Aspects of Nature (Vol. 2 of 2), by Alexander von Humboldt

[24] p. 26.—“The Pothos-form, Aroideæ.”

Caladium and Pothos are exclusively forms of the tropical world; the species of Arum belong more to the temperate zone. Arum italicum, A. dracunculus, and A. tenuifolium, extend to Istria and Friuli. No Pothos has yet been discovered in Africa. India has some species of this genus (Pothos scandens and P. pinnata) which are less beautiful in their physiognomy, and less luxuriant in their growth, than the American species. We discovered a beautiful and truly arborescent member of the group of Aroideæ (Caladium arboreum) having stems from 16 to 21 English feet high, not far from the convent of Caripe, to the East of Cumanas. A very curious Caladium (Culcasia scandens) has been discovered by Beauvois in the kingdom of Benin. (Palisot de Beauvois, Flore d’Oware et de Benin, T. i. 1804, p. 4, pl. iii.) In the Pothos-form the parenchyma is sometimes so much extended that the surface of the leaf is interrupted by holes as in Calla pertusa (Kunth), and Dracontium pertusum (Jacquin), which we collected in the woods round Cumana. The Aroideæ first led attention to the remarkable phenomenon of the fever-heat, which in certain plants is sensible by the thermometer during the development of their inflorescence, and which is connected with a great and temporary increase of the absorption of oxygen from the atmosphere. Lamarck remarked in 1789 this increase of temperature at the time of flowering in Arum italicum. According to Hubert and Bory de St. Vincent the vital heat of Arum cordifolium in the Isle of France was found to rise to 35°and 39° Reaumur, (110°.6 and 119°.6 Fahr.) while the temperature of the surrounding air was only 15°.2 R. (66°.2 F.) Even in Europe, Becquerel and Breschet found as much as 17½° difference, Reaumur (39°.4 Fahr.) Dutrochet remarked a paroxysm, an alternate decrease and increase of vital heat, which appeared to reach a double maximum in the day. Théodore de Saussure observed analogous augmentations of temperature, though to a less amount, only from 0°.5 to 0°.8 of Reaumur’s scale (1°.15 to 1°.8 Fahr.), in plants belonging to other families; for example, in Bignonia radicans and Cucurbita pepo. In the latter plant the use of a very sensitive thermoscope shews that the increase of temperature is greater in the male than in the female plant. Dutrochet, who previous to his early death made such meritorious researches in physics and in vegetable physiology, found by means of thermo-magnetic multiplicators (Comptes rendus de l’Institut, T. viii. 1839, p. 454, T. ix. p. 614 and 781) an increase of vital heat from 0°.1 to 0°.3 Reaumur, (0°.25 to 0°.67 Fahr.) in several young plants (Euphorbia lathyris, Lilium candidum, Papaver somniferum), and even among funguses in several species of Agaricus and Lycoperdon. This vital heat disappeared at night, but was not prevented by placing the plants in the dark during the day-time.

A yet more striking physiognomic contrast than that of Casuarineæ, Needle trees, and the almost leafless Peruvian Colletias, with Aroideæ, is presented by the comparison of those types of the greatest contraction of the leafy organs with the Nymphæaceæ and Nelumboneæ. We find in these as in the Aroideæ, leaves, in which the cellular tissue forming their surface is extended to an extreme degree, supported on long fleshy succulent leaf-stalks; as in Nymphæa alba; N. lutea; N. thermalis (once called N. lotus, from the hot spring of Pezce near Groswardein, in Hungary); the species of Nelumbo; Euryale amazonica of Pöppig; and the Victoria Regina discovered in 1837 by Sir Robert Schomburgk in the River Berbice in British Guiana, and which is allied to the prickly Euryale, although, according to Lindley, a very different genus. The round leaves of this magnificent water plant are six feet in diameter, and are surrounded by turned up margins 3 to 5 inches high, light green inside, and bright crimson outside. The agreeably perfumed flowers, twenty or thirty blossoms of which may be seen at the same time within a small space, are white and rose coloured, 15 inches in diameter, and have many hundred petals. (Rob. Schomburgk, Reisen in Guiana und am Orinoko, 1841, S. 233.) Pöppig also gives to the leaves of his Euryale amazonica which he found near Tefe, as much as 5 feet 8 inches French, or 6 English feet, diameter. (Pöppig, Reise in Chile, Peru und auf dem Amazonenstrome, Bd. ii. 1836, S. 432.) If Euryale and Victoria are the genera which present the greatest extension in all dimensions of the parenchyma of the leaves, the greatest known dimensions of a flower belong to a parasitical Cytinea, the Rafflesia Arnoldi (R. Brown), discovered by Dr. Arnold in Sumatra, in 1818: it has a stemless flower of three English feet diameter, surrounded by large leaf-like scales. Fungus-like, it has an animal smell, resembling beef.

[25] p. 26.—“Lianes, rope-plants, (‘Bush ropes;’ in Spanish, Vejuccos.”)

