. 25. PLANTS CULTIVATED FOR THEIR SEEDS. 405 into a single locality — in Greece, for instance. It was afterwards propagated on the shores of the Mediterranean by the Arabs, as we see from the name qutn or kutn,^ which has passed into the modern languages of the south of Europe as cotone, colon, algodon. Eben el A wan, of Seville, who lived in the twelfth century, describes its cultivation as it was practised in his time in Sicilj^, Spain, and the East.^ Gossypium herbaceum is the species most cultivated in the United States.^ It was probably introduced there from Europe. It was a new cultivation a hundred years ago, for a bale of North American cotton w^as confiscated at Liverpool in 1774, on the plea that the cotton-plant did not grow there.^ The sillcy cotton {sea island) is another species, American, of which I shall presently speak. Tree-Cotton — Gossypium arbor etcm, Linn?eus. This species is taller and of longer duration than the herbaceous cotton ; the lobes of the leaf are narrower, the bracts less divided or entire. The flower is usually pink, with a red centre. The cotton is always white. According to Anglo-Indian botanists, this is not, as it was supposed, an Indian species, and is even rarely cultivated in India. It is a native of tropical Africa. It has been seen wild in Upper Guinea, in Abyssinia, Sennaar, and Upper Egypt.^ So great a number of collectors have brought it from these countries, that there is no room for doubt; but cultivation has so diffused and mixed this species with others that it has been described under several names in works on Southern Asia. * It Is impossible not to remark the resemblance between this name and that of flax ia Arabic, kattan or kittan; it is an example of the con- fusion which takes place in names where there is an aualogy between the prodncts. * De Lasteyrie, Du Cotonnier, p. 290. * Torrey and Asa Gray, Flora of North America^ i. p. 230 ; Darling- ton, Agricultural Botany, p. 16. * Schouw, Naturschilderujigen, p. 152. 5 Masters, in Oliver, Fl. Trop. Afr.yi. p. 211 ; Hooker, Fl. of Brit. Ind., i. p. 347; Schweirifurth and Ascherson, Aufzdhlung, p. 265 (under the name Gossypium nigrum) ; Parlatore, Specie dci Cotoni, p. 25. 40G ORIGIN OF CULTIVATED PLANTS. Parlatore attributed to G. arhoreurti some Asiatic specimens of G. heihaceuW; and a plant but little known which Forskal found in Arabia. He suspected from this that the ancients had known G. arhoreiun as well as G. herhacemn. Now that the two species are better distin- guished, and that the origin of both is known, this does not seem probable. They knew the herbaceous cotton through India and Persia, while the tree-cotton can only have come to them through Egypt. Parlatore himself has given a most interesting proof of this. Until his Avork appeared in 1866, it was not certain to what species belonged some seeds of the cotton plant which Rosellini found in a vase among the monuments of ancient Thebes.^ These seeds are in the Florence museum. Parlatore examined them carefully, and declares them to belong to Gossypiiun arboreivm} Kosellini is certain he was not imposed upon, as he was the first to open both the tomb and the vase. No archaeologist has since seen or read signs of the cotton ])lant in the ancient times of Egyptian civilization. How is it that a plant so striking, remark- able for its flowers and seed, w^as not described nor pre- served habitually in the tombs if it were cultivated ? How is it that Herodotus, Dioscorides, and Theophrastus made no mention of it when writing of Egypt ? The cloths in whicli all the mummies are wrapt, and which were formerly supposed to be cotton, are always linen according to Thompson and many other observers who are familiar Avith the use of the microscope. Hence 1 conclude that if the seeds found by Rosellini were really ancient they were a rarity, an exception to the common custom, perhaps the product of a tree cultivated in a garden, or perhaps they came from Upper Egypt, a country Avhere we know the tree-cotton to be wild. Pliny ^ does not say that cotton was cultivated in Lower Egypt; but here is a translation of his very remarkable passage, which is often quoted. " The upper part of Egypt, towards Ai^abia, produces a shrub which some ' Eo?ellini, Momimenti dell' Egizia, p. 2; Mon. Civ., i. p. GO. ^ Parlatore, Specie dei Cotoni, p. 16. ^ Pliny, Hist. Plant., lib. xix. cap. 1. PLANTS CULTIVATED FOR THEIR SEEDS. 407 call gossijnon and others xylon, whence the name xylina given to the tlircads obtained from it. It is low- growing, and bears a fruit like that of the bearded nut, and from the interior of this is taken a wool fur weaving. None is comparable to tl.is in softness and whiteness." Pliny adds, "The cloth made from it is used by preference for the dress of the Egyptian priests." Perhaps the cotton destined to this purpose was sent from Upper Egypt, or perhaps the author, who had not seen the fabrication, and did not possess a micro- scope, was mistaken in the nature of the sacerdotal I'aiment, as were our contemporaries who handled the grave-cloths of hundreds of mummies before suspecting that they were not cotton. Among the Jews, the priestly robes were commanded to be of linen, and it is not likely that their custom v/as different to that of the Egyptians. Pollux,^ born in Egypt a century later than Pliny, expresses himself clearly about the cotton plant, of which the thread was used by his countrymen ; but he does not say whence the shrub came, and we cannot tell whether it was Gossypium arbor eiini or G. herhacemn. It does not even appear whether tlie plant was cultivated in Lower Egypt, or if the cotton came from the more southern region. In spite of these doubts, it may be suspected that a cotton plant, probably that of Upper Egypt, had recently been introduced into the Delta* The species which Prosper Alpin had seen cultivated in Egypt in the sixteenth century was the tree-cotton. The Arabs, and afterwards Europeans, preferred and trans- ported into different coimtries the herbaceous cotton rather than the tree-cotton, which yields a poorer product and requires more heat. Reoardino^ the two cottons of the old world, I have made as little use as possible of arguments based upon Greek names, such as /3u(tctoc, glv^ov, ^uXov, OOwv, etc., or Sanskrit names, and their derivatives, as carhasa, car pas, or Hebrew names, schesch, buz, wliieh are doubt- fully attributed to the cotton tree. This has been a * Pollux, Onomasticon, quoted by C. Ritter, uhi supra, p. 26. 408 OlllGIN OF CULTIVATED PLANTS. fruitful subject of discussion,^ but the clearer distinction of species and the discovery of their ori^^in greatly diminishes the importance of these questions — to natu- ralists, at least, who prefer facts to words. Moreover, Reynier, and after him C. Ritter, arrived in their re- searches at a conclusion which we must not foriret : that these same names were often applied by ancient peoples to different plants and tissues — to linen and cotton, for example. In this case as in others, modern botany explains ancient words where words and the com- mentaries of philologists may mislead. Barbados Cotton — Gossypiiim barbadense, Linnreus. At the time of the discovery of America, the Spaniards found the cultivation and use of cotton established from the West India Islands to Peru, and from Mexico to Brazil. "J'he fact is proved by all the historians of the epoch. But it is still very dithcult to tell what were tlie species of these American cottons and in what countries they were indigenous. The botanical distinction of the American species or varieties is in the last degree con- fused. Authors, even those who have seen larsre collec- tions of growing cotton plants, are not agreed as to the characters. They are also embarrassed by the difficulty of deciding which of the specific names of Linnaeus should be retained, for the original definitions are insufficient. The introduction of American seed into African and Asiatic plantations has given rise to further complica- tions, as botanists in Java, Calcutta, Bourbon, etc., have often described American forms as species under ditferent names. Todaro admits ten American species ; Parlatore reduced them to three, which answer, he says, to Gosst/- jnuon hirsiituon, G. barbadense, and G. religiosum of Linnaeus ; lastly. Dr. Masters unites all the American forms into a single species which he calls G. barbadense, giving as the chief character that the seed bears only * Revnier, Economic des Arahrs et des Juifs., p. 3G3 ; Bertoloni, Nov. Act. Acad. Bonon., ii. p. 213, and Miscell. Bot., G ; Viviani, in Bihl. Ital., vol. Ixxxi. p. 9i; C. Ritter, Geogr. Verhreitung der Bauynironc, in 4to. ; Targioni, Cenni Storici, -p, 93; Braudis, Der Baumwolle in AHerthum, in 8vo, 1880. PLANTS CULTIVATED FOR THEIR SEEDS. 409 long hairs, wlievcas tlie species of the old world have a short down underneath the h)no;cr hairs.^ The tlower is yellow, with a red centre. The cotton is white or yellow. Parlatore strove to include fifty or sixty of the cultivated forms under one or other of the three heads he admits, from the study of plants in gardens or herbaria. Dr. Masters mentions hut few synonyms, a.nd it is possible that certain forms with which he is not acquainted do not come under the definition of his single species. Where there is such confusion it would be the best course for botanists to seek with care the Gossypia, which are wild in America, to constitute the one or more species solely upon these, leaving to the cultivated species their strancje and often absurd and misleadinsf names. I state this opinion because with regard to no other genus of cultivated plants have I felt so strongly that natural history should be based upon natural facts, and not upon the artificial products of cultivation. If we start from this point of view, which has the merit of being a truly scientific method, we find unfortunately that our know- ledofe of the cottons indiixenous in America is still in a very elementary state. At most we can name only one or two collectors who have found Gossypia really identical with or very similar to certain cultivated forms. We can seldom trust early botanists and travellers on this head. The cotton plant groAvs sometimes in the neighbourhood of plantations, and becomes more or less naturalized, as the down on the seeds facilitates accidental transport. The usual expression of early writers — such a cotton plant grows in such a country — often means a cultivated plant. Linnseus himself in the eighteenth century often says of a cultivated species, " habitat^' and he even says it sometimes without good ground.