©Ijg ^. p. pm pfararg This BOOK may be kept out TWO WE. ONLY, and is subject to a fine of F. CENTS a day thereafter. It is due on day indicated below: 27 J an s> MJL 1 5 11977 SCIENTIFIC PAPERS OF ASA GRAY SELECTED BY CHARLES SPRAGUE SARGENT VOL. I. REVIEWS OF WORKS ON BOTANY AND RELATED SUBJECTS 1834-1887 BOSTON AND NEW YORK HOUGHTON, MIFFLIN AND COMPANY 1889 Copyright, 1889, Br CHARLES SPRAGUE SARGENT. All rights reserved. The Riverside Press, Cambridge : Electrotyped and Printed by 11. 0. Houghton & Co. INTEODUCTION. Asa Geay's first scientific paper was published in 1834 ; his last was written in 1887, a few weeks before the end of his life. The number of his contributions to science and their variety is remarkable, and astonishes his associates even, familiar as they were with his intellectual activity, his vari- ous attainments, and that surprising industry which neither assured position, the weariness of advancing years, nor the hopelessness of the task he had imposed upon himself, ever diminished. Professor Gray's writings may be naturally grouped in four divisions. The first in importance contains his contributions to descriptive botany. These with few exceptions were de- voted to the flora of North America, and although it did not fall to his lot, as it did to that of some of his contemporaries, to elaborate any one of the great families of plants, the extent and character of his contributions to systematic botany will i^lace his name among those of the masters of the science. His works of a purely educational character are only second in importance to his writings on the flora of North America ; and their influence upon the development of botanical knowl- edge in this country, during the half century which elapsed between the publication of the first and the last of the series, has been great and must long be felt. No text-books of sci- ence surpass them in the philosophical treatment of the sub- jects they embrace, or in the beauty and clearness of their style. A series of critical reviews of important scientific publi- cations, and of historical accounts of the lives and labors of botanical worthies, may be conveniently grouped in the third division of Professor Gray's writings ; while in the fourth fall 54 iv INTRODUCTION. a number of papers wliicli owe their existence to the discus- sions which followed the publication of Mr. Darwin's " Origin of Species," — discussions in which Professor Gray took in this country the foremost position. It is not proposed to republish the works of descriptive botany, although some of the early and most important of these memoirs are out of print and quite beyond the reach of the great mass of botanical students. The value of these papers, how^ever, is historical only, as all that they contain of permanent usefulness has already been incorporated in stand- ard -works upon the science, or wall be used in due time to lighten the burden of those upon whom has fallen the task of completing the " Flora of North America." There is even less reason for reprinting any of the earlier editions of the text-books. The last editions contain their author's latest views upon the science, and are still within the reach of stu- dents. Works of this character change necessarily as knowl- edge increases, and the value of every edition of a text-book, except the last, is merely historical. The philosophical essays, or the most important of them, which grew out of the discussion of the Darwinian theory, have already been republished by their author, and another republication of these papers is therefore not proposed at this time, although it is impossible, without having read them, to understand rightly Professor Gray's influence upon the intel- lectual movement of his time. There remain the reviews, the biographical notices, and a few essays upon subjects of general interest to botanists. They have long been out of print and have not been incor- porated in any recent publication. It was believed therefore that a reissue of these papers, or a selection from them, would be a useful contribution to botanical literature, and a proper tribute to the memory of their author ; and for these reasons these volumes have been prepared. Many of the reviews are filled with original and suggestive observations, and, taken together, furnish the best account of the development of botanical literature during the last fifty years that has yet been written. INTRODUCTION. v There can hardly be a question with regard to Professor Gray's value as a critic. His reviews represented the opinion of a just and discriminating mind, thoroughly familiar with all sides of the question before it, critical rather than lauda- tory, loving the truth and its investigators, but the truth above everything else. No other naturalist of his reputation and attainments ever devoted so much time to literary work of this sort, or continued it so uninterruptedly for so many years ; and in our time the criticism and advice of no other botanist has been so eagerly sought or so highly valued by his contem- poraries. The selection of the articles for republication has been an embarrassing and difficult task. The amount of material at my disposal has been overwhelming, and desirable as it might have been to republish it all, it has not been possible to do so within reasonable bounds. More than eleven hundred bib- liographical notices and longer reviews were published by Professor Gray in different periodicals ; and it was necessary in preparing these volumes to exclude a number of papers of nearly as great interest and value as those which are chosen. I have endeavored in making this selection to present, as far as it is possible to do so in a series of papers written indepen- dently of each other during a period of more than fifty years, a history of the growth of botanical science during a period which must remain one of its great eras — a period marked by the gradual change of ideas among naturalists ujDon the origin and fixity of the species which has broadened the field of all biological investigation ; by the establishment and sys- tematic arrangement of vast herbaria gathered from all parts of the world ; by the introduction of improved and more phil- osophical methods of investigation in the laboratory ; and by the growth of popular appreciation for the value of scientific training. I have tried, in making a selection of these articles, to display as far as possible the mental grasp of their author and his varied attainments in all departments of botany ; and to include the reviews of those works which Professor Gray himself believed had played in the two continents, during his vi INTRODUCTION. time, the most important part in elevating the science to which his whole life was devoted. The second volume of this series contains a few essays of general interest, and a selection of the biographical sketches of the principal botanists who have died in recent years. c. s. s. Brookline, April, 1889. CONTENTS. PAGE Introduction i Lindley's Natural System of Botany 1 De Candolle's Prodromus 15 Endlicher's Genera Plantarum . . . . , . . 33 Harvey's South African Plants 36 Siebold's Flora of Japan 37 Moquin-Tandon's Vegetable Monstrosities . . . . .40 Agassiz's Zoological Nomenclator . . . . . . . 41 Von Mohl's Vegetable Cell 51 Ward's Growth of Plants in Glazed Cases 59 Hooker and Thomson's Indian Flora 62 De Candolle's Geographie Botanique 67 Henfrey's Botany 72 Naudin on the Genus Cucurbita 83 Weddell's Monograph of Urticace^ 87 Radlkofer's Fecundation in the Vegetable Kingdom . . 91 Hooker on the Balanophore^ 94 boussingault on the influence of nitrates .... 