THE

AMERICAN

JOURNAL OF SCIENCE AND ARTS.

[SECOND SERIES.]

ART. XXVI.-On the Origination and Distribution of Vegetable Species:-Introductory Essay to the Flora of Tasmania; by Dr. JOSEPH D. HOOKER.

(Continued from p. 25.)

$4. On the General Phenomena of the Distribution of Plants in

Time.

A THIRD class of facts relates to the antiquity of vegetable forms and types on the globe, as evidenced by fossil plants. The chief facts relating to these are the following:

31. The earliest Flora of which we know much scientifically, is that of the Carboniferous formation. We have indeed plants that belonged to an earlier vegetation, but they do not differ in any important respects from those of the carboniferous formation. Now the ascertained features of the coal vegetation may be summed up very briefly. There existed at that time,

Filices, in the main entirely resembling their modern representatives, and some of which may even be generically, though not specifically, identical with them.

Lycopodiaceae; the same in the main characters as those now existing, and, though of higher specialization of stem, of greater stature, of different species, and perhaps also genera, from modern Lycopodiaceae, yet identical with these in the structure of their reproductive organs and their contents, and in the minute anatomy of their tissues.

SECOND SERIES, Vol. XXIX, No. 87.-MAY, 1860.

Coniferæ. The evidence of this order is derived chiefly from the anatomical characters of the Dicotyledonous wood so abundantly found in the coal, and which seems to be identical in all important respects with the wood of modern genera of that order, to which must be added the probability of Trigonocarpon and Naggerathia being Gymnospermous, and allied to Salisburia.* On the other hand, it must not be overlooked that no Coniferous strobili have been hitherto detected in the Carboniferous formation.

Cycadea. Some fragments of wood, presenting a striking similarity in anatomical characters to that of Cycadea, have been found in the carboniferous series.

In the absence of the fructification of Calamites, Calamodendron, Halonia, Anabathra, etc., there are no materials for any safe conclusions as to their immediate affinities, beyond that they all seem to be allied to Ferns or Lycopodiaceae. But the same can hardly be said of the affinities of Volkmannia, Antholites, and others, which have been referred, with more or less probability, to Angiospermous Dicotyledons.

The Permian Flora is for the most part specifically distinct from the Carboniferous, but many of its genera are the same. The prevalent types are Gymnospermous Dicotyledons, especially Cycadece, and a great abundance of Tree-ferns.

The New Red Sandstone, or Trias group, presents plants more analogous to those of the Oolite than to those of the Carboniferous epoch, but they have also much in common with the latter. Voltzia, a remarkable genus of Conifers, appears to be peculiar to this period.

In the Lias numerous species of Cycade have been found, with various Conifers and many Ferns. No other Dicotyledonous or any Monocotyledonous plants have as yet been discov ered, but it is difficult to believe that none such should have existed at a period when wood-boring and herb-devouring insects, belonging to modern genera, were extremely abundant, as has been proved by the researches of Mr. Brodie and Mr. Westwood.t

The Oolite contains numerous Cycadea, Conifera, and Ferns, and more herbivorous genera of insects; and here Monocotyledonous vegetables are recognizable in Podocarya and other Pandaneous plants. A cone of Pinus has been discovered in the Purbeck, and one of Araucaria in the inferior Oolite of Somersetshire.

* Phil. Trans., 1855, p. 149.

See Quarterly Journal of the Geological Society, May, 1854.

These insects include species of the existing common European genera, Elater, Gryllus, Hemerobius, Ephemera, Libellula, Panorpa, and Carabus. Of all conspicuous tribes of plants the Cycadea, Filices, Coniferæ, and Lycopodiacere perhaps support the fewest insects, and the association of the above-named insects with a vegetation consisting solely or mainly of plants of these orders is quite inconceivable.

In the Cretaceous group, Dicotyledons of a very high type appear. A good many species are enumerated by Dr. Debey, of Aix-la-Chapelle, including a species of Juglans, a genus belonging to an order of highly-developed floral structure and complex affinities.†

Characea appear for the first time at this epoch, and are apparently wholly similar in structure to those of the present day.