^ Hernandez, one of the most accurate among sixteenth- century authors, is quoted as having described and figured a Avild Gossypiii'tn in Mexico, but the text * Masters, in Oliver, Flora of Trop. Africa, i. p. 322 ; and in Hooker, Flora of Brit. India, i. p. 347. * He says, for instance, of Go^syphivi herhaceum, ■which is certainly of the old world, as facts known before his time show, " habitat in America," 410 ORIGIN OF CULTIVATED PLANTS. suof2fcsts some doubts as to the wild condition of this plant/ which Parlatore believes to be G. hirsutiun, Linnaeus. Hemsley,^ in his catalogue of Mexican plants, merely says of a Gossypium which he calls harbadense, " wild and cultivated " He gives no proof of the former condition. Macfadyen^ mentions three forms wild and cultivated in Jamaica. He attributes specific names to them, and adds that they possibly all may be included in Linnaeus' G. hirsutmn. Grisebach ^ adn\its that one species, G. harbadense, is wild in the West Indies. As to the specific distinctions, he declares himself unable to estal)lish them with certainty With regard to New Grenada, Triana^ describes a Gossypiiim which he calls G, harbadense, LinniBus, and which he says is " cultivated and half wild along the Rio Seco, in the province of Bogota, and in the valley of the Cauca near Cali ; " and he adds a variety, hirsiUitm, growing (he does not say whether spontaneously or no) along the Ivio Seco. I cannot discover a.ny similar asser- tion for Peru, Guiana, and Brazil ; ^ but the flora of Chili, published by CI. Gay,"^ mentions a Gossypinra, "almost wild in the province of Copiapo," which the writer attributes to the variety G. iieruvianiiin, Cavanilles. Now, this author does not say the plant is wild, and Parlatore classes it with G. relirjiosum, Linnaeus. An important variety of cultivation is that of the cotton with long silky down, called by Anglo-Americans sea island, or long staple cotton, which Parlatore ranks Avith G. harbadense, Linnaeus. It is considered to be of American origin, but no one has seen it wild. In conclusion, if historical records are positive in all that concerns the use of cotton in America from a time far earlier than the arrival of Europeans, the natural * Nascitur in calidis Inimidisque cultis prxcipue locis (Hernandez, Novce Hispanice Thesaarus}, p. 308). ' Hemsley, Biologia Cenirali- Americana, i. p. 123. ^ Macfadyen, Flora of Jamaica, p. 72. * Gnsebacli, Flora of Brit. W. India Is., p. 86. * Ti-iana and P'.anclion, Prodr. Fl. Noro.Granatcnf^'iori to the idea of an American origin. We must remember that MarcG^raf ^ and Piso ^ describe and fioure the plant as used in Brazil, under the name "manduhi, which seems to be indigenous. They quote Monardes, a w^'iter of the end of the sixteenth centurv, as havinor indicated it in Peru under a different name, anchic. Joseph Acosta ^ merely mentions an American name, onani, and speaks of it wdth other species which are not of foreim orio-in in America. The Arachis was not ancient in Guiana, in the West Indies, and in Mexico. Aublet ^ mentions it as a cultivated plant, not in Guiana, but in the Isle of France. Hernandez does not speak of it. Sloane^ had seen it only in a garden, grown from seeds brought from Guinea. He says that the slave- dealers feed the negroes with it on their passage from * Bentham, in Trayis. Linn. Soc, xviii. p. 159; Walpers, Repertojiiayi, i. p. 727. * Marcgiaf and Piso, Brasil., p. 37, edit. 1648. 3 Ibid., edit. 1658, p. 256. * Acosta, Hist. Nat. T7id., French, trans., 1598, p. 165. * Aublet, PL Guyauy p. 705. * Sloane, Jamaica, p. 184. PLANTS CULTIVATED FOR THEIR SEEDS. 413 Africa, wliicli indicates a then very general cultivation in Africa. Pison, in his second edition (1G58, p. 250), not in that of 1()48, o-ives a fimire of a simihxr fruit im- ported from Africa into Brazil under the name niandohi, very near to the name of the Arachis, iniindiibi. From the three Leaflets of the plant it would seem to be the Voandzeia, so often cultivated; but the fruit seems to me to be longer than in this genus, and it has two or three seeds instead of one or two. However this may be, the distinction drawn by Piso between these two subterranean seeds, the one Brazilian, the other Africar, tends to show that the Arachis is Brazilian. " The antiquity and the generality of its cultivation in Africa is, however, an argument of some force, which compensates to a certain degi-ee its antiquity in Brazil, and the presence of six other Arachis in the same country. I would admit its great value if the Arachis had been known to the ancient Egyptians and to the Arabs ; but the silence of Greek, Latin, and Arab authors, and the absence of the species in Egypt in Forskal's time, lead me to think that its cultivation in Guinea, Senegal,^ and the east coast of Africa^ is not of very ancient date. Neither has it the marks of a great antiquity in Asia. No Sanskrit name for it is known,^ but only a Hindu- stani one. Rumphius * says that it was imported from Japan into several islands of the Indian Archipelago. It would in that case have borne only foreign names, like the Chinese name, for instance, which signifies only ' earth -bean.' At the end of the last century it was generally cultivated in China and Cochin-China. Yet, in spite of Rumphius's theory of an introduction into the islands from China or Japan, I see that Thunberg does not speak of it in his Jajxtiiese Flora. Now, Japan has had dealings with China for sixteen centuries, and culti- vated plants, natives of one of the two countries, were commonly early introduced into the other. It is not mentioned by Eorster among the plants employed in the * Guillcrain and Perrottet, Fl. Senegal. * Lonreiro, Fl. Cochin. > * Roxburp^^b, Fl. Ind., iii. p. 280 ; Piddington, Index. * Rumphius, Ue.b. Amb., y. p. 426 414 ORIGIN OF CULTIVATED PLANTS. small islands of the Pacific. All these facts point to an American, I might even say a Brazilian, origin. None of the authors I have consulted mentions havinof seen the plant wild, either in the old or the new world. Those who indicate it in Africa or Asia are careful to say the plant is cultivated. Marcgraf does not say so, writing of Brazil, but Piso sa.ys the species is planted." Seeds of Aracliis have been found in the Peruvian tombs at Ancon,^ which shows some antiquity of existence in America, and supports the opinion I expressed in 1855. Dr. Bretschneider's study of Chinese w^orks ^ over- sets Brown's hypothesis. The Aracliis is not mentioned in the ancient works of this country, nor even in the Pent-sao, published in the sixteenth century. He adds that he believes the plant was only introduced in tlie last century. All the recent floras of Asia and Africa mention the species as a cultivated one, and most authors believe it to be of American origin. Bentham, after satisfying himself that it had not been found wild in America or elsewhere, adds that it is perhaps a form derived from one of the six other species wild in Brazil, but he does not say which. This is probable enough, for a plant provided Avith an efficacious and very peculiar manner of germinating does not seem of a nature to become extinct. It would have been found wild in Brazil in the same condition as the cultivated plant, if the latter were not a product of cultivation. Works on Guiana and other parts of America mention the species as a cultivated one ; Grisebach ^ says, moreover, that in several of the West India islands it becomes naturalized from cultivation. A genus of which all the well-known species are thus placed in a single region of America can scarcely have a species common to both hemispheres ; it would be too * RochoLrune, from the extract in the Botanuclies Centralblatt, 1880, p. lL3k ^ Study and Value of Chinese Botanical Works, p. 18. ' Grisebach, FL Brit. W. Ind. Is., p. 189. PLANTS CULTIVATED FOR THEIR SEEDS. 415 great an exception to the law of geographical botany. But then how did the species (or cultivated variety) pass from the American continent to the old Avorld ? This is hard to sfuess. but I am inclined to believe that the first slave-ships carried it from Brazil to Guinea, and the Portuo'uese from Brazil into the islands to the south of Asia, in the end of the fifteenth century. Coffee — Cofea arahica, Linnaeus. This shrub, belonging to the family of the Rubiace?e, is wild in Abyssinia,^ in the Soudan,'^ and on the coasts of Guinea and Mozambique.^ Perhaps in these latter localities, so far removed from the centre, it may be naturalized from cultivation. No one has yet found it in Arabia, but this may be explained by the difficulty of penetrating into the interior of the country. If it is discovered there it will be hard to prove it wild, for the seeds, which soon lose their faculty of germinating, often spring up round the plantations and naturalize the species. This has occurred in Brazil and the West India Islands,* where it is certain that the coffee plant was never indigenous. The use of coffee seems to be very ancient in Abys- sinia. Shehabeddin Ben, author of an Arab manuscript of the fifteenth century (No. 941 of the Paris Library), quoted in John Ellis's excellent work,^ says that coffee had been used in Abyssinia from time immemorial. Its use, even as a drug, had not spread into the neighbouring countries, for the crusaders did not know it, and the celebrated physician Ebn Baithar, born at Malaga, who had travelled over the north of Africa and Syria at the beginning of the thirteenth century of the Christian era, does not mention coffee.*^ In 159G Bellus sent to de I'Ecluse some seeds from which the Egyptians ex- » Richard, Tentamen Fl. Abyss., i. p. 349; Oliver, Ft. Trop. Afr., iii. p. 180. 2 Ritter, quoted in Flora, 181G, p. 704. ' Meyen, Geogr. Bot., English trans., p. 381; Grisobach, Fl. of Brit. W. Ind. Is., p. 338. "» H. Welter, Essai sur VHistoire du Cafe, 1 vol. in 8vo, Paris, 1868. * Ellis, An Historical Account of Coffee, 1774. • Ebn Baithar, Sondtheimer's trans., 2 vols. Svo, 1842. 416 . ORIGIN OF CULTIVATED PLANTS. tracted the drink cave} Nearly at the same time Prosper Alpin became acquainted with coffee in Egypt itself. He speaks of the ])lant as the " arbor hon, cum f ructu suo huna!' The name hon i-ecurs also in early authors under the forms hunnu, huncho, hunca.^ The names cahiie, cahua, chauhe^ cave,^ refer rather in Egypt and Syria to the prepared drink, whence the French word cafe. The name hunnu, or something similar, is certainly the primi- tive name of the plant which the Abyssinians still call houn.^ If the use of coffee is more ancient in Abyssinia than elsewhere, that is no proof that its cultivation is very ancient. It is very possible that for centuries the berries were sought in the forests, where they were doubtless very common. According to the Arabian author quoted above, it was a mufti of Aden, nearly his contemporary, who, having seen coffee drunk in Persia, introduced the prac- tice at Aden, whence it spread to ^locha, into Egypt, etc. He says tliat the coffee plant grew in Arabia.^ Other fables or traditions exist, accordino; to which it was alwa^^s an Arabian priest or a monk who invented the drink,'' but they all leave us in uncertainty as to the date of the first cultivation of the plant. However this may be, the use of coffee having been spread first in the east, afterwards in the west, in spite of a number of prohibitions and absurd conflicts,** its production became important to the colonies. Boerhave tells us that the Burgermeister of Amsterdam, Nicholas Witsen, director of the East India Company, urged the Governor of Batavia, Van Hoorn, to import coliee berries from Arabia to Batavia. This was done, and in IGOO Van Hoorn sent some living plants to Witsen. These were placed in the Botanical Gardens of Amsterdam, founded by Witsen, where they bore fruit. In 1714, the magistrates of the ' Bellas, Epist. ad Clus., p. 309. « Eauwolf, Clusina. ' Kauwolf ; Bauhin, Hist., i. p. 422. * Bellas, uhi supra, * Richard, Tentamen Fl. Abyss., p. 350. • An extract from the same author in Playfair, Uist. qf Arabia Felix, Bombay, 1859, does not mention this assertion.. 7 Nouv. Diet. d'Hist. Nat., iv. p. 552. 8 Ellis, ubi supra; Nouv. Did., ibid. PLANTS CULTIVATED FOR THEIR SEEDS. 