100 Bentham's Hand-Book of the British Flora 104 Vilmorin's Improvement of Cultivated Plants .... 109 Engelmann on the Buffalo Grass 112 Curtis' s Trees of North Carolina 115 Bentham's Flora of Hong Kong 117 Hooker's Distribution of Arctic Plants 122 De Candolle on the Variation and Distribution of Species . 130 Hooker on Welwitschia 151 Darwin's Movements and Habits of Climbing Plants . . . 158 Watson's Botany of the 40th Parallel 180 Decaisne's Monograph, of the Genus Pyrus 186 Engelmann's Notes on the Genus Yucca 196 Ruskin's Proserpina 199 Emerson's Trees and Shrubs of Massachusetts .... 204 Darwin's Insectivorous Plants 206 viii CONTENTS. Naudin ox Heredity and Variability in Plants . . . .212 Darwin's Cross and Self-Fertilization in the Vegetable King- dom 217 Phytogamy 241 Bentham's Flora of Austraua 246 De Candolle's New Monographs 248 Epping Forest 253 Hooker and Ball's Tour in Marocco 255 Bentham on Euphorbiace^ 259 Henslow on the Self-Fertilization of Plants .... 263 Plant Archeology 269 Watson on North American Liliaceje 278 De Candolle's Phytography 282 Darwin's Power of Movement in Plants 304 De Candolle's Origin of Cultivated Plants 311 Bentham and Hooker's Genera Plantarum 355 Botanical Nomenclature 358 Ball's Flora of the Perl^v'ian Andes 384 REVIEWS. LINDLEY'S NATURAL SYSTEM OF BOTANY.i The cultivators of botany in this country are generally acquainted with the former edition of this work through the American reprint, edited by Dr. Torrey, and published by Messrs. Carvill of New York, in the spring of 1831. Dr. Lindley's treatise was, at the time of its appearance, the only introduction to the Natural System in the English language, if we except a translation of Achille Eichard's " Nouveaux Elemens de la Botanique," which was published about the same period. It is unnecessary to state that a treatise of this kind was greatly needed, or to allude either to the peculiar qualifications of the learned and industrious author for the accomplishment of the task, or the high estimation in which the work is held in Europe. But we may very properly offer our testimony respecting the great and highly favorable in- fluence which it has exerted upon the progress of botanical science in the United States. Great as the merits of the work undoubtedly are, we must nevertheless be excused from adopting the terms of extravagant and sometimes equivocal eulogy employed by a popular author, who gravely informs his readers that no book, since printed Bibles were first sold in Paris by Dr. Faustus, ever excited so much surprise and ^ A Natural System of Botany ; or a systematic view of the Organi- zation, Natural Affinities, and Geographical Distribution of the whole Vegetable Kingdom ; together with the uses of the most important spe- cies in Medicine, the Arts, and rural or domestic economy. By John Lindley. Second edition, with numerous additions and corrections, and a complete list of genera and their synonyms. Loudon : 1836. (Ameri- can Journal of Science and Arts, xxxii. 292.) 2 REVIEWS. wonder as did Dr. Torrey's edition of Lindley's " Introduc- tion to the Natural System of Botany." Now we can hardly believe that either the author or the American editor of the work referred to were ever in danger, as was honest Dr. Faustus, of being burned for witchcraft, neither do we find anything in its pages calculated to j^roduce such astonishing effects, except, perhaps, upon the minds of those botanists, if such they may be called, who had never dreamed of any important changes in the science since the appearance of good Dr. Turton's translation of the "Species Plantarum," and who speak of Jussieu as a writer who " has greatly im- proved upon the natural orders of Linnseus." ^ We have no hesitation, however, in expressing our conviction that no single work has had such a general and favorable influence upon the advancement of botanical science in this coun- try, as the American edition of Dr. Lindley's Introduc- tion to the Natural System. This treatise, however useful, 1 Dr. Lindley is quite right iu his remark that the chief difficulties the student has to encounter in the study of botany, upon the principles of the Natural System, have been very much exaggerated by persons who have written upon the subject without understanding it. To refer to a single instance. In the fifth edition of the Manual of Botany, by Mr. Eaton, an account of the Natural Orders of Jussieu is given, in which the genera Ambrosia and Xanthium are referred to Urticece ; and in a note it is added, " Some botanists place the last two genera in the order Corym- hiferce also iu the Linnsean class Syngenesla. I see no good reason for these innovations." Now Linnffius, in his artificial arrangement, certainly did place these genera (and also Parthenium and Iva) in Monoecia Pen- tamlria ; but the innovator in this instance is Jussieu himself, who never referred these two genera to Urticece^ but places them in his order Corym- hiferce (Compositce), where they truly belong. The descriptions of Natu- ral Orders in Eaton's Manual, purporting to be taken from Jussieu, bear a very remote resemblance indeed to the ordinal characters of the admi- rable Genera Plantarum of that author, while the occasional criticisms on its supposed errors afford the clearest proof that the work was not understood by the author alluded to. It should be recollected that, pre- viously to the reprint of Dr. Lindley's Introduction, Mr. Eaton's Manual was the only work professing to give a view of the Natural System within the reach of the great majority of the botanical students of his country, excepting, perhaps, the American edition of Smith's Grammar of Botany. LINDLEY'S NATURAL SYSTEM OF BOTANY. 3 was indeed not absolutely indispensable to the favored few, who, aided by the works of Jussieu, Brown, De Candolle, the elder and younger Richard, etc., were already successfully and honorably pursuing their studies and investigations ; but to the numerous cultivators of botany throughout the countr}^, who could seldom be expected to possess, or have access to, well-furnished libraries, and to whom the writings of these great luminaries of the science were mostly unknown except by name, this publication was a truly welcome acquisition, conferring advantages which those alone who have pursued their studies under such unfavorable circumstances can fully appreciate. A second and greatly improved edition of this work having appeared within the past year, it occurred to the writer of these remarks that a cursory notice of it might not be unac- ceptable to the readers of the " American Journal of Science." We do not intend, in these observations, to engage in a de- fence of what is called the Natural System of Botany, but take it for granted that the science can by no other method be successfully and philosophically pursued ; or, to employ the forcible language of Linnaeus, " Methodus naturalis pri- mus et ultimus finis botanices est et erit . . . Primum et ultimum in hoc botanicis desideratum est." The few per- sons who remain at this day unconvinced of its advantages are not likely to be affected by any arguments that we could adduce. A somewhat larger number may perhaps be found in this country who admit the importance and utility of the natural arrangement in the abstract, but decline to avail themselves of the advantages it affords in the study of plants, because, forsooth, it is too much trouble to acquire the en- larged views of vegetable structure which are necessary for the application of its principles. It would almost seem, from the views and practice of such botanists, that they considered it the chief object of a classification to afford the means of ascertaining