According to Kunth’s division of the Bauhinieæ, the true genus Bauhinia belongs to the New Continent: the African Bauhinia, B. rufescens, (Lam.) is a Pauletia (Cav.) a genus of which we found some new species in South America. So also the Banisterias, from among the Malpighiaceæ, are properly an American form; although two species are natives of India, and one species, Banisteria leona, described by Cavanilles, is a native of Western Africa. Within the tropics and in the Southern Hemisphere we find among the most different families of plants the twining rope-like climbers which in those regions render the forests at once so impenetrable to man, and on the other hand so accessible and habitable to the Quadrumanæ (or Monkeys) and to the Cercoleptes and the small tiger-cats. The rapid ascent to the tops of lofty trees, the passage from tree to tree, and even the crossing of streams by whole herds or troops of gregarious animals, are all greatly facilitated by these twining plants or Lianes.

In the South of Europe and in North America, Hops from among the Urticeæ, and the species of Vitis from among the Ampelideæ, belong to the class of twining climbers, and between the tropics we find climbing Grasses or Gramineæ. We have seen in the plains of Bogota, in the pass of Quindiu, in the Andes, and in the Quina-producing forests of Loxa, a Bambusacea allied to Nastus, our Chusquea scandens, twine round massive and lofty trunks of trees adorned at the same time with flowering Orchideæ. The Bambusa scandens (Tjankorreh), which Blume found in Java, belongs probably either to the genus Nastus or to that of Chusquea, the Carrizo of the Spanish settlers. Twining plants appear to me to be entirely absent in the Pine-woods of Mexico, but in New Zealand, besides the Ripogonum parviflorum of Robert Brown, (a climber belonging to the Smilaceæ which renders the forests almost impenetrable), the sweet-smelling Freycinetia Banksii, which belongs to the Pandaneæ, twines round a gigantic Podocarpus 220 English feet high, the P. dacryoides (Rich), called in the native language Kakikatea. (Dieffenbach, Travels in New Zealand, 1843, Vol. i. p. 426.)

With climbing Gramineæ and Pandaneæ are contrasted by their beautiful and many-coloured blossoms the Passifloras (among which, however, we even found an arborescent self-supporting species, Passiflora glauca, growing in the Andes of Popayan, at an elevation of 9840 French (10487 English) feet);—the Bignoniaceæ, Mutisias, Alströmerias, Urvilleæ, and Aristolochias. Among the latter our Aristolochia cordata has a crimson-coloured flower of 17 English inches diameter! “flores gigantei, pueris mitræ instar inservientes.” Many of these twining plants have a peculiar physiognomy and appearance produced by the square shape of their stems, by flattenings not caused by any external pressure, and by riband-like wavings to and fro. Cross sections of Bignonias and Banisterias shew cruciform or mosaic figures produced by the mutual pressure and interpenetration of the stems which twine around each other. (See very accurate drawings in Adrien de Jussieu’s Cours de Botanique, p. 77-79, fig. 105-108.)

[26] p. 27.—“The form of Aloës.”

To this group of plants, characterised by so great a similarity of physiognomy, belong; Yucca aloifolia, which extends as far north as Florida and South Carolina; Y. angustifolia (Nutt.) which advances as far as the banks of the Missouri; Aletris arborea; the Dragon-tree of the Canaries and two other Dræcænas from New Zealand; arborescent Euphorbias; Aloë dichotoma (Linn.) (formerly the genus Rhipidodendrum of Willdenow); and the celebrated Koker-boom of Southern Africa with a trunk twenty-one feet high and above four feet thick, and a top of 400 (426 Engl.) feet in circumference. (Patterson, Reisen in das Land der Hottentotten und der Kaffern, 1790, S. 55.) The forms which I have thus brought together belong to very different families: to the Liliaceæ, Asphodeleæ, Pandaneæ, Amaryllideæ, and Euphorbiaceæ; all, however, with the exception of the last, belonging to the great division of the Monocotyledones. A Pandanea, Phytelephas macrocarpa (Ruiz,) which we found in New Granada on the banks of the Magdalena, with its pinnated leaves, quite resembles in appearance a small palm-tree. This Phytelephas, of which the Indian name is Tagua, is besides, as Kunth remarks, the only one of the Pandaneæ found (according to our present knowledge) in the New Continent. The singular Agave-like and at the same time very tall-stemmed Doryanthes excelsa of New South Wales, which was first described by the acutely observing Correa de Serra, is an Amaryllidea, like our low-growing Narcissuses and Jonquils.

In the Candelabra shape of plants of the Aloë form, we must not confound the branches of an arborescent stem with flower-stalks. It is the latter which in the American Aloë (Agave Americana, Maguey de Cocuyza, which is entirely wanting in Chili) as well as in the Yucca acaulis, (Maguey de Cocuy) presents in the rapid and gigantic development of the inflorescence a candelabrum-like arrangement of the flowers which, as is well known, is but too transient a phenomenon. In some arborescent Euphorbias, on the other hand, the physiognomic effect is given by the branches and their division, or by ramification properly so called. Lichtenstein, in his “Reisen im südlichen Africa” (Th. i. S. 370), gives a vivid description of the impression made upon him by the appearance of a Euphorbia officinarum which he found in the “Chamtoos Rivier,” in the Colony of the Cape of Good Hope; the form of the tree was so symmetrical that the candelabrum-like arrangement was regularly repeated on a smaller scale in each of the subdivisions of the larger branches up to 32 English feet high. All the branches were armed with sharp spines.