41/ town sent a flourisliing plant covered with fruit to Louis XIV., who placed it in his garden at Marly. Coffee was also grown in the hothouses of the king's garden ^ in Paris. One of the professors of this establishment, Antoine de Jussieu, had already published in 1713, in the Memoires de VAcademie des Sciences, an interesting- description of the plant from one which Pancras, director of the Botanical Garden at Amsterdam, had sent to him. The first coffee plants grown in America were intro- duced into Surinam by the Dutch in 1718. The Governor of Cayenne, de la Motte-Aigron, having been at Suri- nam, obtained some plants in secret and multiplied them in 1725.^ The coffee plant was introduced into Mar- tinique by de Clieu,^ a naval officer, in 1720, according to Deleuze f in 1723, according to the Notices StcUistiques SUV les Colonies Fvanqaises,^ Thence it was introduced into the other French islands, into Guadaloupe, for in- stance, in ]730.^ Sir Nicholas Law^es first grew it in Jamaica.^ From 1718 the French East India Companj^ had sent plants of Mocha coffee to Bourbon ;'^ others say ^ that it was even in 1717 that a certain Dufougerais- Grenier had coffee plants brought from Mocha into this island. It is known how the cultivation of this shrub has been extended in Java, Ceylon, the West Indies, and Brazil. Nothing prevents it from spreading in nearly all tropical countries, especially as the coffee plant thrives * This detail is borrowed from Ellis, Diss. Caf., p. IG. In the KoHces Sfatistiqiies siir les Colonies Franraises (ii. p. 46) I find : "About 171G or 1721, fresh seeds of the coffee having been brought secretly from Surinam, in spite of the precautions of the Dutch, the cultivation of this colonial product became naturalized at Cayenne." * The name of this sailor has been spelt in several ways — Declieux, Duclieux, Desclieux. From the information supplied me at the minis- tere de la guerre, I learn that de Clieu was a gentleman, and a connec- tion of the Comte de Maurepas. Ke was born in Normandy, went into tlie navy in 1702, and retired in 1760, after a distinguished career. Ho died in 1775. The official reports have not neglected to mention tlie important fact th'at he introduced the coffee plant into the French colonies. ^ Deleuze, Hist, d^l Museum, i. p. 20. * Not. Stat. Col. Frav^.,i. p. 30. * Ibid., i. p. 209. * Martin, Stat. Col. Brit. Emp. ^ Nouv. Diet. Ilist. Nat., iv. p. 135, ' Not. Stat. Col. Franp., ii. p. 81. 418 ORIGIN OF CULTIVATED PLANTS. on sloping ground and in poor soils where other crops cannot flourish. It corresponds in tropical agriculture to the vine in Europe and tea in China. Further details may be found in the volume published by H. Welter ^ on the economical and commercial history of coffee. The author adds an interesting chapter on the various fair or very bad substitutes used for a com- modity which it is impossible to overrate in its natural condition. Liberian Coffee — Coffea liherica, Hiern.^ Plants of this species have for some years been sent from the Botanical Gardens at Kew into the EnQ^lish colonies. It grows wild in Liberia, Angola, Golungo Alto,^ and probably in several other parts of western tropical Africa. It is of stronger growth than the common coffee, and the berries, which are larger, yield an excellent product. The official reports of Kew Gardens by the learned director. Sir Joseph Hooker, show the progress of this introduction, which is very favourably received, especially in Dominica. Madia — Madia sativa, Molina. The inhabitants of Chili before the discovery of America cultivated this annual species of the Composite family, for the sake of the oil contained in the seed. Since the olive has been extensively planted, the madia is despised by the Chilians, who only complain of the plant as a Aveed which chokes their gardens.^ The Europeans began to cultivate it with indifferent success, owing to its bad smell. The madia is indigenous in Chili and also in Cali- fornia.^ There are other examples of this disjunction of habitation between the two countries.^ ' H. Welter, Essai sur VHistoire du Cafd, 1 vol. 8vo, Paris, 18G8. ' In Hiern, Trans. Linn. Soc, 2nd series, vol. i. p. 171, pi. 2i. This plate is reproduced in the Report of the lloyal Botanical Gardens at Kew for 1876. 3 Oliver, Fl. Trop. Afr., iii. p. 181. * CI. Gay, Fl. Chilena, iv. p. 268. * Asa Gray, in Watson, Bot. of California, i. p. 359, * A. de Candolle, Geogr. Bot. Rais., p. 1017. PLANTS CULTIVATED FOR THEIR SEEDS, 419 Nutmeg — Myristica fragrans, Houttuyn. The nutmeg, a little tree of the order Myristicece, is wild in the Moluccas, principally in the Banda Islands.^ It has long been cultivated there, to judge from the considerable number of its varieties. Europeans have received the nutmeg by the Asiatic trade since the Middle Ages, but the Dutch long possessed the monopoly of its cultivation. When the English owned the Moluccas at the end of the last century, they carried live nutmeof trees to Bencoolen and into Prince Edward's Islands.^ It afterwards spread to Bourbon, Mauritius, Madagascar, and into some of the colonies of tropical America, but with indifferent success from a commercial point of view. Sesame — Sesamiim indicitm, de Candolle ; S. inclicum and S. orientale, Linnseus. Sesame has long been cultivated in the hot regions of the old world for the sake of the oil extracted from the seeds. The order Pedcdinece to which this annual belongs is composed of several genera distributed through the tropical parts of Asia, Africa, and America. Each genus has only a small number of species. Sesamum, in the widest sense of the name,^ has ten, all African except perhaps the cultivated species whose origin we are about to seek. The latter forms alone the true genus Sesamum, which is a section in Bentham and Hooker's work. Botanical analogy points to an African origin, but the area of a considerable number of plants is known to extend from the south of Asia into Africa. Sesame has two races, the one with black, the other with white seed, and several varieties differing in the shape of the leaf. The difference in the colour of the seeds is very ancient, as in the case of the poppy. The seeds of sesame often sow themselves outside plantations, and more or less naturalize the species. This has been observed in regions very remote one from the * Rumphms, Amhoin,, u. p. 17 ; Blume, Runiphia, i. p. ISO, ' Roxburgh, FL Indica, iii. p, 845, 3 BeBtyiam and Hooker, Genera PL, \i. p. 1059. 19 420 ORIGIN OF CULTIVATED PLANTS. other ; for instance, in India, the Sunda Isles, Egypt, and even in the West India Islands, where its cultivation is certainly of modern introduction.^ This is perhaps the reason that no author asserts he has found it in a wild state except Blume,^ a trustworthy observer, Avho men- tions a variety with redder flowers than usual growing in the mountains of Java. This is doubtless an indica- tion of origin, but we need others to establish a proof. I shall seek them in the history of its cultivat'oa. The country where this began should be the ancient habitation of the species, or have had dealings with this ancient habitation. That its cultivation dates in Asia from a very early epoch is clear from the diversity of names. Sesame is called in Sanskrit tila,^ in Malay widjin, in Chinese inoa (Rumphius) or chi-'ina (Bretschneider), in Japanese koha.^ The name sesam is common to Greek, Latin, and Arabic, with trifling variations of letter. Hence it might be inferred that its area was very extended, and that the cultiv^ation of the plant was begun independently in several different countries. But we must not attribute too much importance to such an argument. Chinese works seem to show that sesame was not introduced into China before the Christian era. The first certain mention of it occurs in a book of the fifth or sixth century, entitled Tsi-inin-ycw-chou.^ Before this there is confu- sion between the name of this plant and that of flax, of which the seed also yields an oil, and which is not very ancient in China.^ Theophrastus and Dioscorides say that the Eg^^ptians cultivated a plant called sesame for the oil contained in its seed, and Pliny adds that it came from India."^ He * Pickering, Chronol. Ilistory of Playrfs, p. 223 ; Eumphius, Herh. Amh., V. p. 204 ; Miquel, Flora Indo-Batara, ii. p. 7(>0 ; Schwcinfurth and Ascherson, Aufzahlung, p. 273 ; Grisebach, Fl. Brit. W. Ind. Is., p. 458. 2 Blutne, Bijdragen, p. 778. » Eoxburgh, Fl. Ind., edit. 1832, vol. iii. p. 100; Piddington, Index. * Thunberg, Fl. Jap., p. 254. « Bretschneider, letter of Aug. 23, 1801. * Ihid., On Studij, etc., p. 16. ^ Theophrastus, lib. viii. cap. 1, 5 ; Dioscorides, lib. ii. cap. 121 ; Pliny, Hist., lib. xviii, cap. 10. PLANTS CULTIVATED FOR THEIR SEEDS. 421 also Speaks of a sesame wild in Egypt from which oil was extracted, but this was probably the castor-oil plant.^ It is not proved that tlie ancient Egyptians before the time of Theophrastus cultivated sesame. No drawing or seeds have been found in the monuments. A drawing from the tomb of Rameses III. show the custom of mixing- small seeds with flour in making pastry, and in modern times this is done with sesame seeds, but others are also used, and it is not possible to recognize in the drawing those of the sesame in particular.^ If the Egyptians had known the species at the time of the Exodus, eleven hundred years before Theophrastus, there would probably have been some mention of it in the Hebrew books, because of the various uses of the seed and especially of the oil. Yet commentators have found no trace of it in the Old Testament. The name senisem or smisiin is clearly Semitic, but only of the more recent epoch of the Talmud,^ and of the agricultural treatise of Alawwam,* compiled after the Christian era began. It was perhaps a Semitic people who introduced the plant and the name se7)ise7ii (whence the sesam of the Greeks) into Egypt after the epoch of the great monuments and of the Exodus. They may have received it with the name from Babylonia, where Herodotus says^ that sesame was cultivated. An ancient cultivation in the Euphrates valley agrees with the existence of a Sanskrit name, tila, the tilu of the Brahmans (Rheede, Malabar, i., ix., pp.. 105-107), a word of which there are traces in several modern languages of India, particularly in Ceylon.*' Thus we are carried back to India in accordance with the origin of which Pliny speaks, but it is possible that India itself may have received the species from the Sunda Isles before the arrival of the Aryan conquerors. Rumphius gives * Plinj, Hist., lib. xv. cap. 7. * Wilkinson, Manners and, Customs of Ancient Egyptians, vol. ii. ; Unf?er, Pflanzen des Alien jEgypfens, p. 45. ' Reynier, Econ. Pub. des Arahes et des Jui/s, p. 431 ; Low, Aramaeische Pflavzennamen, p. 376. * E. Meyer, Geschichte der Botanik, iii. p. 75. ' Herodotus, lib. i. cap. 193. ° Tbwaites, Emim., p. 209. 422 ORIGm OF CULTIVATED PLANTS. three names for the sesame in these islands, very different one from the other, and from the Sanskrit word, which supports the theory of a more ancient existence in the archipelago than on the continent. In conclusion, from the fact that the sesame is wild in Java, and from historical and philological arguments, the plant seems to have had its origin in the Sunda Isles. It was introduced into India and the Euphrates valley two or three thousand years ago, and into Egypt at a less remote epoch, from 1000 to 500 B.C. It was transported from the Guinea coast to Brazil by the Portuguese,^ but it is unknown how long it has been cultivated in the rest of Africa. Castor-oil Plant — Ricinus comominis, Linn^us. The most modern works and those in highest repute consider the south of Asia to be the original home of this Eui^horhiacea ; sometimes they indicate certain varieties in Africa or America without distinsfuishinof the wild from the cultivated plant. I have reason to believe that the true origin is to be found in tropical Africa, in accordance with the opinion of Ball.