the name of an unknown plant by the slightest examination of its structure, and with the least possible ex- penditure of thought. In the first edition, Dr. Lindley entered into some detailed 4 REVIEWS. explanations to show the fallacy of the common opmion that the artificial system of Linnaeus is easy, and the Natural Sys- tem difficult of application. The sentiments of the public have undergone so great a change upon this subject within the last five or six years that he finds it no longer necessary to adduce these considerations, and accordingly commences at once with a development of the principles on which the Natu- ral System is founded, namely, ''That the affinities of plants may be determined by a consideration of all the points of resemblance between their various parts, proj)erties, and qualities; that thence an arrangement may be deduced in which those species will be placed next each other which have the greatest degree of relationship; and that consequently the quality or structure of an imperfectly known plant may be determined by those of another which is well known. Hence arises its superiority over arbitrary or artificial systems, such as that of Linnaeus, in which there is no combination of ideas, but which are mere collections of isolated facts, not hav- ing any distinct relation to each other." (^Preface, p. vii.) We have never met with a more clear and succinct account of the principles upon which the primary divisions of the vegetable kingdom rest than that comprised in the follow- ing extract. Those acquainted with the first edition will per- ceive that the author has changed his opinions respecting the number of these primary divisions or classes; the Gymno- spermce^ or Flowering plants with naked ovules (comprising the Coniferce^ Cycadece^ and, according to Brongniart and Lindley, the Eqidsetacece)^ and the RMzantJim^ as ori- ginally established by Blume, being here admitted to the rank of independent classes. Their claim to this rank, how- ever, can as yet be hardly considered as fully established. " One of the first things that strikes an inquirer into the structure of plants is the singular fact that while all species are capable of propagating their race, the mode in which this important function is accomplished is essentially different in different cases. The great mass of plants produce flowers which are succeeded by fruits, containing seed, which is shed or scattered abroad, and grows into new individuals. But in LINDLEY'S NATURAL SYSTEM OF BOTANY. 5 Ferns, Mosses, Muslirooms, and the like, neither flowers, nor seeds properly so called, can be detected ; but propagation is effected by the dispersion of grains or spores which are usually generated in the substance of the plant, and seem to have little analogy with true seeds. Hence the vegetable world separates into two distinct groups, the Flowering and the Flowerless. Upon examining more closely into the respec- tive peculiarities of these two groups, it is found that flow- ering plants have sexes, while flowerless plants have none ; hence the former are called Sexual and the latter Asex- ual. Then again the former usually possess a highly devel- oped system of spiral or other vessels, while the latter are either altogether destitute of them, or have them only in the highest orders, and then in a peculiar state : for this reason flowering plants are also called Vascular, and flowerless Cellular. More than this, all flowering plants when they form stems, increase by an extention of their ends and a dis- tention or enlargement of their sides ; but flowerless plants appear to form their stems simply by the addition of new mat- ter to their points ; for this reason, while the former are prin- cipally Exogens or Endogens, the latter are called Acrogens. Flowering plants are also for the most part furnished with respiratory organs or stomates, while flowerless plants are to a great extent destitute of them. No one then can doubt that in the vegetable kingdom, two most essentially distinct divi- sions exist, the Flowering and the Flowerless, and that these differ not in one circumstance only, but are most essentially unlike in many points both of organization and physiology. " In like manner. Flowering plants are themselves divi- sible into equally well-marked groups. Some of them grow by the addition of the new woody matter to the outside of their stem beneath the bark ; these are Exogens : others grow by the addition of new woody matter to the inside of their stem near the centre ; these are Endogens. But Exogens have two or more cotyledons to their embryo, and hence are called Dicotyledons ; while Endogens have only one cotyle- don, and are, therefore. Monocotyledons. Exogens have the young external wood connected with the centre by medullary 6 REVIEWS. processes; Endogens, having no occasion for sucli a provi- sion, are destitute of it. In Exogens the leaves have their veins disposed in a netted manner ; in Endogens the veins run parallel with each other. The number of parts in the flower of an Exogen is usually five or its multiples ; in an En- dogen it is usually three or its multiples. In germination the young root of Exogens is a mere extension of the radicle ; but of Endogens it is protruded from within the radicle ; hence the former have been named Exorliizce^ and the latter Endo- rJiizoi. In this case, then, as in the last, we have two groups differing entirely from each other in their germination, the structure of their stem and leaves, their mode of growth, the arrangements of the parts of the flower, and in the or- ganization of their embryo. It is impossible, therefore, not to recognize such groups also as natural. " To this separation of the vegetable kingdom into Ex- ogens, Endogens, and Acrogens, or by whatever sjaionymous names these groups may be known, many botanists confine themselves. But there are two others, of subordinate im- portance, perhaps, but nevertheless characterized by circum- stances of a similar nature, and therefore, I think, to be esteemed of equal dignity with them. In true Exogens and Endogens, the fertilizing principle is communicated to the young seeds through the medium of a stigma which termi- nates a case or pericarp in which they are inclosed. But in some plants otherwise Exogens, the fertilizing principle of the pollen is applied immediately to the seeds, without the intervention of any pericarpial apparatus, and they bear the same relation to other Exogens as frogs and similar reptiles to other animals. These plants, therefore, are separated as a distinct class, under the name of Gymnosperms. Like the other groups of the same grade, these are also found to pos- sess peculiarities of a subordinate nature. For instance, they have in many cases more cotjdedons than two, whence they have been called Polycotyledons ; their radicle usually adheres to the albumen in w^hich the embryo lies, and that circumstance has given rise to the name Synorhizce. The veins of their leaves, when they have any veins, are either LINDLEY'S NATURAL SYSTEM OF BOTANY. 1 simple or forked ; in which respect they approach Endogens on the one hand, and Acrogens on the other. And finally, their vascular system is very imperfect compared with that of other Exogens of an equal degree of development. " The other group, called RMzanthce^ is far less correctly known, but it seems to stand as it were between Endogens and Acrogens of the lowest grade ; agreeing with the latter in the absence or very imperfect state of the vascular system, in a general resemblance to Fungi, and in the ap^mrent seeds being mere masses of sporules ; but apparently according with Endogens in the ternary number of their floral enve- lopes, and in the presence of fully developed sexes. " Certainly there is no possibility of obtaining such im- portant primary groups as these by any kind of artificial con- trivance." (^Preface., pp. x.