Palms, Yuccas, Aloes, tall-stemmed Ferns, some Aralias, and the Theophrasta where I have seen it growing luxuriantly, different as they are in the structure of their flowers, yet offer to the eye in the nakedness (absence of branches) of their stems, and in the ornamental character of their tops or crowns, a certain degree of physiognomic resemblance.

The Melanoselinum decipiens (Hofm.), which is sometimes upwards of 10 or 12 feet high, and which has been introduced into our gardens from Madeira, belongs to a peculiar group of arborescent umbelliferous plants to which Araliaceæ are otherwise allied, and with which other plants which will doubtless be discovered in course of time will be associated. Ferula, Heracleum, and Thapsia, do indeed attain a considerable height, but they are still herbaceous plants. Melanoselinum is still almost entirely alone as an umbelliferous tree; Bupleurum (Tenonia) fruticosum (Linn.) of the shores of the Mediterranean; Bubon galbanum of the Cape, and Crithmum maritimum of our sea-shores, are only shrubs. On the other hand, the tropical zone, in which, according to the old and very just remark of Adanson, Umbelliferæ and Cruciferæ are almost entirely wanting in the plains, presented to us on the high ridges of the American Andes the smallest and most dwarf-like of all umbelliferous plants. Among 38 species of plants which we collected at elevations where the mean temperature is below 10° Reaumur (54°.5 Fah.), there vegetate almost like mosses, and as if they made part of the rock and of the often frozen earth, at an elevation of 12600 (13430 English) feet above the level of the sea, Myrrhis andicola, Fragosa arctioïdes, and Pectophytum pedunculare, intermingled with which there is an equally dwarfed Alpine Draba. The only umbelliferous plants growing in the low grounds within the tropics observed by us in the New Continent were two species of Hydrocotyle (H. umbellata and H. leptostachya) between Havannah and Batabano; therefore at the extreme limits of the torrid Zone.

[27] p. 27—“The form of Gramineæ.”

The group of arborescent grasses which Kunth, in his able treatise on the plants collected by Bonpland and myself, has combined under the name of Bambusaceæ, is among the most beautiful adornments of the tropical world. (Bambu, also called Mambu, is a word in the Malay language, but appears according to Buschmann to be of doubtful origin, as the usual Malay expression is buluh, in Java and Madagascar wuluh, voulu.) The number of genera and species which form this group has been extraordinarily augmented by the zeal of botanists. It is now recognised that the genus Bambusa is entirely wanting in the New Continent, to which on the other hand Guadua, from 50 to 60 French or about 53 to 64 English feet high, discovered by us, and Chusquea, exclusively belong; that Arundinaria (Rich) is common to both continents, although the species are different; that Bambusa and Beesha (Rheed.) are found in India and the Indian Archipelago, and Nastus in the Island of Bourbon, and in Madagascar. With the exception of the tall-climbing Chusquea the forms which have been named may be said to replace each other morphologically in the different parts of the world. In the northern hemisphere, in the valley of the Mississipi, the traveller is gratified, long before reaching the tropics, with the sight of a form of bamboo, the Arundinaria macrosperma, formerly called also Miegia, and Ludolfia. In the Southern Hemisphere Gay has discovered a Bambusacea, (a still undescribed species of Chusquea, 21 English feet high, which does not climb, but is arborescent and self-supporting) growing in southern Chili between the parallels of 37° and 42° S. latitude; where, intermixed with Drymis chilensis, a uniform forest covering of Fagus obliqua prevails.

While in India the Bambusa flowers so abundantly that in Mysore and Orissa the seeds are mixed with honey and eaten like rice, (Buchanan, Journey through Mysore, Vol. ii. p. 341, and Stirling in the Asiat. Res. Vol. xv, p. 205) in South America the Guadua flowers so rarely, that in four years we were only twice able to procure blossoms; once on the unfrequented banks of the Cassiquiare, (the arm which connects the Orinoco with the Rio Negro and the Amazons River,) and once in the province of Popayan between Buga and Quilichao. It is striking to see plants in particular localities grow with the greatest vigour without producing flowers: it is thus with European olive trees which have been planted for centuries between the tropics near Quito, 9000 (about 9590 English) feet above the level of the sea, and also in the Isle of France with Walnut-trees, Hazel-nuts, and, as at Quito, olive trees (Olea europea): see Bojer, Hortus Mauritianus, 1837, p. 291.