^ The difficulties with which the question is attended arise from the antiquity of cultivation in different countries, from the facility with which the plant sows itself and becomes naturalized on rubbish-heaps and in waste ground, lastly from the diversity of its forms, which have often been described as species. This latter point need not detain us, for Dr. J. MUller's careful monograph ^ proves the existence of sixteen varieties, scarcely heredi- tary, which pass one into the other by many transitions, and constitute, therefore, but one species. The number of vaiieties is the sign of a very ancient cultivation. They differ more or less as to capsules, seeds, inflorescence, etc. Moreover, they are small trees in hot countries, but they do not endure frost, and become annuals north of the Alps and in similar regions. They are in such cases planted in gardens for ornament, * Piso, Brazil., edit. 1658, p. 211. ' Ball, Florae Maroccana Spicilegium, p. 664. ' MUlJer, Argov., ia D.C., Prodromus, vol. xv. part 2, p. 1017. PLANTS CULTIVATED FOR THEIR SEEDS. 423 while in the tropics, and even in Italy, they are grown for the sake of the oil contained in the seed. This oil, which is more or less purgative, is used for lamps in Beniral and elsewhere. In no country has the species been found wild with such certainty as in Abyssinia, Sennaar, and the Kordofan. The expressions ot authors and collectors are distinct on this head. The castor-oil plant is common in rocky places in the valley of Chire, near Goumalo, says Quartin Dillon ; it is wild in those parts of Upper Sennaar which are flooded during the rains, says Hartmann.^ I have a specimen from Kotschy, No. 24:3, gathered on the northern slope of Mount Kohn, in the Kordofan. The indications of travellers in Mozambique and on the coast of Guinea are not so clear, but it is possible that the natural area of the species covers a great part of tropical Africa. As it is a useful species, and one very conspicuous and easily propagated, the negroes must have early diffused it. However, as we draw near the Mediterranean, it is no longer said to be indigenous. In Egypt, Schwein- furth and Ascherson ^ say the species is only cultivated and naturalized. Probably in Algeria, Sardinia, and Morocco, and even in the Canaries, where it is principally found in the sand on the sea-shore, it has been naturalized for centuries. I believe this to be the case Avith speci- mens brought from Djedda, in Arabia, by Schimper, which were gathered near a cistern. Yet Forskal^ gathered the caster-oil plant in the mountains of Arabia Felix, which may signify a wild station. Boissier* indicates it in Beluchistan and the south of Persia, but as " subspontaneous," as in Syria, Anatolia, and Greece. Bheede ^ speaks of the plant as cultivated in Malaliar and growing in the sand, but modern Anglo-Indian authors do not allow that it is wild. Some make no * Richard, Tentamen Fl. Abyss., ii. p. 250; Scliweinfurth, Plantce Kiloficce a Uartmann, etc., p. 13. ^ Schweinfurth and Ascherson, Aufzahhing, p. 262. ' Furskal, Fl. Arabica, p. 71. * Boissier, FL Orient.^ iv. p. 1143 * Rheede, Malabar, ii. p. 57, t. 32. 424j ohigin of cultivated plants. mention of the species. A few speak of tlie facility with Avhich the species becomes naturalized from cultiva- tion. Loureiro had seen it in Cochin-China and in China " cultivated and uncultivated," which perhaps means escaped from cultivation. Lastl}^, for the Sunda Islands, Rumphius ^ is as usual one of the most interesting authorities. The castor-oil plant, he says, grows especially in Java, where it forms immense fields and produces a great quantity of oil. At Amboyna, it is planted here and there, near dwellings and in fields, rather for medicinal purposes. The wild species grows in deserted gardens (in desertis liortis) ; it is doubtless sprung from the cultivated plant (sine duhio degeneratio doniestica). In Japan the castor-oil plant grows among shrubs and on the slopes of Mount Wuntzen, but Franchet and Savatier add,^ " probably introduced." Lastly, Dr. Bretschneider mentions the species in liis work of 1870, p. 20; but what he says here, and in a letter of 1S81, does not argue an ancient cultivation in China. The species is cultivated in tropical America. It becomes easily naturalized in clearings, on rubbish-heaps, etc. ; but no botanist has found it in the conditions of a really indigenous plant. Its introduction must have taken place soon after the discovery of America, for a common name, lamoiiroii, exists in the West India Islands; and Piso gives another in Brazil, oihamhu- giiacu, Jiguero inferno in Portuguese. I have received the largest number of specimens from Bahia ; none are accompanied by the assertion that it is really indigenous. In Egypt and Western Asia the culture of the species dates from so remote an epoch that it lias given rise to mistakes as to its origin. The ancient Egyptians practised it extensively, according to Herodotus, Pliny, I)iodorus, etc. There can be no mistake as to the species, as its seeds have been found in the tombs.^ The Egyptian name was kild. Theophrastus and Dioscorides mention * Rumpliins, JJerh. Anih., vol. iv. p. 03. ' Franchet and Savatier, Env.ni. Japon., i. p. 424. ^ Unger, Pjlanzen des Alien JEjyptens, p. 61. PLANTS CULTIVATED FOE THEIR SEEDS. 425 it, and it is retained in modem Greek/ while the Arabs have a totally different name, kerua, kerroa, chavua.^ Roxburgh and Piddington quote a Sanskrit name, eranda, erimcla, which has left descendants in the modern languages of India. Botanists do not say from what epoch of Sanskrit this name dates ; as the species belongs to hot climates, the Aryans cannot have known it before their arrival in India, that is at a less ancient epoch than the Egyptian monuments. The extreme rapidity of the growth of the castor-oil plant has suggested different names in Asiatic language, and that of Wunderhauini in German. The same circum- stance, and the analogy with the Egyptian name kihi, have caused it to be supposed that the hikajon of the Old Testament,^ the growth, it is said, of a single night, was this plant. I pass a number of common names more or less absurd, as pahna Christi, girasole, in some parts of Italy, etc., but it is worth while to note the origin of the name castor oil, as a proof of the English habit of accept- ing names without examination, and sometimes of dis- torting them. It appears that in the last century this plant was largely cultivated in Jamaica, where it was once called agno casto by the Portuguese and the Spaniards, being confounded with Vitex agnus castus, a totally different plant. From casto the English planters and London traders made castor.^ Walnut — Juglans regia, Linnjeus. Some years ago the walnut tree was known to be wild in Armenia, in the district to the south of the Caucasus and of the Caspian Sea, in the mountains of the north and north-east of India, and in Burmah.^ ^ Theophrastus, Hist,, lib. i. cap. 19 ; Dioscorides, lib. ir. cap. 171 ; Fraas, Syn. Ft. Class., p. 92. 2 Nemnich, Polyglott. Lexicon ; Forskal, Fl. JEgypt., p. 75. 3 Jonah iv. 6. Pickering, Chron. Hist. Plants, p. 225, writes Jcyhwi/n. * Fliickiger and Hanbui'y, Fharmacographiuy p. 511. * A. de Caudolle, Prodr., xvi. part 2, p. 13G; Tcbihatclieff, Asic Mineure, i. p. 172 ; Ledebour, Fl. Ross., i. p. 507 ; Koxburgh, Fl. Lid., iii. p. 630; Boissier, Fl. Orient., iv. p. 1160; Jirandis, Forest^Flora of N.W. India, p. 498; Kurz, Forest Flora of Brit. Burmah, p. 390. 426 ORIGIN OF CULTIVATED PLANTS. C. Kocli ^ denied that it was indigenous in Armenia and to the south of the Caucasus, but this has been proved by several travellers. It has since been discovered wild in Japan,^ which renders it probable that the species exists also in the north of China, as Loureiro and Bunnre said,^ but without particularizing its wild character. Heldreich* has recently placed it beyond a doubt that the walnut is abundant in a wild state in the mountains of Greece, whicli agrees with passages in Theoplirastus ^ which had been overlooked. Lastl}^, Heuffel saw it, also wild, in the mountains of Banat.'' Its modern natural area extends, then, from eastern temperate Europe to Japan. It once existed in Europe further to the west, for leaves of the walnut have been found in the quater- nary tufa in Provence.'' Many species of Juglans existed in our hemisphere in the tertiary and quaternary epochs ; there are now ten, at most, distributed throughout North America and temperate Asia. The use of the walnut and the planting of the tree may have begun in several of the countries where the species was found, and cultivation extended gradually and slightly its artificial area. The walnut is not one of those trees which sows itself and is easily naturalized. The nature of its fruit is perhaps against this ; and, moreover, it needs a climate where the frosts are not severe and the heat moderate. It scarcely passes the northern limit of the vine, and does not extend nearly so far south. The Greeks, accustomed to olive oil, neglected the walnut until they received from Persia a better variety, called Jcaruon hasilikon^ or Pcrsikon? The Romans * C. Koch, Dendrologip, i. p. 584. ^ Fraiichet and Savatier, Enum. Plant. Jap., i. 453. ' Loureiro, Fl. Cochin., p. 702; Bunge, Emim., p. 62. * Heldreich, Verhandl. Bot. Vereins Brandenb., 1879, p. 147. * Theophrastus, Hist. Plant., lib. iii. cap. 3, 6. These passac:os, and others of ancient writers, are quoted and interpreted by Heldreich better than bv Hehn and other scholars. « HeufTel, Abhandl. Zool. Bot. Ges. in Wien, 1853, p. 194. ' De S^porta, 33 rd Sess. du Cong res Scient. de France. * Dio-corides, lib. i. cap. 176. ® ri'ny, Hist. Plant., lib. xv. cap. 22. PLANTS CULTIVATED FOR THEIR SEEDS. 427 cultivated the walnut from the time of their kings ; they considered it of Persian origin.^ I'hey had an old custom of throwino: nuts in the celebration of weddini^s. Archaeology confirms these details. The only nuts which have hitherto been found under the lake-dwellini^s of Switzerland, Savoy, or Italy are confined to a single locality near Parma, called Fontinellato, in a stratum of the iron age.^ Now, this metal, very rare at the time of the Trojan war, cannot have come into general use among the agricultural population of Italy until the fifth or sixth century before Christ, an epoch at which even bronze was perhaps still unknown to the north of the Alps. In the station at Lagozza, walnuts have been found in a much higher stratum, and not ancient.^ Evidently the walnuts of Italy, Switzerland, and France are not descended from the fossil plants of the quater- nary tufa of which I spoke just now. It is impossible to say at what period the walnut was first planted in India. It must have been early, for there is a Sanskrit name, akschoda, akhoda, or akhota. Chinese authors say that the walnut was introduced among them from Thibet, under the Han dynasty, by Chang-kien, about the year 140-150 B.C.