The Tertiary strata present large assemblages of plants of so many existing genera and orders, that it can hardly be doubted. but that even the earliest Flora of that period was almost as complex and varied as that of our own. In the lowest Eocene beds are found Anonaceae, Nipa, Acacia, and Cucurbitaceae. In the Bagshot sands some silicified wood has been found, which may confidently be referred to Banksia, and which is, in fact, scarcely distinguishable from recent and fossil Australian Banksia wood.§

In the brown coal of the Eocene and Miocene periods, Fanpalms, Conifers, and various existing genera of Myricea, Laurinece, and Platanea are believed to have been identified. Wesel and Weber describe from the brown coal of the Rhine a rich and varied Flora, representing numerous families never now seen associated, and including some of the peculiar and characteristic genera of the Australian, South African, American, Indian, and European floras.

* Quart. Journ. Geol. Soc., vii, pt. 1, misc., p. 110.

Professor Oswald Heer, of Zurich, in an interesting little paper (Quelques Mots sur les Noyers), in Bibl. Univ. Genev., Sep. 1858, argues from the fact of the early appearance of Juglans in the geological series, that this genus must be a low type of the Dicotyledonous class to which it belongs. The position of Juglans is unsettled in the present state of our classification of Dicotyledonous orders, as it has equal claims to be ranked with Terebinthacea, which are very high in the series, and with Cupuliferæ, which are placed very low; and were the grounds for our thus ranking these orders based on characters of ascertained relative value, such an argument might be admissible; but the system which sunders these orders is a purely artificial one, and Juglans with its allies would prove it so, if other proofs were wanting; for it absolutely combines Terebinthaceae and Cupulifera into one natural group, in which (as in so many others) there is a gradual passage from great complexity of floral organs to great simplicity.

I am far from considering the identification of these and the other genera which I have enumerated in various strata as satisfactory, but I conclude that they may be taken as evidence of as highly developed and varied plants having then existed as are now represented by these genera.

§ I am indebted to the late Robert Brown for this fact, and for the means of comparing the specimens, which are beautifully opalized. I ascertained that he was satisfied with the evidence of this wood having really been dug up near Staines, though it is so perfectly similar in every respect to the opalized Banksia-wood of Tasmania as to suggest to his mind and my own the most serious doubts as to its English origin.

See Quart. Journ. Geol. Soc., xv, misc. 3, where an abstract is given, with some excellent cautions, by C. J. F. Bunbury, Esq. The Australian genera include Encalyptus, Casuarina, Leptomeria, Templetonia, Banksia, Dryandra, and Hakea. I am not prepared to assert that these identifications, or the Australian ones of the Mollasse, are all so unsatisfactory that the evidence of Australian types in the brown coal and Mollasse should be altogether set aside; but I do consider that not one of the above-named genera is identified at all satisfactorily, and that many of them are not even problematically decided.

In the Mollasse and certain Miocene formations at Eningen and elsewhere in Germany, Switzerland, and Tuscany,* 900 species of Dicotyledonst have been observed, all apparently different from existing ones. They have been referred, with more or less probability, to Fan-palms, Poplars (three species), evergreen Laurine, Ceratonia, Acacia, Tamarindus, Banksia, Embothrium, Grevillea, Cupressus, several species of Juglans (one near the North American J. acuminata, another near the common walnut of Europe and Asia, J. regia, and a third near the North American J. cinerea); also a Hickory, near the Carya alba (a genus now wholly American), and a Pterocarya closely allied to P. Caucasica.

The rise of the Alps was subsequent to this period; and in the European deposits immediately succeeding that event, in Switzerland (at Durnten and Utznach) are found evidences of the following existing species,-Spruce, Larch, Scotch Fir, Birch, a Hazel (different from that now existing), Scirpus lacustris, Phragmites communis, and Menyanthes trifoliata.

The glacial epoch followed, during and since which there has probably been little generic change in the vegetation of the globe.