As some of the Bambusaceæ (arborescent grasses) advance into the temperate zone, so within the tropics they do not suffer from the temperate climate of the mountains. They certainly grow more luxuriantly as social plants from the sea coast to the height of about 2560 English feet; for example, in the province de las Esmeraldas, west of the Volcano of Pichincha, where Guadua angustifolia (Bambusa Guadua in our Plantes équinoxiales, T. i. Tab. xx.) produces in its interior much of the siliceous Tabaschir (Sanscrit tvakkschira, ox-milk). In the pass of Quindiu we saw the Guadua growing at an elevation which we found by barometric measurement to be 5400 (5755 English) feet above the level of the Pacific. Nastus borbonicus is called by Bory de St. Vincent a true Alpine plant; he states that it does not descend lower on the declivity of the Volcano in the Island of Bourbon than 3600 (3837 English) feet. This recurrence or repetition as it were at great elevations of the forms characteristic of the hot plains, recalls the mountain group of palms before pointed out by me (Kunthia Montana, Ceroxylon andicola, and Oreodoxa frigida), and a grove or thicket of Musaceæ sixteen English feet high (Heliconia, perhaps Maranta), which I found growing isolated at an elevation of 6600 (7034 English) feet, on the Silla de Caraccas. (Rélation hist. T. i. p. 605-606.) As, with the exception of a few isolated herbaceous dicotyledones, grasses form the highest zone of phænogamous vegetation round the snowy summits of lofty mountains, so also, in advancing in a horizontal direction towards either pole of the Earth, the phænogamous vegetation terminates with grasses.

To my young friend Joseph Hooker, who, but just returned with Sir James Ross from the frozen antarctic regions, is now exploring the Thibetian portion of the Himalaya, the geography of plants is indebted not only for a great mass of important materials, but also for excellent general deductions. He calls attention to the circumstance that phænogamous flowering plants (grasses) approach 17½° nearer to the Northern than to the Southern pole. In the Falkland Islands near the thick masses of Tussack grass (Dactylis cæspitosa, Forster, according to Kunth a Festuca), and in Tierra del Fuego or Fuegia, under the shade of the birch-leaved Fagus antarctica, there grows the same Trisetum subspicatum which extends over the whole range of the Peruvian Cordilleras, and over the Rocky Mountains to Melville Island, Greenland, and Iceland, and which is also found in the Swiss and Tyrolese Alps, in the Altai mountains, in Kamtschatka, and in Campbell Island, south of New Zealand; therefore, from 54° South to 74½° North latitude, or through 128½° of latitude. “Few grasses,” says Joseph Hooker, in his Flora Antarctica, p. 97, “have so wide a range as Trisetum subspicatum, (Beauv.) nor am I acquainted with any other Arctic species which is equally an inhabitant of the opposite polar regions.” The South Shetland Islands, which are divided by Bransfield Strait from D’Urville’s Terre de Louis Philippe and the Volcano of Haddington Peak, situated in 64° 12´ South latitude and 7046 English feet high, have been very recently visited by a Botanist from the United States of North America, Dr. Eights. He found there (probably in 62° or 62¼°, S. latitude) a small grass, Aira antarctica (Hooker, Icon. Plant. Vol. ii. Tab. 150) which is “the most antarctic flowering plant hitherto discovered.”

In Deception Island, of the same group, S. lat. 62° 50´, lichens only are found, and not a single species of grass; and so also farther to the south-east, in Cockburn Island (lat. 64° 12´), near Palmer’s Land, there were only found Lecanoras, Lecideas, and five Mosses, among which was our German Bryum argenteum: “this seems to be the ultima Thule of antarctic vegetation.” Farther to the south, land-cryptogamic, as well as phænogamic, vegetation is entirely wanting. In the great bay formed by Victoria Land, on a small island which lies opposite to Mount Herschel (S. lat. 71° 49´), and in Franklin Island, 92 geographical miles North of the great volcano Mount Erebus, 12400 English feet high (latitude 76° 7´ South), Hooker found not a single trace of vegetable life. It is quite different in respect to the extension even of the forms of higher vegetable organisation in the high northern latitudes. Phænogamous plants there approach 18½° nearer to the pole than in the southern hemisphere: Walden Island (N. lat. 80½°) has still ten species. The antarctic phænogamous vegetation is also poorer in species at corresponding distances from the pole (Iceland has five times as many flowering plants as the southern group of Auckland and Campbell Islands), but this less varied antarctic vegetation is from climatic reasons more luxuriant and succulent. (Compare Hooker, Flora antarctica, p. vii., 74, and 215, with Sir James Ross, Voyage in the Southern and Antarctic Regions, 1839-1843, Vol. ii. p. 335-342.)

[28] p. 28.—“Ferns.”

If, with a naturalist deeply versed in the knowledge of the Agamæ, Dr. Klotzsch, we estimate the whole number of cryptogamic species hitherto described at 19000, this gives to Fungi 8000 (of which the Agarici constitute 1-8th); Lichens, according to J. von Flotow of Hirschberg, and Hampe of Blankenburg, at least 1400; Algæ 2580; Mosses and Liver-worts, according to Carl Müller of Halle, and Dr. Gottsche of Hamburgh, 3800; and Ferns 3250. We are indebted for this last important result to the thorough investigation of all that is known concerning this group of plants by Professor Kunze of Leipsic. It is remarkable that of the entire number of described Filices the family of Polypodiaceæ, alone, comprises 2165 species; while other forms, even Lycopodiaceæ and Hymenophyllaceæ, only count 350 and 200. There are, therefore, almost as many described ferns as described grasses.