* This was per- haps a perfected variety. Moreover, it seems probable, from the actual records of botanists, that the wild walnut is rare in the north of China, and is perhaps wanting in the east> The date of its cultivation in Japan is un- known. The walnut tree and walnuts had an infinite number of names among ancient peoples, wdiich have exercised the science and imagination of philologists,^ but the origin of the species is so clear that we need not stay to consider them. Areca — Areca Catechu, Linmeus. ' Pliny, Hht. Plant., lib. xv. cap. 22. * Heer, Pflmzen der Pfahlhauten, p. 31. ' Sordelli, Sulle piante della torhiera, etc., p, 39. * Bretschneider, Study and Value, etc., p. IG; and letter of Aug. 23, 1881. ^ Ad. Pictet, Origines Ittdo-Europ., edit. 2, vol. i. p. 289 ; Helm, Cul. turpjlanzen und Hausthiere, edit. 3, p. 311. 42 S OrtlGIN OF CULTIVATED PLANTS. The areca palm is mucli cultivated in the countries where it is a custom to chew betel, that is to say through- out Southern Asia. The nut, or rather the almond which forms the piincipal part of the seed contained in the fruit, is valued for its aromatic taste ; chopped, mixed with lime, and enveloped in a leaf of the pepper-betel, it forms an agreeable stimulant, which produces a flow of saliva and blackens the teeth to the satisfaction of the natives. The author of the principal work on the order Palm- acese, de Martins,^ says of the origin of this species, " Its country is uncertain (non constat) ; probably the Sunda Isles." We may lind it possible to affirm some- thing positive by referring to more modern authors. On the continent of India, in Ceylon and Cochin-China, the species is always indicated as cultivated.^ So in the Sunda Isles, the Moluccas, etc., to the south of Asia. Blume,^ in his work entitled llnmplda, says that the " habitat " of the species is the ^lalay Peninsula, Siam, and the neighbouring islands. Yet he does not appear to have seen the indigenous plants of which he speaks. Dr. Bretschneider * believes that the species is a native of the Malay Archipelago, principally of Sumatra, for he says those islands and the Philippines are the only places where it is found wild. The hrst of these facts is not confirmed by Miquel, nor the second by Blanco,^ who lived in the Philippines. Blume's opinion aj^pears the most probable, but we must still say with Martins, "" The country is not proved." The existence of a num- ber of Malay names, innang, jamhe, etc., and of a San- skrit name, gouvaka, as well as very numerous varieties, show the antiquity of cultivation. The Chinese received it. Ill B.C., from the south, with the Malay nsLvne, 2^in-lang, ^ Martius, Hist. Nat. Palmarum, in folio, Tol. iii. p. 170 (published without date, but before 1851). ' Roxburc^h, PL InO., iii. p. fil6 ; Bx'andis, Forest Fl. of India, p. 551 ; Kurz, Forest Fl. of Brit. Burmah, p. 537 ; Thwaites, Enum. Zeylan., p. 327 ; Loureiro, FL Cochin-Ch., p. 695. ' Blume, Iliimphia, ii. p. 67; Miquel, FL Indo-Batava., iii. p. 9; vppl. de Sumatra, p. 253. * Bretschneider, Studij and Valve, etc., p. 2S. * Blanco, FL di Filipinas, edit. 2. PLANTS CULTIVATED FOR THEIR SEEDS. 429 The Telinga name, areh, is the origin of the botanical name Areca. Elaeis — Elceis guineensis, Jacquin. Travellers who visited the coast of Guinea in the first half of the sixteenth century ^ already noticed this pahn, from which the negroes extracted oil by pressing the fleshy part of the fruit. The tree is indigenous on all that coast.^ It is also planted, and the exportation of palm-oil is the object of an extensive trade. As it is also found wild in Brazil and perhaps in Guiana,^ a doubt arose as to the true origin. It seems the more likely to be American that the only other species which with this one constitutes the genus Elceis belong^s to New Granada.^ Kobert Brown, however, and the authors who have studied the family of palms, are unanimous in their belief that Elceis guineensis was introduced into America by the negroes and slave-traders in the traffic between the Guinea coast and the coast of America. Many facts confirm this opinion. The first botanists who visited Brazil, Piso and Marcgraf and others, do not mention the Elseis. It is only found on the littoral, from Rio di Janeiro to the mouth of the Amazon, never in the interior. It is often cultivated, or has the appearance of a species escaped from the plantations. Sloane,^ who explored Jamaica in the seventeenth century, relates that this tree was introduced in his time into a plantation which he names, from the coast of Guinea. It has since become naturalized in some of the West India Islands.'' Cocoa-nut Palm — Cocos mucifera, Linnaeus. The cocoa-nut palm is perhaps, of all tropical trees, the one which yields the greatest variety of products. Its * Da Mosto, in Ramusio, i. p. lOl, quoted by R. Brown. ' Brown, Bof. of Congo, p. 55. ' Martius, Hist. Nat. Palmaram, ii. p. 02; Dnide, in Fl. Bra. •<{!., ^nsc. 85, p. 457. I find no author who asserts that this palui is wild in (Juiana, as Martius afiirms it to be in Brazil. * Elceis melanocarpa, Gaertncr. The fruit also contains oil, but it does not appear that the species is cultivated, as the number of oleaginous plants is considerable in all countries. * Sloane, Nat. Hist, of Jamaica, ii. p. 113. f Grlsebach, Flora of Brit. W. Inch Is., p. 522, 430 OEIGIN OF CULTIVATED PLANTS. wood and fibres are utilized in various ways. The sap extracted from the inner part of the inflorescence yields a much-prized alcoholic drink. The shell of the nut forms a vessel, the milk of the half-ripe fruit is a pleasant drink, and the nut itself contains a great deal of oil. It is not surprising that so valuable a tree has been a good deal planted and transported. Besides, its dispersion is aided by natural causes. The woody shell and fibrous envelope of the nut enable it to float in salt water without injury to the germ. Hence the possibility of its transportation to great distances by currents and its naturalization on coasts where the temperature is favourable. Unfortu- nately, this tree requiies a warm, damp climate, such as exists only in the tropics, or in exceptional localities just without them. Nor does it thrive at a distance from the sea. The cocoa-nut abounds on the littoral of the warm reo-ions of Asia, of the islands to the south of this con- tinent, and in analogous regions of Africa and America ; but it may be asserted that it dates in Bi-azil, the West Indies, and the west coast of Africa from an introduction which took place about three centuries ago. Piso and Marcgraf ^ seem to admit that the species is foreign to Brazil without saying so positively. De Martius,'-^ who has published a very important work on the Palmace?e, and has travelled through the provinces of Bahia, Per- nambuco, and others, Avhere the cocoa-nut abounds, does not say that it is wild. It was introduced into Guiana by missionaries.^ Sloane ^ says it is an exotic in the West Indies. An old author of the sixteenth century, Martyr, whom he quotes, speaks of its introduction. This probably took place a few years after the discovery of America, for Josej^h Acosta^ saw the cocoa-nut palm at Porto Rico in the sixteenth centur}^ De Martius says that the Portuguese introduced it on the coast of Guinea. Many travellers do not even mention it in this » Piso, Brai^iL, p. fi5 ; Marcgraf, p. 138. 2 Martius, Hist. Nat. Palmarnw^ 3 vols, in folio; see vol. ii. p. 125. ' Aublet, Guyane, suppl., p. 102. * Sloane, Jamaica, ii. p. 9. ^ J. Acosta, Rist. Nat. des hides, French trans., 1598, p. 178. PLANTS CULTIVATED FOR THEIR SEEDS. 431 region, where it is apparently of no great importance. More common in Madagascar and on the east coast, it is not, however, named in several works on the plants of Zanzibar, the Seychelles, Mauritius, etc., perhaps because it is considered as cultivated in these parts. Evidently the species is not of African origin, nor of the eastern part of tropical America. Eliminating these countries, there remain western tropical America, the islands of the Pacific, the Indian Archipelago, and the south of Asia, where the tree abounds with every appear- ance of beino; more or less wild and lonor established. The navigators Dampier and Vancouver^ found it at the beginning of the seventeenth century, forming woods in the islands near Panama, not on the mainland, and in the isle of Cocos, situated at three hundred miles from the continent in the Pacific. At that time these islands were uninhabited. Later the cocoa-nut palm was found on the western coast from Mexico to Peru, but usually authors do not say that it was wild, excepting Seemann,^ however, who saw this palm both wild and cultivated on the Isthmus of Panama. Accordinor to Hernandez,^ in the sixteenth century the Mexicans called it coyolli, a word which does not seem to be native. Oviedo,* writing in 1526, in the first years of the con- quest of Mexico, says that the cocoa-nut palm was abun- dant on the coast of the Pacific in the province of the Cacique Chiman, and he clearly describes the species. This does not prove the tree to be wild. In southern Asia, especially in the islands, the cocoa-nut is both wild and cultivated. The smaller the islands, and the lower and the more subject to the influence of the sea air, the more the cocoa-nut predominates and attracts the atten- tion of travellers. Some take their name from the tree, among others two islands close to the Andamans and one near Sumatra. ^ Vafer, Voyage de Dampier, edit. 1705, p. 186 ; Vancouver, French edit., p. 325, quoted by de Martius, Hist. Nat. Palmarum, i. p. 1S8. * Seemaiin, Bot. of Herald.^ p. 20-i. ' Hernandez, Thesaurus Mexic, p. 71. He attiibutes the same name, p. 75, to the cocoa-nut palm of the Philippine Islands. * Oviedo, Ramusio's traus., iii. p. 53. 432 ORIGIN OF CULTIVATED PLANTS. The cocoa-nut occurring with every appearance of an ancient wild condition at once in Asia and western America, the question of origin is obscure. Excellent authors have solved it differently. De Martins believes it to have been transported by currents from the islands situated to the west of Central America, into those of the Asiatic Archipelago. I formerly inclined to the same hypothesis,^ since admitted without question by Grise- bach ; ^ but the botanists of the seventeenth century often regarded the species as Asiatic, and Seemann,^ after a careful examination, says he cannot come to a decision. I will give the reasons for and against each hypothesis. In favour of an American origin, it may be said — 1. The eleven other species of the genus Cocos are American, and all those which de Martius knew well are Brazilian.* Drude,^ who has studied the Palmaceae, has written a paper to show that each genus of this family is proper to the ancient or to the new world, excepting the genus El?eis, and even here he suspects a transport of the E. guineensis from America into Africa, which is not at all probable. (See above, p. 429.) The force of this argument is somewhat diminished by the circumstance that Cocos nmcifera is a tree which groAvs on the littoral and in damp places, while the other species live under different conditions, frequently far from the sea and from rivers. Maritime plants, and those which grow in marshes or damp places, have commonly a more vast habitation than others of the same irenus. 2. The trade winds of the Pacific, to the south and yet more to the north of the equator, drive floating bodies from America to Asia, a direction contrary to that of the general currents.^ It is known, moreover, from the un- * A. de Canclolle, GSogr. Bot. Baisovnee, p. 976. * Grisebach, Vegetation der Erde, pp. 11, 323. ' Seemann, Flora Vitiensis, p. 275. * The cocoa-nut called Maldive belong's to the genus Lodoicen. Coco mamillaris, Blanco, of the Philippines is a variety of the culti. vated Cocos micifera. ' Drade, inBot. Zeitung, 187C, p. 801 ; and Flora Brasilicnsis,ia.sc. 85, p. 405. ^ Stieler, Hand Atlas, edit. 1867, map 3. PLANTS CULTIVATED FOR THEIR SEEDS. 