-xii.) The grand natural divisions of the natural kingdom are, therefore, perfectly obvious, and may be very clearly defined. With our present knowledge of vegetable structure no great difficulty is experienced in characterizing the orders of natu- ral families, and all subordinate groups. The great desi- deratum has ever been to effect such an arrangement of the orders under the primary classes that each family should be placed next to those which it most nearly resembles. This might easily be accomplished, if the idea once so strongly in- sisted upon by poets and metaphysicians, of a chain of beings, a regular gradation, by a single series, from the most perfect and complicated to the most simple forms of existence, had any foundation in truth. On the contrary, nothing is more evident than that almost every order, or other group, is allied not merely to one or two, but often to several others, which are sometimes widely separate from each other ; and, indeed, these several points of resemblance, or affinity, are oc- casionally of about equal importance. A truly natural lineal arrangement is therefore impracticable, since by it only one or two out of several points of agreement can be indicated. As this method is, however, the only one that can be followed in books, all that can be done is to arrange the orders in such a manner as to offer the least possible interruption to their 8 REVIEWS. natural affinities. The number of orders is so large that practical convenience seems to require their arrangement into groups subordinate to the primary classes ; and when man- ifestly natural assemblages cannot be recognized, we are obliged to employ those which, being less strongly marked, and distinguished by a smaller number of characters, are apparently of a more artificial nature. The arrangement em- ployed by the learned Jussieu, in his celebrated "Genera Plantarum," although to a considerable extent artificial, has been almost universally adopted, until within the last few years. In this method Dicotyledonous plants are primarily divided into three groups : the first including those with a polype- talous corolla ; the second, those with a monopetalous corolla ; and the third, those destitute of a corolla. These sections are subdivided (as also the Monocotyledons) by means of char- acters taken from the insertion of the stamens (or corolla), whether hypogynous, perigynous, or epigynous. The arrange- ment here pursued, which is too well known to require further notice, is substantially adopted by De Candolle, the difference being more in appearance than in reality. Dr. Lindley dis- carded these subdivisions in the first edition of his work ; but the new distribution of the orders therein proposed possesses few advantages, and indeed seems not to have satisfied the author himself. In the same year with the publication of the work just mentioned, the " Ordines Plantarum " of Bartling appeared, in which a more natural arrangement of the orders is attempted by the formation of aggregate or compound or- ders, as originally proposed, and in several instances success- fully accomplished, by Robert Brown. An analogous plan was pursued by Agardh in his *' Aphorismi Botanici " (1817), and again in his " Classes Plantarum " (1825) ; but these at- tempts, however ingenious, do not seem to have obviated, in any considerable degree, the inconveniences of lineal arrange- ment. We now return to our author, whose views upon this sub- ject have been materially modified since the original publica^ tion of his Introduction of the Natural System. The method LIN D LEY'S NATURAL SYSTEM OF BOTANY, 9 now employed was first sketched in the " Nixus Plantarum " (1832), and afterwards in the " Key to Structural, Physio- logical, and Systematic Botany"^ (1835), and is more fully developed and illustrated in the work before us. He now admits, as we have already seen, five primary classes, two of which, however, are much smaller than the others and of sub- ordinate importance, and may be considered as transition classes, namely, Gymnospermce^ which connect Exogens with the higher Acrogens, and Rhizanthce, which form the transi- tion from Endogens to Acrogens of the lowest grade. The great class Exogenm (^Dicotyledones of Jussieu) is divided into three subclasses, namely : — 1. Polyioetalm ; those with the floral envelopes consisting of both calyx and corolla, the latter couiposed of distinct petals. 2. Mo7iopetal(B ; those with the petals combined in a mono- petalous corolla. 3. Incom.pl etce ; those always destitute of a corolla, the calyx also often incomplete or absent. Thus far this mode of subdivision is nearly the same with that of Jussieu ; Dr. Lindley, however, neglecting altogether the character afforded by the insertion of the stamens, divides the polypetalous orders into seven, and the Monopetalm and Incompletce each into five sections or groups. As a specimen of this plan, we copy the names of the groups of the first sub- class, with their synoptical characters. 1. Albuminosce. Embryo very considerably shorter and smaller than the albumen. 2. Uj^igy 1710 see. Ovary inferior, usually having an epigy- nous disk. 3. Parietosce. Placentation parietal. 4. Calycosce. Calyx incompletely whorled ; two of the sepals being exterior. 5. SyncaiposcB. None of the characters of the other groups, and with the carpels compactly united. ^ This excellent little work consists of an augmented edition of the author's Outlines of the First Principles of Botany, with a revised trans- lation of the Nixus Plantarum. 10 REVIEWS. 6. Gynohaseosce. Carpels not exceeding five, diverging at tlie base, arranged in a single row around an elevated axis or gynobase. Stamens usually separate from the calyx. 7. Apocarposm. None of the characters of the other groups, but with the carpels distinct, or sej)arable by their faces, or solitary. Next, every group is divided into smaller groups, each of which includes one, two, or several orders. These minor groups are called Alliances.^ and are distinguished by the termination ales. Thus, under the Albuminose group, we have : — Alliance 1. Banales., comprising the Hcmunculacece., Pa- paver acece (with its suborder, as Lindley, following Bern- hardi, considers it, Fumariece)^ Nymphceacem^ (to which Hydropeltidem is improperly joined), and Nelumhlacece ; Alliance 2. Anonales, which comprehends the Nutmeg tribe, the Ajionacece, Magnoliacece, etc. ; Alliance 3. Umhellales, including the Umbelliferous tribe, with the nearly allied Araliacece ; Alliance 4. Grossales^ consisting chiefly of the Grossu- lacece or currant tribe ; and lastly. Alliance 5. Pittosptorales^ which strikes us as a singularly heterogeneous assemblage, bringing together into one group the Vitaceo2^ Pittosporacece^ Olacacece, Francoacem., and Sar- raceniacem. All the subclasses and groups, both of Exogens and Endo- gens, are subdivided in a similar manner ; but we cannot here proceed further with our enumeration. It will be borne in mind that the chief object of an arrangement of this kind is to facilitate the study of the natural orders, by dividing the extensive primary classes into sections of convenient size, and to dispose these groups, and the orders they comprise, as nearly in accordance with their respective affinities and rela- tionships as a lineal arrangement will allow. It is impossible, in the present state of our knowledge, to say how far the views of our author will ultimately be approved. Every at- tempt of the kind must necessarily be very imperfect, so long as the structure of only a limited portion of the whole vegeta- ble kingdom has been attentively and completely examined ; LINDLEY'S NATURAL SYSTEM OF BOTANY. 