It is remarkable that in the ancient classic writers, Theophrastus, Dioscorides, and Pliny, no notice occurs of the beautiful form of arborescent ferns; while from information derived from the companions of Alexander, Aristobulus, Megasthenes, and Nearchus, mention is made of Bamboos “quæ fissis internodiis lembi vice vectitabant navigantes;” of the Indian trees “quarum folia non minora clypeo sunt;” of the fig-tree of which the branches take root round the parent stem; and of Palms “tantæ proceritatis, ut sagittis superjici nequeant.” (Humboldt, de Distributione geogr. Plantarum, p. 178 and 213.) I find the first description of tree-ferns in Oviedo’s Historia de las Indias, 1535, fol. xc. This experienced traveller, who had been placed by Ferdinand the Catholic as director of the gold-washings in Hayti, says: “Among the many ferns there are some which I reckon among trees, for they are as thick and as tall as pines (Helechos que yo cuento por arboles, tan gruesos como grandes pinos y muy altos). They grow chiefly in the mountains and where there is much water.” The height is exaggerated. In the dense forests round Caripe even our Cyathea speciosa only attains a height of 30 to 35 (32 to 37 English) feet; and an excellent observer, Ernst Dieffenbach, in the northernmost of the three islands of New Zealand saw no stems of Cyathea dealbata of more than 40 (42½ English) feet in height. In the Cyathea speciosa and the Meniscium of the Chaymas missions we observed, in the midst of the shadiest primeval forest, in very luxuriantly growing individuals, the scaly stems covered with a shining carbonaceous powder. It seemed like a singular decomposition of the fibrous parts of the old frond stalks. (Humboldt, Rel. hist. T. i. p. 437.)

Between the tropics, where, on the declivities of the Cordilleras, climates are placed successively in stages one above another, the proper zone of the tree-ferns is between three and five thousand feet (about 3200 and 5330 English) above the level of the sea. In South America and in the Mexican highlands they seldom descend lower towards the plains than 1200 (about 1280 Eng.) feet. The mean temperature of this happy zone falls between 17° and 14°.5 Reaumur (70°.2 and 64°.6 Fahr.) This region enters the lowest stratum of clouds, or that which floats next above the sea and the plains; and hence, besides great equality of temperature, it also enjoys uninterruptedly a high degree of humidity. (Robert Brown, in Appendix to Expedition to Congo, p. 423.) The inhabitants, who are of Spanish descent, call this zone “tierra templada de los helechos.” The Arabic word for fern is feledschun, f being changed into h in helechos according to Spanish custom: perhaps the Arabic feledschun is connected with “faladscha,” “it divides;” in allusion to the finely divided margins of fern leaves or fronds. (Abu Zacaria Ebn el Awam, Libro de Agricultura, traducido por J. A. Banqueri, T. ii. Madr. 1802, p. 736.)

The conditions of mild temperature and an atmosphere nearly saturated with vapour, together with great equability of climate in respect to both temperature and moisture, are fulfilled on the declivities of the mountains, in the valleys of the Andes, and above all in the mild and humid atmosphere of the southern hemisphere, where arborescent ferns extend not only to New Zealand and Van Diemen Island (Tasmania), but even to the Straits of Magellan and to Campbell Islands, or to a latitude almost corresponding to that of Berlin in the northern hemisphere. Of tree-ferns, Dicksonia squarrosa grows vigorously in 46° South latitude, in Dusky Bay (New Zealand); D. antarctica of Labillardière in Tasmania; a Thyrsopteris in Juan Fernandez; an undescribed Dicksonia with stems from 12 to 15 (nearly 13 to 16 English) feet in the south of Chili, not far from Valdivia; and a Lomaria of rather less height in the Straits of Magellan. Campbell Island is still nearer to the south pole, in 52½° lat., and even there the stem of the Aspidium venustum rises to 4 feet (4 feet 3 inches, English) before the fronds branch off.

The climatic relations under which Ferns in general flourish, are manifested in the numerical laws of their quotients of distribution taken in the manner alluded to in an earlier part of the present volume. In the low plains of the great continents within the tropics, the quotient for ferns is, according to Robert Brown, and according to late researches, 1-20th of all the species of phænogamous plants growing in the same region; in the mountainous parts of the great continents in the same latitudes it is from 1-8th to 1-6th. But a very different ratio is found in the small islands dispersed over the wide ocean. The proportion of ferns to the whole number of Phanerogamæ increases there in such a manner that in the groups of islands between the tropics in the Pacific the ferns equal a fourth,—and in the solitary far detached islands in the Atlantic Ocean, St. Helena, and Ascension,—almost equal the half of the entire phænogamous vegetation. (See an excellent memoir of D’Urville entitled Distribution géographique des Fougères sur la surface du Globe, in the Annales des Sciences Nat. T. vi. 1825, p. 51, 66, and 73). From the tropics (where in the great continents D’Urville estimates the ratio generally at 1:20) we see the relative frequency of ferns decrease rapidly in the temperate zone. The quotients are: for North America and for the British Islands ¹⁄₃₃, for France ¹⁄₅₈, for Germany ¹⁄₅₂, for the dry parts of the south of Italy ¹⁄₇₄, and for Greece ¹⁄₈₄. Towards the colder regions of the north we see the relative frequency increase again rapidly; that is to say, the number of species of ferns decreases much more slowly than does the number of species of phænogamous plants. At the same time, the luxuriance, abundance, and mass of individuals in each species augments the illusive impression of absolute numbers. According to Wahlenberg’s and Hornemann’s Catalogues the relative numbers of Filices are, for Lapland ¹⁄₂₅, for Iceland ¹⁄₁₈, and for Greenland ¹⁄₁₂.