433 expected arrival of bottles containing papers on different coasts, tliat cliance has much to do with these transports. The arguments in favour of an Asiatic, or contrary to an American origin, are the following: — 1. A current between the third and fifth parallels, north latitude, flows from the islands of the Indian Archipelago to Panama.^ To the north and south of this are currents which take the opposite direction, but they start from regions too cold for the cocoa-nut, and do not touch Central America, where it is supposed to have been long indigenous. 2. The inhabitants of the islands of Asia were far bolder navi'^ators than the American Indians. It is verv possible that canoes from the Asiatic Islands, containing a provision of cocoa-nuts, were thrown by tempests or false manoeuvres on to the islands or the west coast of America. The converse is highly improbable. 3. The area for three centuries has been much vaster in Asia than in America, and the diflerence was yet more considerable before that epoch, for we know that the cocoa-nut has not long existed in the east of tropical America. 4. The inhabitants of the islands of Asia possess an immense number of varieties of this tree, which points to a very ancient cultivation. Blume, in his Rumphia, enumerates eighteen varieties in Java and the adjacent islands, and thirty-nine in the Philippines. Nothing similar has been observed in America. 5. The uses of the cocoa-nut are more varied and more habitual in Asia. The natives of America hardly utilize it except for the contents of the nut, from which they do not extract the oil. 6. The common names, very numerous and original in Asia, as we shall presently see, are rare, and often of European origin in America. 7. It is not probable that the ancient Mexicans and inhabitants of Central Ameiica would have neglected to spread the cocoa-nut in several directions, had it existed among them from a very remote epoch. The trifling ' Stieler, ibid., map 9. 43 Jf ORIGIN OF CULTIVATED PLANTS. breadth of the Isthmus of Panama would have facilitated the transport from one coast to the other, and the species would soon have been established in the West Indies, at Guiana, etc., as it has become naturalized in Jamaica, Antigua,^ and elsewhere, since the discovery of America. 8. If the cocoa-nut in America dated from a cfeolosical epoch more ancient than the pleiocene or even eocene deposits in Europe, it would probably have been found on both coasts, and the islands to the east and west equally. 9. We cannot tind any ancient date of the existence of the cocoa-nut in America, but its presence in Asia three or four thousand years, ago is proved by several Sanskrit names. Piddington in his index only quotes one, narikela. It is the most certain, since it recurs in modern Indian lan-inal condition to such a degree that they could no longer be recognized among wild plants. It appears, on the contrary, that the forms anterior to cultivation have commonly remained side by side with those wdiich cultivators employed and propa- gated from century to century. This may be explained in two ways : 1. Tlie period of four thousand years is short compared to the duration of most of the specific forms in plianerogamous plants. 2. The cultivated species receive, outside of cultivated ground, continual j-einforcen^ents from the seeds wliieh man, birds, and different natural agents disperse and transport in a thousand ways. Natuializations produced in this manner often confound the w41d j^lants with the cultivated ones, and the more easily that they fertilize each other since they belong to the same species. This fact is clearly demonstrated in the case of a plant of tlie oM world cultivated in America, in gardens, and which, later, becomes naturalized on a large scale in the open country or the woods, like the cardoon at Buenos Ayre>^, and the oranofes in several American countries. Cultivation widens areas, and supplements the deficits which the natural reproduction of the species may present. There are, however, a few exceptions, which are worth men- tioning in a separate article. Article. IV. — Cultivated Plants which are Extinct, or becoming Extinct in a Wild State. These species to which I allude present three remark- able characters : — 1. They have not been found w^ild, or only once or twice, and often doubtfully, although the regions wdience they come have been visited by several botanists. 2. They have not the faculty of sowing themselves, and propagating indefinitely outside cultivated ground. GENERAL OBSEHVATIONS AND CONCLUSIONS. 450 In othfir terms, in sncli cases they do not pass out of tlie condition of adventitious plants. 3. It cannot be supposed that they are derived within historic times from certain allied species. These three characters are found united in the follow- ing species: — Bean (Faha vuhjarw), chick-pea {Cicer arietinum), ervilla {Ervum Ervilia), lentil (Erviim lens), tobacco {JSHcotiaria tahacum), wheat (Triticum vul- gwre), maize {Zea tnnys). The sweet potato (Convol- vulus batatas) should be added if the kindred species were better known to be distinct, and the carthamine {Garthamus tinctorius) if the interior of Arabia had been explored, and we had not found a mention of the plant in an Arabian author. All these species, and probably others of little-known countries or genera, appear to be extinct or on their w^ay to become so. Supposing they ceased to be cultivated, they would disappear, whereas tlie majority of culti- vated plants have become somewhere natuialized, and would persist in a wild state. The seven species mentioned just now, excepting tobacco, have seeds full of fee u la, which are the food of birds, rodents, and different insects, and have not the power of passing entire through their alimentary canal. This is probal)]y the sole or pi'incipal cause of their inferiority in the struggle for existence. Thus my researches into cultivated plants show that certain species are extinct or becoming extinct since the historical epoch, and that not in small islands but on vast continents without any great modifications of climate. This is an important result for the history of all organic beings in all epochs. Article V. — Concluding Remarks. 1. Cultivated plants do not belong to any particular category, for they belong to fifty-one diff'erent families. They are, however, all phanerogamous except the mush- room (Agaricus cavipestris). 460 oraaiN of cultivated plants. 2. The characters which have most varied in cultiva- tion are, beginning with the most variable : a. The size, form, and colour of the fleshy parts, whatever organ they belong to (root, bulb, tubercle, fruit, or seed), and the abundance of fecula, sugar, and other substances wliich are contained in these parts ; h. The number of seeds, which is often in inverse ratio to the development of the fleshy parts of the plant ; c. The form, size, or pubes- cence of the floral organs which persist round the fruits or seeds ; d. The rapidity of the phenomena of vegeta- tion— whence often results the quality of ligneous or herbaceous plants, and of perennial, biennial, or annual. The stems, leaves, and flowers vary little in plants cultivated for those oro-ans. The last formations of each yearly or biennial growth vary most ; in other terms, the results of vegetation vary more than the orG:ans which cause ve , kidnev, 338 Beetroot, 58 Beuincasa, 2(38 Beta vulgaris, 58 Bird. cherry. 205 Bird's foot, 113 liittcr orange, 1*^3 Bixa Orellana, 401 Black currant, 278 Biassica campestris, 3G Napus, 30 oleracea, 30, 83 Rapa, 30 Bread-fruit, 298 Broad bean, 316 Bromelia Ananas, 311 Buckwheat, common, 3 J8 , notch-seeded, 351 , Tartarj, 353 Bullace, 214 Bullock's heart, 174 o Cabbage, 83 Cacao, 313 Caimito, 285 Calabash, 21-5 Cannabis sativa, 118 Capsicum annuuin, 289 frutescens, 2'JO Cardoon, 92 Carica Papaya, 273 Carob, 334 Carthaniine, 164 Caryophvllus aromaticus, 161 Cashew, 'l 98 Cassis, 278 Castanea vulgaris, 353 Cast or- oil plant, 422 Catha edulis, 134 Celery, 89 Cerasus vulgaris, 207 Ceratonia Siliqua, 331 Chayote, 273 Chenopodium Quinoa, 351 CheiTy, bird, 205 , sour, 207 Chervil, 90 Chestnut, 353 Chickling vetch, 110 Chick-pea, 323 Chicorium Endivia, 97 Intvbus, 90 Chicory, 96 China grass, 146 Chinese pear, 233 Cliirimoj^a, 174 Chives, 72 Chocho, 273 Chrysophyllum Ca'i'mito, 285 Cinnamon, 146 Cinnamonum zeylanicum, 116 Citron, 178 Citrullus vulgaris, 202 Citrus Aurantium, 18S decumana, 177 medica, 178 nobilis, 188 Clove, 161 Clover, crimson, 106 , Egyptian, 1(>7 , purple, 105 Coca, 135 Cochlearia Armoraciii, 33 Cocoa-nut palm, 429 Cocos nucifera, 429 Coffee, 415 Coffea arabica, 418 liberica, 418 Colocasia, 73 Convolvolus Batatas, 53 mammosa, 57 Corchorus capsularis, 130 olitorius, 130 Com salad, 91 Corn spurry, 114 Cotton, Barbados, 408 , herbaceous, 452 , tree, 408 Cress, garden, 166 Crocus sativum, 86 Cucumber, 264 Cucumis Anguria, 2G7 Melo, 258 sativas, 26 1 Cucurbita citrullus, 202 ficifolia, 257 Lagenaria, 245 maxima, 249 INDEX. 4G5 Cucnrbita Melopopo, pepo, 253 moscViata, 257 Currant, black, 278 , red, 277 Custard apple, 168 Cydouia vulgaris, 236 Cynara Carduncnlus, 93 Cytisus Cajan, 33:i ScolyBius, 92 D Date-palm, 301 Dioscorea, 76 Uulichos Lablab, 346 Labia, 347 Soja, 330 Dyer's iudigo, 136 E Egyptian clover, 107 lupin, 327 wheat, 259 Eloois guineensis, 429 Eleusine Coracaua, 384 Endive, 97 Ervilla, 107 Ervum Eivilia, 107 lens, 321 Erythroxylon Coca, 135 Eugrenia Jambos, 240 malaccensis, 241 F Faba vtilgaris, 316 Fagopyrum emarginatum, 351 esculontum, 348 tataricum, 350 Fenugreek, 112 Ficus Carica, 295 Field-pea, 327 Fig, 295 Fig-leaved pumpkin, 257 Fig, Indian, 274 Flat-podded pea, 109 Flax, 119 Fragaria chiloensis, 205 vesca, 203 virgiiiiana, 205 French honeysuckle, 104 G Garcinia JMangostana, 118 Garden cress, 86 pea, 328 Garlic, 63 Glycine soya, 330 subterranea, 317 Gombo, 189 Gooseberrv, 276 Gossypium arborenm, 108 barbadeusc, 408 herbaceum, 402 Gourd, 245, 249 , snake, 273 , towel, 269 Grass, China, 3-16 Grass, guinea, 115 Green gram, 31.() Guava, 241 Uaricot bean, 338 Ilcdysarium coronarinni, 104 Helianthus tubcrosus, 42 Hemp, 148 Henna, 138 Hibiscus esculentus, 189 Holcus saccharatus, 382 Sorghum, 380 Hop, 162 Hordeum distich on, 367 hexastichon, 369 vulgare, 368 Horse-radish, 33 Humulus Lupulus, 162 Ilex pnraguarionsis, 135 Indian fig, 274 Indigo, American, 137 , dyer's, 136 , silver, 137 Tndigofera argentea, 137 cerulca, 137 tinctoria, 136 Ipomea mammosa, 57 466 INDEX. Jact-fruit, 299 Jambosa Malaccensis, 241 vulgaris, 240 Jatropha uianihot, 59 Jerusalem artichoke, 42 Juylans regia, 425 Jujube, commou, 194- , Indian, 197 , Lotus, 196 Jute, 130 Kidney bean, 338 , tnotli, 344 , three-lobed, 315 Kierj, 352 Khat, 134 Konjak, 76 Lablab, 347 Lagenaria vulgaris, 245 Lamb's lettuce, 91 Lathyrus CiccrJi, 109 "Ocbrus, 110 — sativus, 111 7 Lattuca scariola, 95 Lawsonia alba, 138 Leek, 101 Lemon, 178 Lens esculenta, 221 Lentil, 321 Lcpidnm sativum, 86 Lettuce, 95 • , lamb's, 91 Linum usitatissimum, 119 Litchi, 314 Longan, 315 Lotos jujube, 196 Labia, 347 Lucern, 102 Lucuma Caimito, 285 mammosa, 286 Lupin, 325 Lupinus albus, 325 termis, 327 Lycopersicum esculcntum, 290 M 413 iMadder, 41 Madia sativa, Maize, 387 Malay apple, 241 Mammee, 199 americana, 189 Sapota, 286 Mandarin, 188 Mandubi, 411 ]\rangifera indica, 200 Mango, 200 Mangosteen, 188 Manioc, 59 Manihot utilissima, 59 iLaranta arundinacea, 81 Marmalade plum, 286 Mate, 135 Medicago sativa, 103 Melon, 258 , pumpkin, 256 , water, 2152 , white gourd, 2:18 Millet, common, af4» "^7 6 , Italian, 2^8 "S 7^ Jfrmoi-dica cylindrica, 2C9 Monkey-nut, 411 Moras alba, 149 nigra, 152 Mulberrv, 149 Mung, 346 Musk pumpkin, 356 Myristica fragrans, 419 N Nephelium lappaceum, 315 litchi, 314 longana, 315 New Zealand spinacli, 89 Nicotiana tabacum, 139 Nutmeg, 419 Oats, 37$ Oehro, 189 Ochrus, 111 Oil.palm, 429 Olca europea, 279 INDEX. 