11 and the author is well aware " that this part of the work will require many great changes and improvements before it can be considered at all established." Notwithstanding the ob- jections to which it is liable in many particulars, we agree with the author in the opinion, " that even in its present state it will be found to be attended with many advantages, and that every step which may be taken in determining the limits of the natural groups subordinate to the primary classes must be a decided gain to the science. So rapid is the advance of our knowledge of the vegetable kingdom, and so numerous are the new types of structure that present themselves to the sys- tematic botanist, that it is to be feared lest another chaos should be brought on by the masses of imperfectly grouped species with which the science will soon abound." The names of natural orders, as first established, do not appear to have been framed in accordance with any uniform rule, as to derivation or mode of termination. They were sometimes intended to express some characteristic feature (Ex. Leguminosce^ Lahiatce^ CriLciferoe^ Umhellifermy Coni- ferce^ etc.), but more commonly some genus was selected as the type of the family, which was designated either by the plural of the genus simply (as Myrti^ Lilia^ Irides^ Uiq^Jiorbice), or with a slight prolongation (as Orchidece, JasminecB^ etc.), or with the termination still further modified (as in Cyper- oidem, Aroidece^ Boraginece^ or RannnGulacece^ Rosacece^ Cucurbit acece^ etc.). The derivation of the name of the order from some prominent genus is now the universal practice ; and for the sake of uniformity as well as to distinguish such names from those of genera in the plural number, the termina- tion acexE is given to orders, and that of ece to suborders, etc. The advantages of uniformity in this respect is manifest, and Dr. Lindley therefore insists upon the adoption of the rule in all cases. In the " Key to Botany," published the year pre- vious to the appearance of the second edition of the present work, the termination in acece is employed, not only in the names of orders formed from those of genera, but also in the few still in use which relate to some peculiarity in the habit of the family. Thus instead of Cruciferce^ Umhelliferoe^ 12 REVIEWS. Conlfero3, etc., we have Criiciacece, Umhellacecej and Cona- cece. These are, however, very properly abandoned in the work before us, in which the author inclines to give up the old and familiar names of these orders, and to substitute those formed in the customary manner from well-known genera. Brassicacece, Apiacece, and Pinacece may certainly be as good names as any other when we once get accustomed to them, but it seems hardly necessary to make any change in names of this kind. Dr. Lindley, as we have already seen, gives to the names of Alliances the termination ales^ and to groups that of osce. The chief advantage of this system is, that the name of any group at once indicates its rank and importance. The value of this work is greatly increased by the complete list of genera (so far as known at the time of publication), with the principal synonyms, appended to each order and properly arranged under their several suborders, sections, etc. This laborious and difficult task is upon the whole very faith- fully executed. We observe, however, several errors, typo- graphical and otherwise, which are not noticed in the appen- dix ; and in a few instances the same genus is referred to two different orders. The whole catalogue will doubtless be ren- dered more perfectly accurate in a future edition. The whole number of genera comprised in this enumera- tion, exclusive of synonyms, is 7840. Sprengel's " Systema Yegetabilium," which was finished in 1827, contains (exclu- sive of the appendix) only 3598 genera, or not quite half the number now known ; while the twelfth edition of " Systema Naturae " (the last of Linnseus himself) comprises 1228 genera, or only about a third more than are now known in a single family. This great and rapid increase is perhaps chiefly owing to the discovery of new plants ; but it is also attributable in a good degree to the more accurate knowledge of those already known. In either case, it is the natural result of the progress of discovery ; and instead of embarrassing the student, as is often supposed, does in reality render the study of the science much more clear and satisfactory. Notwithstanding the great increase of genera within the last few years, it may LINDLEY'S NATURAL SYSTEM OF BOTANY. 13 be safely said that at no previous period could a really useful knowledge of the vegetable kingdom be acquired with so lit- tle labor. In hazarding this remark, it is of course taken for ofranted that the student will avail himself of all the advan- tages of modern physiological botany and of the natural sys- tem : for so rapid has been the discovery of new and strange forms of structure, for which the artificial arrangement of Linnseus makes no provision, that the student who takes that system as his guide has indeed a hopeless task before him. The essential characters of the orders appear to have been very carefully revised in this edition, as also the remarks upon their affinities, geographical distribution and sensible properties. Did our limits allow, we might call the attention of our readers more particularly to this part of the work. We cannot bring our remarks to a close, however, without sug- gesting what we consider a very desirable improvement upon the manner in which the seed is described, not only in this, but in almost all modern systematic works. It is very necessary that an organ which affords such important characters, both as to its situation in the fruit, and particularly as to its inter- nal structure, should be described wdth the greatest possible clearness and precision, and in a uniform manner. The prevalent fault of which we complain is thus noticed, as long ago as the year 1811, by that most acute botanist, the late L. C. Richard. " Cffisalpinus, Adanson, Jussieu, and Gaertner, always take into view the direction of the embryo relative to the pericarp merely. This method appears to me improper : first, because it does not indicate with precision that direction which is most important to be understood ; secondly, because the pericarpic direction of the embryo is often difficult to be ascertained, and is sometimes variable or even wholly different in the seeds of the same fruit. I have already shown by numerous examples in my ' Analyse du Fruit,' that the best method is to indicate the direction of the seed relative to the pericarp, and of the embryo relative to the seed." In very many descriptions, the direction of the embryo rel- ative to the seed can only be inferred from the pericarpic 14 REVIEWS. direction, or, wliicli is still more objectionable, the same struc- ture is described by very different language in different in- stances, thus rendering unnecessarily complicated an investi- gation which of itself is not usually difficult. We may adduce as an example the five orders comprised in the alliance i?a- nales., which stands at the commencement of Dr. Lindley's treatise. We have no means of ascertaining, from the essen- tial character of any one of these orders, either the spermic direction and position of the embryo, or the situation of the chalaza and micropyle relative to the hilum, from which the former may