Such, according to the present state of our knowledge, are the natural laws manifested in the distribution of the pleasing form of Ferns. But it would seem as if in the family of Ferns, which has so long been regarded as a cryptogamic family, we had quite recently arrived on the traces of another natural law, a morphological one of propagation. Count Leszczyc-Suminski, who happily unites the gift of microscopic examination with distinguished artistic talent, has discovered in the prothallium of ferns an organisation by which fructification is effected. He distinguishes a bisexual arrangement in the ovule-like cell on the middle of the theca, and in the ciliated antheridia or spiral threads before examined by Nägeli. The fertilisation is supposed to take place not by pollen tubes but by the moveable ciliated spiral threads. (Suminski zur Entwickelungs-geschichte der Farrnkräuter, 1848, S. 10-14.) According to this view, Ferns, as Ehrenberg expresses it (Monatl. Berichte der Akad. zu Berlin, Januar 1848, S. 20), would be produced by a microscopic fertilisation taking place on the prothallium as a receptacle; and throughout the whole remainder of their often arborescent development they would be flowerless and fruitless plants, forming buds or bulbs; the spores or sori on the under side of the frond not being seeds but flower buds.

[29] p. 28.—“Liliaceæ.”

The principal seat of this form is Africa, where it is both most varied and most abundant, and where these beautifully flowering plants are assembled in masses and determine the aspect and character of the country. The New Continent does, indeed, also possess superb Alstromeriæ and species of Pancratium, Hæmanthus, and Crinum (we augmented the first-named of these genera by nine, and the second by three species); but these American Liliaceæ grow dispersed, and are less social than our European Irideæ.

[30] p. 28.—“Willow Form.”

Of the leading representative of this form, the Willow itself, 150 different species are already known. They are spread over the northern hemisphere from the Equator to Lapland. They appear to increase in number and diversity of form between the 46th and 70th degrees of north latitude, and especially in the part of north of Europe where the configuration of the land has been so strikingly indented by early geological changes. Of Willows as tropical plants I am acquainted with ten or twelve species, which, like the willows of the southern hemisphere, are deserving of particular attention. As Nature seems as it were to take pleasure in multiplying certain forms of animals, for example Anatidæ (Lamellirostres) and Columbæ, in all the zones of the earth; so are Willows, the different species of Pines, and Oaks, no less widely disseminated: the latter (oaks) being always alike in their fruit, though much diversified in the forms of their leaves. In Willows, the similarity of the foliage, of the ramification, and of the whole physiognomic appearance, in the most different climates, is unusually great,—almost greater than even in Coniferæ. In the southern part of the temperate zone of the northern hemisphere the number of species of willows decreases considerably, yet (according to the Flora atlantica of Desfontaines) Tunis has still a species of its own resembling Salix caprea; and Egypt reckons, according to Forskäl, five species, from the catkins of whose male flowers a medicine much employed in the East, Moie chalaf (aqua salicis) is obtained by distillation. The Willow which I saw in the Canaries is also, according to Leopold von Buch and Christian Smith, a peculiar species, common however to that group and to the Island of Madeira,—S. canariensis. Wallich’s Catalogue of the plants of Nepaul and of the Himalaya cites from the Indian sub-tropical zone thirteen species, partly described by Don, Roxburgh, and Lindley. Japan has its indigenous willows, one of which, S. japonica (Thunb.) is also found as a mountain plant in Nepaul.

Previous to my expedition, the Indian Salix tetrasperma was the only known intertropical species, so far as I am aware. We collected seven new species, three of which were from the elevated plains of Mexico, and were found to extend to an elevation of 8000 (about 8500 English) feet above the level of the sea. At still greater elevations,—for example, on the mountain plains situated between 12000 and 14000 feet, (about 12790 and 14920 English,) which we often visited,—we did not find, either in the Andes of Mexico or in those of Quito and Peru, anything which could recall the small creeping alpine willows of the Pyrenees, the Alps, and Lapland (S. herbacea, S. lanata, and S. reticulata). In Spitzbergen, where the meteorological conditions have much analogy with those of the Swiss and Scandinavian snow-mountains, Martins described two dwarf willows, of which the small woody stems and branches creep on the ground, and which lie so concealed in the turf-bogs that their small leaves are only discovered with difficulty under the moss. The species found by me in Peru in 4° 12´ S. latitude, near Loxa, at the entrance of the forests where the best Cinchona bark is collected, and described by Willdenow as Salix humboldtiana, is the one which is most widely distributed in the western part of South America. A sea-shore species, S. falcata, which we found on the sandy coast of the Pacific, near Truxillo, is, according to Kunth, probably only a variety of the above; and possibly the fine and often pyramidal willow which accompanied us along the banks of the Magdalena, from Mahates to Bojorque, and which, according to the report of the natives, had only extended so far within a few years, may also be identical with Salix humboldtiana. At the confluence of the Rio Opon with the Magdalena, we found all the islands covered with willows, many of which had stems 64 English feet high, but only 8 to 10 inches in diameter. (Humboldt and Kunth, Nova Gen. Plant. T. ii. p. 22, tab. 99.) Lindley has made us acquainted with a species of Salix from Senegal, and therefore in the African equinoctial zone. (Lindley, Introduction to the Natural System of Botany, p. 99.) Blume also found two species of Salix near the equator, in Java: one wild and indigenous, S. tetrasperma; and another cultivated, S. sieboldiana. From the southern temperate zone I know only two willows described by Thunberg, (S. hirsuta and S. mucronata); they grow by the side of Protea argentea (which has itself very much the physiognomy of a willow), on the banks of the Orange River, and their leaves and young shoots form the food of the hippopotamus. Willows are entirely wanting in Australia and the neighbouring islands.