32. So much for the main facts hitherto regarded as estab lished in Vegetable Palæontology. They are of little value as compared with those afforded by the Animal Kingdom, even granting that they are all well made out, which is by no means the case. In applying them theoretically to the solution of the question of creation and distribution, the first point which strikes us is the impossibility of establishing a parallel between the successive appearances of vegetable forms in time, and their complexity of structure or specialization of organs, as represented by the successively higher groups in the natural method of classification. Secondly, that the earliest recognizable Cryptogams

*During the printing of this sheet I have received from my friend M. DeCandolle a very interesting memoir on the tertiary fossil plants of Tuscany, by M. C. Gaudin and the Marquis C. Strozzi, in which some of the genera here alluded to are described. The age of these Tuscan beds is referred by Prof. O. Heer to a period intermediate between those of Utznach and Eningen, The most important plants described are, Coniferæ 6 sp., Salix 2, Liquidambar 1, Alnus 1, Carpinus 1, Populus 2, Fagus 1, Quercus 5, Ulmus 2, Planera 1, Ficus 1, Platanus 1, Oreodaphne 1, Laurus 2, Persea 1, Acer 2, Vitis 1, Juglans 4, Carya 1, Pterocarya 1. There are 49 extinct species in all, of which 46 are referred, without even a mark of doubt or caution, to existing genera, and this in almost all cases from imperfect leaves alone! Without questioning the good faith or ability of the authors of this really valuable and interesting memoir, I cannot withhold my protest against this practice of making what are at best little better than surmises, appear under the guise of scientifically established identifications. What confidence can be placed in the positive reference of supposed fossil Fungi to Sphæria, or of pinnated leaves to Sapindus, and other fragments of foliage to existing genera of Laurineæ, Ficus, and Vitis? O. Heer, Sur les Charbons feuilletés de Durnten et Utznach, in Mem. Soc. Helvet. Sc. Nat. 1857; Bibl. Univers. Genev., August, 1858.

should not only be the highest now existing, but have more highly differentiated vegetative organs than any subsequently appearing; and that the dicotyledonous embryo and perfect exogenous wood with the highest specialized tissue known (the coniferous, with glandular tissue*), should have preceded the monocotyledonous embryo and endogenous wood in date of appearance on the globe, are facts wholly opposed to the doctrine of progression, and they can only be set aside on the supposition that they are fragmentary evidence of a time further removed from that of the origin of vegetation than from the present day; to which must be added the supposition that types of Lycopodi aceœ, and a number of other orders and genera, as low as those now living, existed at that time also.

33. Another point is the evidence,† said to be established, of genera now respectively considered peculiar to the five continents having existed cotemporaneously at a comparatively recent geological epoch in Europe, and the very close affinity, if not identity, of some of these with existing species. The changes in the level and contour of the different parts of the earth's surface which have occurred since the period of the chalk, or even since that preceding the rise of the Alps, imply a very great amount of difference between the past and present relations of sea and land and climate; and it is no doubt owing to these changes that the Araucariae, which once inhabited England, are no longer found in the northern hemisphere, and that the Australian genera which inhabited Europe at a period preceding the rise of the Alps have since been expelled.

34. Such facts, standing at the threshold of our knowledge of vegetable palæontology, should lead us to expect that the problem of distribution is an infinitely complicated one, and suggest the idea that the mutations of the surface of our planet, which replace continents by oceans, and plains by mountains, may be insignificant measures of time when compared with the duration of some existing genera and perhaps species of plants, for some of these appear to have outlived the slow submersion of continents.

* The vexed question of the true position of Gymnospermous plants in the Natural System assumes a somewhat different aspect under the view of species being created by progressive evolution. In the haste to press the recent important dis coveries in vegetable impregnation and embryogeny into the service of classification, the long-established facts regarding the development of the stem, flower, and reproductive organs themselves of Gymnospermous plants have been relatively underrated or wholly lost sight of; and if an examination of the doctrines of progression and variation lead to a better general estimation of the comparative value of the characters presented by these organs, the acceptance or rejection of the doctrines themselves is, in the present state of science, a matter of secondary import

ance.

See fifth foot-note of p. 307 (1): what I have there said of the supposed identifications of the Australian genera applies to many of those of the other enumerated quarters of the globe.