be inferred. It is commonly stated that the em- bryo is situated at the base of the albumen ; but it is not specified whether the radicle is next the hilum (as in Pa- paveracece^ JSfympTiceacece^ etc.), or points in the opposite di- rection (as in Nelumbiacece and Cahomhaceoe) ; a matter of essential importance, since the seeds result in the one case from the ripening of anatropous, and in the other of orthotro- pous, ovules. The students of botany in this country are greatly indebted to the learned editor and the enterprising publishers of the first American edition of this work. May we hope to have our obligations increased by the reprint of this greatly im- proved edition ? DE CANDOLLE'S PRODROMUS. The second part of the seventh volume of De Candolle's " Prodromus " ^ — with which our notices may appropriately commence — was published at the very close of last year, and comprises the following orders, namely, Stylideoe^ Loheliaceoe^ Campanulacece^ CyiMacem (a very small order, founded on the Cape genus Cyphia, and here first proposed by Alphonse De CandoUe), Goodenoviece^ Gesneriacece^ Napoleonece^ Vac- cinece, £Jricacece, Epacridece^ Pyrolacece^ Francoacece^ and Monotropem. Of these, the Loheliacece^ Campanulacece^ and CypMacece were elaborated by Professor Alphonse De Can- doUe, the well-known son of the distinguished author ; the Vac- cinea3 by Professor Dunal of Montpelier ; and the tribe Ericece (the Heath-tribe) was prepared by Mr. Bentham. It will be observed that De Candolle has disposed the Ericacem nearly in the manner first proposed in the " Theorie Elementaire," considering the Vacciniece, Monotropece, Pyrolacece^ etc., as so many distinct families ; a view, however, which will not probably be ultimately adopted. Among the uncertain or little known Ericaceous plants, De Candolle has introduced the genus Pickeringia of Nuttall (which was founded upon Cyrilla paniGulata of the same author, published in the fifth volume of this Journal) ; this, however, has been long since ascertained to be a species of Ardisia, which belongs to a very different order ; and Mr. Nuttall has accordingly recently dedi- 1 De Candolle. Prodromus Systematis Naturalis Regni Vegetahilis, etc. Pars VII., sectio ii. Paris, 1839 (American Journal of Science and Arts, xxxix. 168). — Pars VIII. Paris, 1844 (Ibid., xlvii. 198). — Pars IX. Paris, 1845 (Ibid., 2 ser., i. 174). — Pars XI. Paris, 1847 (Ibid., 2 ser., v. 449). — Pars XII. Paris, 1848 (Ibid., 2 ser., vii. 309). — Pars XIII., sectio ii. Paris, 1849 (Ibid., 2 ser., viii. 300). — Pars XIV., sectio ii. Paris, 1857 (Ibid., 2 ser., xxv. 290). — Pars XVI., sectio i. Paris, 1870 (Ibid., 2 ser., xlvii. 125). — Pars XVII. etc. Paris, 1873 (Ibid., 3 ser., vii. 66). 16 REVIEWS. cated to Dr. Pickering a curious Leguminous plant from Cali- fornia. The genus Galax, De Candolle has appended to Py- rolacece (tribe Galacea?^^ a view which seems to be confirmed by an unpublished plant from the mountains of North Caro- lina, which, in compliment to an assiduous and well-known American botanist, will bear the name of Shortia galacifolia. The prior portion of the seventh volume (published in 1838), as well as the whole of the fifth (1836) and sixth (1837), is exclusively devoted to the immense family of the Compositce (the class Syngenesia of Linnaeus), which fills more than 1700 closely printed pages, the immediate prepa- ration of which occupied the indefatigable author for seven years ! We may take this family as a fair example of the increase in the number of known species within the last eighty years. The whole number of Syngenesious plants described by Linnaeus in the first edition of the "Species Plantarum " (published in 1753), including the few Compo- sitce referred to other classes, is 555, which is about 150 less than the now described species of the single genus Senecio. We have not time nor space to enumerate the species of the order in succeeding systematic works, so as to show the pro- gressive increase. Suffice it to say that the whole number known to Linnaeus and published during his lifetime cannot exceed 800 species, while the number described by De Can- dolle is in round numbers about 8700, which are disposed in 893 genera. If to these we were to add the species which have been since published, or are being published in works now in progress, and also the very numerous unpublished spe- cies which exist in all large collections, making at the same time reasonable allowance for nominal species, the number of Compositce at present known would scarcely fall short of 10,000, which considerably exceeds the whole number of both flowering and flowerless plants described by Linnaeus or his contemporaries. Of the 8700 species given by De Can- dolle, more than 3000 are described for the first time in this work. In the general disposition of the order, the clear and simple classification of Lessing is to a great degree adopted. It is first divided into three great series, namely : — DE CANDOLLE'S PRODROMUS. 17 1. TuBULiFLOR^ ; those with the perfect flowers tubular and regularly five (or rarely four) toothed. 2. Labiatiflor^ ; those with bilabiate, or two-cleft, per- fect flowers. 3. LiGULiFLOR^ ; which have all the flowers strap-shaped. The first series includes about four fifths of the whole fam- ily, which are arranged in five tribes, namely, Vermoniacece^ Eiqoatoriacece^ Asteroidece^ Senecionideoi and Cynarece. The second series consists exclusively of the Mutisiacece and the JV^assauviacece, chiefly South American plants ; a single spe- cies of Chaptalia is, we believe, the only North American representative. The third series, comprising the Chicoracece^ so readily known by their milky juice, and by having all their florets ligulate, contains many North American repre- sentatives. So many orders or separate genera of Monopetalous plants have been the subject of recent monographs, and so much valuable assistance is also engaged for the ensuing portions of the " Prodromus," that several volumes may be expected at no very distant period. It may not be improper to state that Mr. Boissier of Geneva is engaged in the preparation of the Plu7nhaghiece ; Mr. Duby of Geneva will prepare the PAmidacecE ; Professor Dunal of Montpelier, the Solancce ; Mr. Decaisne of Paris, the Asclepiadece ; and Mr. Bentham, the ScrophidarinecB and Lahiatm, We can at length announce the publication of this im- portant volume (viii.) ; the first of the series under the editor- ship of the son of the great Genevan botanist,^ and which he has appropriately dedicated to the memory of his illustrious father. We are glad to state, that arrangements have been made to expedite the publication of the succeeding volumes. The printing of the ninth, it is said, has already commenced, and its appearance may be expected in the autumn of the present year. It will contain the Loganiacece^ Bignoniaceoe^ Cyrtandracem^ Sesamem^ and Borraginece^ from the notes prepared by the late Professor De Candolle ; the HydropJiyl- lacece, by Alphonse De Candolle ; the Gentianacece, by Grise- 1 The elder De Candolle died September 9, 1841. 18 REVIEWS. bach; the Polemojiiacece, by Bentham, and the Convolvula- cece, by Choisy. The tenth vohmie will be occupied with the Solanacece, by Dunal, and the Scrop/ndarinece, by Bentham ; in the elaboration of which orders, these two distinguished botanists are now actively engaged. The first order in the volume before us, the Lentihulariece^ is prepared by the edi- tor. The North American species of Utricularia are dis- tributed into three sections, namely: 1. Megacista, where the verticillate foliage is floated by inflated petioles ; 2. Len- tibularia, where the capillary segments of the submersed foliage are utriculiferous ; and 3. Oligocista, where the leaves are few, undivided, and disappear after flowering ; the roots strike into the soil or mud, and generally bear the utri- culi, when these are present. U. resujnnata, discovered by B. D. Greene, Esq., and first mentioned in the Massachusetts Catalogue of Plants and Animals, 1835, is wrongly placed by De CandoUe among the yellow-flowered species of the second section. It has purple flowers and should stand next U. 