[31] p. 29.—“Myrtaceæ.”

An elegant form, with stiff, shining, thickly set, generally unindented, small leaves, studded with pellucid dots. Myrtaceæ give a peculiar character to three districts of the earth’s surface,—the South of Europe, particularly the calcareous and trachytic islands which rise above the surface of the Mediterranean;—the continent of New Holland, adorned with Eucalyptus, Metrosideros, and Leptospermum;—and an intertropical region, part of which is low, and part from nine to ten thousand feet high (about 9590 to 10660 English), in the Andes of South America. This mountain district, called in Quito the district of the Paramos, is entirely covered with trees which have a myrtle-like aspect and character, even though they may not all belong to the natural family of Myrtaceæ. Here, at the above-named elevation, grow the Escallonia myrtilloides, E. tubar, Simplocos alstonia, some species of Myrica, and the beautiful Myrtus microphylla which we have figured in the Plantes équinoxiales, T. i. p. 21, Pl. iv. We found it growing on mica slate, and extending to an elevation of more than ten thousand English feet, on the Paramo de Saraguru, near Vinayacu and Alto de Pulla, which is adorned with so many lovely alpine flowering plants. Myrtus myrsinoides even extends in the Paramo de Guamani up to 10500 (11190 English) feet. Of the 40 species of the Genus Myrtus which we collected in the equinoctial zone, and of which 37 were undescribed, much the greater part belonged, however, to the plains and lower mountains. From the mild tropical mountain climate of Mexico we brought back only a single species (Myrtus xalapensis); but the Tierra templada, towards the Volcano of Orizaba, must no doubt contain several more. We found M. maritima near Acapulco, quite on the sea-coast of the Pacific.

The Escallonias,—among which E. myrtilloides, E. tubar, and E. floribunda, are the ornament of the Paramos, and by their physiognomy remind the beholder strongly of the myrtle-form,—once constituted, in combination with the European and South American Alp-roses (Rhododendrum and Befaria), and with Clethra, Andromeda, and Gaylussaccia buxifolia, the family of Ericeæ. Robert Brown (see the Appendix to Franklin’s Narrative of a Journey to the Shores of the Polar Sea, 1823, p. 765), has raised them to the rank of a separate family, which Kunth places between Philadelpheæ and Hamamelideæ. The Escallonia floribunda offers in its geographical distribution one of the most striking examples, in the habitat of the plant, of proportion between distance from the equator and vertical elevation above the level of the sea. In making this statement I again support myself on the authority of my acute and judicious friend Auguste de St.-Hilaire (Morphologie végétale, 1840, p. 52):—“Messieurs de Humboldt et Bonpland ont découvert dans leur expédition l’Escallonia floribunda à 1400 toises par les 4° de latitude australe. Je l’ai retrouvé par les 21° au Brésil dans un pays élevé, mais pourtant infiniment plus bas que les Andes du Pérou: il est commun entre les 24°.50´ et les 25°.55´ dans les Campos Geræs, enfin je le revois au Rio de la Plata vers les 35°, au niveau même l’ocean.”

Trees belonging the group of Myrtaceæ,—to which Melaleuca, Metrosideros, and Eucalyptus belong in the sub-division of Leptospermeæ,—produce partially, either where the leaves are replaced by phyllodias (leaf-stalk leaves), or by the peculiar disposition or direction of the leaves relatively to the unswollen leaf-stalk, a distribution of stripes of light and shade unknown in our forests of round-leaved trees. The first botanical travellers who visited New Holland were struck with the singularity of the effect thus produced. Robert Brown was the first to show that this strange appearance arose from the leaf-stalks (the phyllodias of the Acacia longifolia and A. suaveolens) being expanded in a vertical direction, and from the circumstance that the light instead of falling on horizontal surfaces, falls on and passes between vertical ones. (Adrien de Jussieu, Cours de Botanique, p. 106, 120, and 700; Darwin, Journal of Researches, 1845, p. 433). Morphological laws in the development of the leafy organs determine the peculiar character of the effects produced, the outlines of light and shade. “Phyllodias,” says Kunth, “can, according to my view, only occur in families which have compound pinnated leaves; and in point of fact they have as yet only been found in Leguminosæ, (in Acacias). In Eucalyptus, Metrosideros, and Melaleuca, the leaves are simple (simplicia), and their edgewise position arises from a half turn or twist of the leaf-stalk (petiolus); it should be remarked at the same time that the two surfaces of the leaves are similar.” In the comparatively shadeless forests of New Holland the optical effects here alluded to are the more frequent, as two groups of Myrtaceæ and Leguminosæ, species of Eucalyptus and of Acacia, constitute almost the half of all the greyish green trees of which those forests consist. In addition to this, in Melaleuca there are formed between the layers of the inner bark easily detached portions of epidermis which press outwards, and by their whiteness remind the European of our birch bark.