4G7 Olive, 279 Onion, 66 , spring or Welsh, G8 Onobrychis sativa, lOl Opuntia ficns ludica, 274) Orange, 181 , bitter, 185 , sweet, 183 Ornithopus sativus, 113 Oryza sativa, 385 Palm, cocoa-nut, 429 oil, 429 Panicum italicura, 378 maximum, 115 miliaceum, 376 Papava somniferam, 397 Papaw, 293 Papaya vulgaris, 293 Parsley, 90 Pea, 327 , field, 327 , garden, 328 nut, 411 , pisreoD, 382 Peach, 221 Pear, 229 , avocado, 273 , Chinese, 233 , prickly, 27-i , sand, 233 , snowy, 232 Pepper, red, 288 Persea gratissima, 293 Persica vulgaris, 221 Petroselinum sativum, 90 Phaseolus aconitifolius, 315 lunatos, 344 Mungo, 346 vulgaris, 338 Phconix dactylifera, 301 Pigeon-pea, 332 Pine-apple, 311 Pistachio nut, 316 Pistacia vera, 316 Pisum arvense, 327 Ochrus, 111 sativum, 328 Plum, 211 21 Polygonum emarginafum, 351 fagoj^yrum, 3 18 tataricum, 353 Pomegranate, 327 Poppy, 397 Portulaca oleracea, 87 Potato, 45 , sweet, 83 Prickly pear, 274 Prunus Amygdalnp, 213 Armeniaca, Zlb • avium, 205 Cerasus, 207 domestica, 212 insititia, 214 Persica, 221 Psidium guayava, 241 Pumpkin, fig-leaved, 257 , musk or melon, 256 Pnnica Granatum, 237 Pui'slane, 87 Pyrus communis, 229 mains, 233 nivalis, 233 sinensis, 233 Q Quince, 236 Quinoa, 351 Radish, 29 , horse, 33 Rambutan, 315 Ilaphanus sativus, 29 llhus Coriaria, 133 Ribes Grossularis, 276 nigrum, 278 rubrum, 277 Uva-crispa, 276 Rice, 385 Ricinus communis, 423 Rocambole, 72 Rose-apple, 240 Rubia tinctorum, 41 Rve, 370 468 INDEX. S Saccharattun officinale, 154 Saffron, 106 Sainfoin, 104 , Spanish, 104 Salsify, 44 Sapodilla, 286 Sapota achras, 286 Scandix cerefolium, 90 Scorzonera hispanica, 44 Secale cereale, 370 Sechinm edule, 272 Sesame, 419 Sesamuin indicam, 419 Setaria Italica, 380 Shaddock, 177 Shallot, 68 Siam Sisaram, 39 Skirret, 39 Smyminm Olus-atrum, 91 Snake gourd, 272 Solanum Commersonii, 46 immite, 49 niaglia, 49 tuberosum, 45 veiTucosus, 49 Sorghum sa<;charatus, 383 vulgaris, 360 Sour sop, 173 Soy, 330 Spanish sainfoin, 104 Spelt, 362 Spergula arvensis, 114 Spinach, 98 , New Zealand, 87 Spinacia oleracea, 98 Spondias dulcis, 202 Spurry, com, 114 Straw'berrv, 203 , Chili"', 205 , Virginian, 205 Sugar apple, 168 cane, 154 Sumach, 133 Sweet potato, 83 sop, 168 Tahiti apple, 202 Tare, 108 Tea, 117 Tetragonia expansa, 89 Thea sinensis, 117 Theobroma Cacao, 313 Tobacco, 139 Towel gourd, 269 Trigonella Foennm-grrocum, 112 Trifolium Alexandrinum, 107 incamatum, 146 pratense, 105 Triticum a3stivum, 35 i compositum, 359 dicoccam, 365 durum, 360 hybernnm, 354 moncoccum, 365 polonicum, 361 spelta, 262 vulgare, 354 Turnip, 36 Valerianella nlitoria, 89 Vetch, chickling, 110 , common, 108 Vicia ervilla, 107 sativa, 108 Vine, 191 Vitis vinifera, 191 Voandzeia subterranca, 317 Walnut, 245 Wheats, 35 i Yams, 76 W z Zea Mays, 387 Zizyphus jujube, 197 Lotus, 196 vulgaris, 191 %C. FALLACIES: ' A View of Logic from tlie Practical Side. By ALFBED SIDGWICK, B. A., Oxon. 12mo. Cloth, $1.75. This book is intended mainly for the general reader. That is to say, it requires no previous technical training, and is written as much as pos- sible from the unprofessional point of view. " A book intended for popular use, and one now very much needed in these days of half education, when so many persons are the prey of illusions. It treats mainly of the methods of proof, shows what evidence is, and the different sorts needed to produce belief, and what are the fallacies most suited to deceive." — Hartford Cmirant. "Like all the others in this series, this volume is intended for the general reader, but the trained logician will find it useful and suggestive." — New York Herald. "An important treatise on a topic that deserves the attention of all thinking people. ITie author writes mainly for the general reader; no previous technical training, but only a fair degree of intelligence and application, is needful to follow the train of his thought." — Cultivator and Country Gentleman. " Even among educated men logic is apt to be regarded as a dry study, and to be neglected in favor of rhetoric ; it is easier to deal with tropes, metaphors, and words, than with ideas and arguments— to talk than to reason. Logic is a study; it re- quires time and attention, but it can be made interesting, even to general readers, as this work by Mr. Sidgwick upon that part of it included in the name of 'Fallacies' shows. Logic is a science, and in this volume we are taught the practical side of it. The author discusses the meaning and aims, the subject-matter and process of proof, unreal assertions, the burden of proof, non-sequiturs, guess-work, argument by exam- ple and sign, the reductio ad absurdum, and other branches of his subject ably and fully, and has given us a work of real value. It is furnished with a valuable appendix, and a good index, and we should be glad to see it in the hands of thinking men who wish to understand how to reason out the truth, or to detect the fallacy of an argu- ment"— The Churchman. "Its perusal would save many a man from being misled." — Louisville {Ky.) Chris- tian Observer. "The author has bestoAved much labor upon the production, and the originality of his ideas is refreshing. lie holds that to combat fallacy is the raison d'^etre of logic, hence, instead of touching logic directly, he treats, in a systematic manner, of those fallacies which logic combats." — Ilarrisburg {Pa.) Telegraph. New York: D. APPLETON & CO., 1, 3, & 5 Bond Street THE ORGANS OF SPEECH, And their Application in the Formation of Articulate Sounds. By GEORG HERMANN VON MEYER, Professor of Anatomy at the University of Zurich. Witli numerous Illustrations. - - - 12mo, clotli, §1.75. " At once philosophical and practical, suitable as a text-book in a medical col- lege or for reading at home. Persons engaged in philological studies, and all professional musicians, will find it full of extremely useful facts and suggestions." —New York Journal of Commerce. " This volume comprises the author''s researches into the anatomy of the vo- cal organs, with special reference to the point of view and needs of the philolo- gist and the trainer of the voice. It seeks to explain the origin of artif'-Jate sounds, and to outline a system in which all elements of all languages may be co-ordinated in their proper place. The work has obviously a special value for students in the science of the transmutations of lantruage, for etymologists, elo- cutionists, and musicians."— A'isto York Home Journal. " The author perceives in the sounds made by animals meanings analogous to words, and in support of this is the fact that in the legends of all nations an im- portant part is played by wise men who understand the language of the brute creation. With patient thoroughness Professor von Meyer describes minutely the vocal apparatus, and the sounds produced by the complex combinations of its simple parts."— Ciw^/i/iaii Commercial Gazette. " It is surprising to note what different sounds, and shades of tone and mean- ing, can be produced by volition in the use of the human organs of speech." — Hartford Daily Times. "The book presents a happy combination of the Teutonic thoroughness of treatment with the method and lucidity of statement which especially distinguit-h French scientists. The author's expositions are remarkable for their clearness and avoidance of technicalities ; while their meaning is rcndend more apparent by the use of numerous diagrams."— jErice. New York: D. APPLETON & CO., 1, 3, & 5 Bond Street. Ants, Bees, and Wasps. A Record of Observations on the Habits of the Social Hy7nenopiera. By Sir JOHN LUBBOCK, Bart., M. P., F. K. S., etc., Author of " Origin of Civilization, and the Primitive Condition of Man/'' etc., etc. "With Coloeed Plates. 12mo. Cloth, $2.00. "This volume contains the record of various experiments made with ants, bees, and wasps durinjJT the last ten years, with a view to test their mental condition and powers of sense. The principal point in which ^ir John's mode of experiment differs from those of Iluber, Forel, McCook, and others, is that he has carefully watched and marked particular insects, and has had their nests under observation for lonsr periods — one of hi.s ants' nests havino^ been under constant inspection ever since 1ST4. His observations are made principally upon ants because they show more power ard flexi- bility of mind; and the value of his studies is that they belong to the department of original research." " We have no hesitition in saying that the author has presented us with the most valuable series of observations on a special subject that has ever been produced, charm- ingly written, full of logical deductions, and, when we consider his multitudinous en- gagements, a remarkable illustration of economy of time. As a contribution to insect psychology, it will be long before this book tinds a parallel." — London Athenceum. " These studiu.s, when handled by such a master as Sir John J.ubbock,rise far above the ordinary dry treatment of such toi)ics. The work is an effort made to discover what are the gener.al. not the special, laws which govern communties of insects com- posed of inhabitants as numerous as the human beings living in London and I'eking.and who labor together in the utmost harmony for the common good. That there are re- markable analogies between societies of ants and human beings no one can doubt. If, according to Mr. Orote. 'positive morality under some form or other has existed in every society of which the world has ever had experience,' the present volume is an effort to show whi'ther this passage be correct or not." — New Tork Times. " In this work the reader will And tlie record of a series of experiments and obser- vations more thorough and ingenious than those instituted by any of the accomplished author's predecessors. . . . Sir^John has been a close observer ot the habits of ants for many years, generally having from thirty to forty communities under his notice, and not only watchint,' each of these in its carefully isolated glass house, but, by the use of paint-marks, following the fortunes of individuals. . . . One notable result of this sys- tem has been the correcting of previous theories as to the age to which ants attain : in- stead of living merely a year, as the popular belief has been, some of Sir John's queens and workers arc thriving after being under observation since 1874 and 1875." — New York World. "Sir John Lubbock's book on 'Ants, Bees, and Wasps' is mainly devoted to the crawlers, and not the fliers, though he has some observations upon hone3'-bees and more interesting ones upon the un|)opular wasp, which he fondly deems to be capable of gratitude. Darwin made a strong case for the monkeys, but Ij»ibbnck may yet make us out to be. as Irishmen say, ' The sons of our ants.' For he begins his enter- taining book thus: ' The anthropoid apes no doubt approach nearer to man in bodily structure than do any other animals, but, when we consider the Imbits of ants, their large communities and elaborate habitations, their roadways, their possession of do- mestic animals, and, even in some cases, of slaves, it must be admitted that i\\e\ have a fjiir claim to rank next to man in the scale of intelligence.' " — Sprim field liepublicun. For sale by all booksellers ; or sent brj mail, post-paid, on receipt of price. New York : D. APPLETON & CO., 1, 3, & 5 Bond Street DISEASES OF MEMORY AN ESSAY IN THE POSITIVE PSYCHOLOGY, By TH. RIBOT, Author of "Heredity," etc. Translated from the French bt WILLIAM HUNTINGTON SMITH. 