The distribution of Myrtaceæ is very different in the two continents. In the New Continent, and especially in its western portion, it scarcely extends beyond the 26th parallel of north latitude, according to Joseph Hooker (Flora antarctica, p. 12); while in the Southern Hemisphere, according to Claude Gay, there are in Chili 10 species of Myrtus and 22 species of Eugenia, which, intermixed with Proteaceæ (Embothrium and Lomatia), and with Fagus obliqua, form forests. The Myrtaceæ become more abundant beyond 38° S. lat.,—in the Island of Chiloe, where a Metrosideros-like species of Myrtus (Myrtus stipularis) forms almost impenetrable thickets under the name of Tepuales; in Patagonia; and in Fuegia to its extremity in 56½° S. lat. In the Old Continent they prevail in Europe as far as the 46th parallel of North latitude: in Australia, Tasmania, New Zealand, and the Auckland Islands, they advance to 50½° South latitude.

[32] p. 29.—“Melastomaceæ.”

This group comprises the genera Melastoma (Fothergilla and Tococa Aubl.) and Rhexia (Meriana and Osbeckia), of which we found, on either side of the equator in tropical America alone, 60 new species. Bonpland has published a superb work on Melastomaceæ, in two volumes, with coloured drawings. Some species of Rhexia and Melastoma ascend in the Andes, as alpine or Paramos shrubs, as high as nine and ten thousand five hundred (about 9600 and 11190 English) feet: among these are Rhexia cernua, R. stricta, Melastoma obscurum, M. aspergillare, and M. lutescens.

[33] p. 29.—“Laurel-form.”

To this form belong the genera of Laurus and Persea, the Ocoteæ so numerous in South America, and (on account of physiognomic resemblance), Calophyllum and the superb aspiring Mammea, from among the Guttiferæ.

[34] p. 29.—“How interesting and instructive to the landscape painter would be a work which should present to the eye the leading forms of vegetation!”

In order to define somewhat more distinctly what is here only briefly alluded to, I permit myself to introduce some considerations taken from a sketch of the history of landscape painting, and of a graphical representation of the physiognomy of plants, which I have given in the second volume of Kosmos (Bd. ii. S. 88-90; English edit. vol. ii. p. 86-87).

“All that belongs to the expression of human emotion and to the beauty of the human form, has attained perhaps its highest perfection in the northern temperate zone, under the skies of Italy and Greece. By the combined exercise of imitative art and of creative imagination, the artist has derived the types of historical painting at once from the depths of his own mind, and from the contemplation of other beings of his own race. Landscape painting, though no merely imitative art, has, it may be said, a more material substratum and a more terrestrial domain: it requires a greater mass and variety of distinct impressions, which the mind must receive within itself, fertilize by its own powers, and reproduce visibly as a free work of art. Hence landscape painting must be a result at once of a deep and comprehensive reception of the visible spectacle of external nature, and of this inward process of the mind.”

“Nature, in every region of the earth, is indeed a reflex of the whole; the forms of organised beings are repeated everywhere in fresh combinations; even in the icy north, herbs covering the earth, large alpine blossoms, and a serene azure sky, cheer a portion of the year. Hitherto landscape painting has pursued amongst us her pleasing task, familiar only with the simpler form of our native floras, but not, therefore, without depth of feeling, or without the treasures of creative imagination. Even in this narrower field, highly gifted painters, the Caracci, Gaspar Poussin, Claude Lorraine, and Ruysdael, have with magic power, by the selection of forms of trees and by effects of light, found scope wherein to call forth some of the most varied and beautiful productions of creative art. The fame of these master-works can never be impaired by those which I venture to hope for hereafter, and to which I could not but point, in order to recall the ancient but deeply-seated bond which unites natural knowledge with poetry and with artistic feeling; for we must ever distinguish in landscape painting, as in every other branch of art, between productions derived from direct observation, and those which spring from the depths of inward feeling and from the power of the idealising mind. The great and beautiful works which owe their origin to this creative power of the mind applied to landscape-painting, belong to the poetry of nature, and like man himself, and the imagination with which he is gifted, are not rivetted to the soil, or confined to any single region. I allude here more particularly to the gradation in the form of trees from Ruysdael and Everdingen, through Claude Lorraine to Poussin and Annibal Caracci. In the great masters of the art we perceive no trace of local limitation; but an enlargement of the visible horizon, and an increased acquaintance with the nobler and grander forms of nature, and with the luxuriant fulness of life in the tropical world, offer the advantage not only of enriching the material substratum of landscape painting, but also of affording a more lively stimulus to less gifted artists, and of thus heightening their powers of production.”