12mo. Cloth, Sl.SO. ''Not merely to scientific, but to all thinking men, this volume will prove intensely interesting."— iVew York Observer. "M. Ribot has bestowed the most painstaking attention upon his theme, and numerous examples of the conditions considered greatly inci'ease the value and interest ot the volume." — FhUadeljihia North American. " 'Memory,' says M. Ribot, ' is a general function of the nervous system. It is based upon the faculty possessed by the nervous elements of conserving a received modification and of forming associations.' And again: 'Memory Ih a biological fact. A rich and extensive memory is not a collection of impressions, but an acctimulation of dynamical associations, very stable and very responsive to proper stimuli. . . . The brain is like a laboratory full of movement where thousands of operations are going on all at once. Unconscious cerebration, not being subject to restrictions of time, operating, so to speak, only in space, may act in several directions at the same moment. Consciousness is the narrow gate through which a very small part of all this work is able to reach us.' M. Ribot tiius reduces diseases of memory to law, and his treatise is of extraordinary i n te resV—P/iiladeljjhia Press. "The ireneral deductions reached by M. Ribot from the data here collected are summed up in the formulation of a law of regression, based upon the phjfeio- logical principle that "degeneration first afi"ects what has been most recently formed,' and upon the psychological principle that ' the complex disappears before the simple because it has not been repeated so often in experience.' According to this law of regression, the loss of recollection in cases of general dissoiution of the memory follows an invariable path, proceeding from recent events to ideas in general, then to feelings, and lastly to acts. In the best- known cases of partial dissolution or aphasia, forgetfulness follows the same course, beginning with proper names, passing to common nouns, then to ad- jectives and verbs, then to interjections, and lastly to gestures. M. Ribot sub- mits that the exactitude of his laws of regression is verified in those rare cases where proiiressive dissolution of the memory is followed by recovery, recollec- tions being observed to return in an inverse order to that in which they dis- appeared."—A^erf York Sun. "To the general reader the work is made entertaininir by many illustrations connected with such names as Linnaeus, Newton, Sir Walter Scott, Horace Ver- net, Gustave Dore, and many others." — Hanisburg Telegraph. "The whole subject is presented with a Frenchman's vivacity of style."— Providence Journal. '"It is not too much to say that in no single work have so many cnrious CJsas been brought together and interpreted in a scientific manner." — Boston Evening Traveller. '' Specially interesting to the general reader." — Chicago Interior. For sale by all booksellers ; or sent by mail, post-paid, on receipt of 2 rice. New York: D. APrLETON & CO., 1, 3, & 5 Bond Street. MYTH AND SCIENCE By TITO VIGNOLI. 12mo. Cloth, $l.SO. Contexts : The Ideas and Sources of Myth ; Animal Sensation and Perception ; Human Sensation and Perception ; Statement of the Prob- lem ; The Animal and Human Exercise of the Intellect in the Perception of Things ; The Intrinsic Law of the Faculty of Apprehension ; The His- torical Evolution of Myth and Science ; Of Dreams, Illusions, Normal and Abnormal Hallucinations, Delirium, and Madness. " Ilis book is insfenious ; . . . liis theory of hovr science gradually dif- ferentiated from and conquered myth is extremely well wrouglit out, and is probably in essentials correct." — Saturday Review. " Tito Viijnoli's treatise is a valuable contribution to the public book- table at the present moment, when the issues between faith and fact are so much discussed. The author holds that the myth-making faculty is a con- stant attendant of human progress, and that its action is manifest to-day in the most highly cultivated peoples as well as in the most undeveloped. The ditference is, that its activity in the former case is limited, or rather neutralized, by the scientific faculties, and consequently is no longer allowed to grow into legends and mythologies of the primitive pattern. The author traces both myth and science to their common source m sensation and per- ception, whicii he treats under the separate titles of ' animal ' and ' human.' He makes clear the distinctive operations of perception and apprehension, and traces, in a wide survey of history and human life, a most interesting array of examples illustrating the evolution of myth and science." — New York Home Journal. " The book is a strong one, and fiir more interesting to the general reader than its title would indicate. The learning, the acuteness, the strong reasoning power, and the scientific spirit of the author, command admira- tion."— New York Christian Advocate. " An essay of such length as to merit a different title, and of sufficient originality to merit more than common attention." — Chicago Times. " An attempt made, with much ability and no small measure of success, to trace the origin and development of the myth. The author has pursued his inquiry Avith much patience and ingenuity, and has produced a very readable and luminous treatise." — Philadelphia North American. " A very interesting work, which, first published in Italy, created a great deal of interest there, and will scarcely do less in this country." — Mostoii Post. " This intensely interesting volume." — Albany (New York) I^ess. " It is a curious if not startling contribution both to psychology and to the early history of man's development." — New York World. For sale hy all booksellers ; or sent by mail, post-paid, on receipt of price. New York: D. APPLETON & CO., 1, 3, & 5 Bond Street. THE BRAIN AND ITS FUNCTIONS. By J. LUYS, Physician to the Hospice de la Salpetri^re. With Illustrations. 12mo, cloth. Price, $1.50. " No livinpT physiolo'rist is better entitled to speak with authority upon the stnicture ar-d functiors of the brain than Dr. Luys. His studies on the anatomy of the nervous tvstem are acknowledged to be the fullest and most systematic ever undertaken.''" — /St. Jameses Gazette. "Dr. Luys, at the head of the prreat French Insane Asylum, is one of the most eminent and successful investigators of cerebral science now living; and he has given unquestionably the clearest and most interesting brief account yet made of the struct- ure and operations of the brain." — Popular Science Monthly. " It is not too much to say that M. Luys has gone further than any other investi- gator into this great field of study, and only those who are at least dimly aware of the vast changes going on in the realm of psychology can appreciate the importance of his revelations. Particularly interesting and valuable are the chapters dealing with the genesis and evolution of memory, the development of automatic activity, and the de- velopment of the notion of personality." — Boston Evening Traveller. "Thanks to his method of cutting the brain into thin sections, hardening them with chromic acid, photographing them, and then examining the plates through the micro- Bcope, he has succeeded in gaining a knowledge of the structure of the brain which is amazing in extent and startling in its character. But, however advanced his anatoiuy, his physiology is still more so. He has reached conclusions which will be of high im- portance in the treatment of mental diseases and derangements."— ^05/erate9 in producing the phenomena of cerebral physiology, and, carrying the data of contem- porary physiology into the domain of speculative psj'chology. he has endeavored to show that the most complex acts of psycho-intellectual activity are all definitely re- solvable into regular processes and obey regular laws of evolution."— j!/on^r<;a/ Gazette. For sale by all booksellers ; or sent by mail, post-paid, on receipt qf inicc. New York: D. APPLETON & CO., 1, 3, & 5 Bond Street. The Concepts and Theories OF Modern Physics. By J. B. STALLO. 12mo, cloth $1.75. *' Judge Stallo's work is an inquiry into the validity of those mechanical conceptions of the universe which are now held as fundamental in physical science. He takes up the leading modern ductrines which are based upon this mechanical conception, such as the atomic constitution of matter, the kinetic theory of gases, the conservation of energy, the nebular hypothesis, and other views, to find how much stands upon solid empirical ground, and how much rests upon metaphysical speculation. Since the appearance of Dr. Draper's ' Religion and Science,' no book has been published in the country calculated to make so deep an impression on thoughtful and edu- cated readers as this volume. . . . The range and minuteness of the au- thor's learning, the acutcness of his reasoning, and the singular precision and clearness of his style, are qualities which very seldom have been jointly exhibited in a scientific treatise." — Seio York ISun. " Judge J. B. Stallo, of Cincinnati, is a German by birth, and came to this country at about the age of seventeen. He was early familiar with science, and he lectured for some years in an Eastern college : but at length he adopted the profession of law. He is also remembered by many as an author, having a number of years ago written a metaphysical treatise of marked ability for one of hLs vouthtid years. His present book must be road deliberately, must be studied to be appreciated ; but the students of science, as well as those of metaphysics, are certain to be deeply interested in its logical developments. It is a timely and telling contribution to the philosophy of science, imperatively called"^ for by the present exigencies in the progress of knowledge. It is to be commended equally for the solid value of its contents and the scholarly finish of its execution." — The I'op- ular Science Monthly. " The book is of vital interest to a much larger class than specialists — to all, in fact, who value clear thinking or are mterested in the accuracy more than the proi^ress of scientific thought. It deals with the results and theories of physical science, and in no sense with the processes of the labf>- ratory. It is written with a clearness that is uncommon in philosopliic works and with a desire to find truth, conscious of the fact that a prime prerequisite of finding it is to clear the way of accumulated and fast-settling untruths. It is a scientific rebuke, as severe as it is lucid, of the scientists who leave their apparatus and go star-gazing : here is the pit into which they have fallen." — iVie^tf York World. " The volume is an important contribution to scientific discussion, arid is marked by closeness of reasoning, and clearness and cogency of state- ment."— Boston Journal. For 6-j.U by all booksellers; or sent by mail, post paid, on receipt of price. New York: D. APPLETO.N & CO., 1, 3, & 6 Bond St. -^ ENTIFIC SERIES i^^J^w•'CTl'lJWMlnwlwl^^l^llil■lllt»ll■ll■*-« ruiannjiaffin^iifcjrii^tjiiiiiia