'purpurea. The name U. Greenei^ Oaks, in " Hovey's Magazine " must stand as a synonym, as there is no good reason for changing the prior name imposed by the discoverer. The order of Primulacece is elaborated by M. Duby of Geneva, who follows Endlicher in the general distribution of the family. From some inadvertence, Glaux maritima is not cited as an American plant. Naumhurgia thyrsijlora, McGuch, = Lysimachia tliyrsiflora, Linn., and L. cajntata, Pursh. The L, revoluta, Nutt., is referred to L. longifolia, Pursh. The common Samolus, of the southwestern United States, which has smaller flowers than the true S. Valerandi, is referred to S, Jlorihundus, HBK. S. ehracteatus is not noticed as a plant of the United States, although it is com- mon along our southern borders, nor is it disting-uished even as a subgenus, although, on account of its nearly free ovary and want of sterile filaments, a recent writer (M. Baudo, in Ann. Sci. Nat., Dec. 1843) has separated it, to form his genus Samodia. In the Myrsinacece, elaborated by the editor, we meet with two North American species, both natives of Florida, namely, Myrsine Floridana^ A. DC., DE CANDOLLE'S PRODROMUS. 19 and Ardisia Picheringia^ Torr. and Gr. To the small order Theophrastacece^ A. DC, our author has joined Jacquinia, a West Indian genus, one species of which extends into Florida. In the order Sapotacece^ the editor has proposed one new North American species of Bumelia. In Ebenacece we have only our Persimmon. From this the order Styraceoe (embracing Symplocinece and Halesiacece of Don) is dis- tinguished chiefly by the position of the cells of the ovary opposite the lobes of the calyx. Hopea is kept as a mere section of Symplocos ; including a dozen Asiatic species as well as our S. tinctoria. The order Oleacem is published from the manuscripts of the late Professor De Candolle. The American species of Fraxinus still require the labors of a monogra^^her. The order Jasminece is made to comprehend Bolivaria (of which there is at least one Texan species) and Menodora ; and the family Bolivariacem is shown to have been founded upon misconceived characters. For the elaboration of the Apocyjiacece, we are indebted to the younger De Candolle. The only North American genera are Amsonia (is not Echites Fraseri^ Roem. and Schultes, the A. ciliata, Walt. ?), Apocynum and Forsteronia (i^. dif- formis, J)C.,^ Echites difformis, Walt.). The order Asclepiadece has been very faithfully studied by Decaisne. All the North American representatives belong to the tribe of true Asclepiadece, with the exception of Gonolobus, of which we have several species (one of them, collected by Dr. Short, forms the new G. tilicef Gnus'), and one, or possibly two species of Chthamalia, Decaisne. 3fe- tastelma Fraseri is probably a native of the West Indies, not of Carolina. Enslenia albida, we notice, is about to be figured in the forthcoming volume of Delessert's Icones ; as also is Podostigma. Acerates includes ten, chiefly North American species. Asclepias is reduced to forty-four species, all of which are American, and the greater part extra-tropi- cal. We are happy to learn that the plates of the fifth volume of the " Icones Selectse " of the liberal Delessert — chiefly devoted to the illustration of the eighth volume of the " Pro- dromus " — are already in the hands of the engraver. 20 REVIEWS. A year ago we had the pleasure to notice the eighth volume of this indispensable work, the 1st of the series under the editorship of Professor Alphonse De Candolle. The ninth volume, now before us, was issued on the 1st of January last ; and the forthcoming portions are in course of prepa- ration under such favorable circumstances that we may now confidently look for the appearance of a volume a year, and for the full completion of this " Species Plantarum," according to the natural system, at no very distant period. We have already mentioned the arrangements that are made to secure this desirable consummation, and by which the work becomes as it were a series of separate monographs, prepared by the most skilful hands, under the superintendence of a common editor. Every botanist is aware of the improvement of the successive volumes as they appeared from the unrivalled hands of the elder De Candolle ; and a further improvement is mani- fest in the later portions, elaborated or revised by his son, es23ecially in the introduction of characters drawn from aesti- vation, placentation, the structure of the ovule, and other points which have only quite recently been turned to special account by systematic botanists. A particular account of a volume which is or soon will be in the hands of every working botanist, cannot be necessary, and we have not time at present for special enumeration. The ninth volume commences with the Loganiacece, by Alphonse De Candolle. The genus Coelostylis, Torr. and Gr., is correctly reduced to Spigelia. Under this order we have a tribe created for the long- vexed Gelsemium, which we suspect is not yet finally at rest. Next follows the Gentianacece, elaborated by Grisebach, whose recent monograph of that family, which forms the basis of the present arrangement, was duly noticed in this Journal. The order Bignoniacece is edited from the manuscripts of the elder De Candolle ; as are also the orders Sesamece and Cyr- tandracece, which last has been reduced by Mr. Brown to Ges- neriacece. The Hydrophyllaceoe are elaborated by Alphonse De Candolle, in which, by attributing generic importance to the presence or absence of the appendages or nectariferous scales within the tube of the corolla, the number of genera is DE CANDOLLE'S PRODROMUS. 21 perhaps too greatly increased. The Polemoniacece are admi- rably worked out by Bentham, who has reduced to sections of Gilia his Hugelia, Fenzlia, Linanthus, Dactylophyllum, Lep- tosiphon, Leptodactylon, and the Ipomopsis, Michx. The elaboration of the Convolvulacece by Professor Choisy does not appear to give entire satisfaction to botanists. The term " infelicissime intricatus " is perhaps still applicable to their family ; and the genera are probably unduly increased in number. Of the Borragiiiece, printed from the elder De Candolle's manuscripts, with valuable notes and additions by the editor, we have the first three tribes, namely, Cordiece, Ehretiea^^ and Heliotropece, But for the true Borragece we must wait until the appearance of the tenth volume, which is already in press. The long-expected eleventh volume of this work was pub- lished at the close of November last, or rather in December, and is now in the hands of botanists throughout Europe and America. It comprises, first, the Orohanchacece, by Mr. Renter of Geneva ; the true genera of which are Epiphegus of the United States ; Phelipsea, containing thirty-eight sjiecies of the Old World, and two of North America west of the Mississippi; Conopholis (for the Orohanclie Americana^ L.) ; Orobanche of over ninety species belonging to the temperate regions of the Old World ; Boschniakia of two liigli north- west American and Siberian species ; Clandestina of one south European species ; Lathraea of two Old World species ; and Anoplanthus, Endl. (Anoplon, Wallr.), which is made to embrace not only the three genuine species of North America, but also a separate section corresponding with Anblatum, Endl., of two sjDccies of Central Asia. The earlier recent name of this last genus is Gymnocaulis, Nutt. ; the next is Anoplon, by Wallroth. This was changed by Endlicher to Anoplanthus, because there is an Anoplon in zoology — an insufficient reason — with which, however, we need not trouble ourselves, since all three must give way to the early and ex- cellent name of Aphyllon, founded by Mitchell, and published in 1748, on the typical species afterwards called Orobanche uniflora by Linnaeus. This name, the revival of which is 22 REVIEWS. demanded not only by statute law, but also for the avoidance of the intricately conflicting names recently imi^osed, was adopted in the MSS. for this family prej^ared for the " Flora of North America," and also in the writer's " Manual of the Botany of the Northern United States." Besides A. uni- Jioru7n^ we have in the AYest and North, A. comosiim, and A. fasciculatwn. To the Genera affinia vel dubia, Reuter ap- pends Obolaria, which is correctly described, except that the insertion of the ovules over the whole inner surface of the ovary is overlooked ; the present writer's illustration of this genus not having reached Geneva until after this family was printed off. The great family of this volume, occupying almost four hundred pages, is the order Acanthacece, which is contributed by Nees von Esenbeck. This chiefly tropical or subtropical family, founded less than forty years ago upon a small number of genera and species, now ranks among the largest of the monopetalous series, and is arranged by Nees under two sub- orders, eleven tribes, and 146 genera (including those of the appendix). We have few Acanthacece in the United States, so that an analysis of the family would not interest our readers. We are pleased to find that the writer's reference of the RuelUa justiciceflora^ Hook, (the Eberlea of Riddell), to the genus Hygrophila, R. Br. (vide PI. Lindheim, p. 22, note), is confirmed by Nees von Esenbeck. It appears that the species is also Mexican, and had been already described by Schlechtendal. Our Ruellige belong to Dipteracanthus and Calophanes. Our Diantliera Americana^ L., with its allies, is included in the large genus Rhytigiossa, established by Nees (in Lindl. Introd. Nat. Syst., ed. 2) a few years ago, on some Cape of Good Hope species, from which the author suggests that the American species may differ gen- erically. However that may be, w^e urgently protest against this suppression of the old Gronovian and Linnsean genus, Dianthera, which was founded upon our species ; and which name, although unaccountably overlooked by Endlicher, who is usually careful, as well as by Nees, who is careless, as to questions of priority, must surely be continued for the genus, DE CANDOLLE'S PRODROMUS. 23 however bounded, which includes the Justicia pedunculosa and J. humilis of Michaux. The remainder of the volume, contributed by Schauer of Breslau, comprises, first, Phrymacece^ another of these incon- venient little orders established on a single genus of a single species, which, however, differs remarkably from Verhenacem as well as Lahiatcje^ by the monomerous ovary, with a single orthotropous ovule erect from the base of the cell, an embryo with the radicle superior, and convolute cotyledons ; and finally, the Verhenacece, composed of three tribes (the Ver- henece^ Vlticece, Avicenniece)^ ten subtribes, and forty-two gen- era. The genus Verbena, with which we are principally con- cerned in the United States, comprises seventy-one species, besides a dozen dubious ones appended at the close. We had nearly forgotten the small family Myoporacece^ which is elaborated by Professor Alphonse De Candolle him- self. These twelve genera are all Australian or Oceanic, except Bontia, which is a Caribbean genus of a single species. The twelfth volume, which will contain the Labiatm by Bentham, is promised for June. The twelfth volume of the " Prodromus," delayed somewhat by the convulsions of the continent, will be welcome to bot- anists. It concludes the series of 3Ionopetalce^ with the ex- ception of the small family Plantaginacem^ and the large one of Solanacece^ which last, Professor Dunal — from whom it has long been due — appears to find by no means easy to elab- orate. Of the present volume, all but one hundred pages are occupied by the Lahiatce^ from the hand of the most unwearied and best of monographers, Mr. Bentham. The plants of this large order, arranged in 101 genera, are thrown into eight tribes, instead of the eleven in the " Labiatarum Genera et Species ; " the Satureiece now being made to comprehend the Mentlioidem and the Melissinece of the earlier work, and the Scutellarinece being merged in the Stachydece. The Ocimoi- dece.) comprising nineteen genera, are represented in extra- tropical North America solely by one or two species of Hyptis, which inhabit our southeastern frontiers. Of the SatureiecB we have Mentha, Lycopus, a single Cu- 24 REVIEWS. nila, Pycnanthemiim (the whole seventeen species), an anom- alous Satureia (^S. riglda, Bartr.), three species of INIicrome- ria ; while to Calamintha is now referred the former 31. gla- bella and 31. Nuttallii^ as well as the Gardoqida Hookevi^ Benth., with the new C. canescens^ Torr and Gr., MSS., and C. Caroliniaiia^ Sweet (Thymus, Michx.) ; also two species of Dicerandra (of which D. densijlora is a new one from Florida) ; the California Pogogyne ; Hedeoma, including H. ciliata QKeithia ciliata, Benth., Lab.) ; and Collinsonia, of which six species are recognized. Of the tribe 3Io7iardece we possess a small portion of the vast genus Salvia, which is increased to 407 species ; Audi- bertia of California, of six species ; Monarda, of six species (J/, did y ma and 31. jistidosa being retained nearly as in the earlier monograph) ; and Blephilia, of two species. The tribe Nepetece affords us Lophanthus ; a single Dra- cocephalum ; and a Cedronella. The tribe StacJiydece furnishes Prunella, for which Al- phonse De Candolle restores, with satisfactory reasons given, the orthography, Brunella ; Scutellaria, of eighty-six species, sixteen of them North American ; Physostegia, of which two species are admitted ; Brazoria, Engelm. — Gray (in which the synonym '' Physostegia tnmcata, Hook. Botanical Maga- zine, t. 3494," should be cited under B. scutellarioides, as shown in Chlor. Bor.-Amer., and not under B. truncata) ; Macbridea and Synandra, each of a single species ; and Stachys (from which Betonica is now excluded), of 168 species. The tribe Prasicc belongs entirely to the Old World and to the Sandwich Islands ; and the tribe Prostantlierece is exclu- sively Australian. Of the Ajugece we have Isanthus, Trichostemma (in which a corrected view is taken of the inflorescence of the typical section), and Teucrium. The small order, Selaginacece, contributed by Professor Choisy of Geneva, consists of eight almost exclusively South African genera, to which the sub-arctic genus Gj^mnandra is doubtfully appended. Of three Siberian species of this genus T 4"K-r»«-rff DE CANDGLLE'S PRODROMUS. 25 two are found on our northwest coast, two are Himalayan, and one has recently been found at Aucher-Eloy in the moun- tains of Armenia. The order Stilbacece, prepared by Professor Alphonse De Candolle himself, consists of three genera each of a single known species, and of one with five species ; all of them natives of the Cape of Good Hope. The Globulariacece, by the same author, comprises the typical genus, with eight sj)ecies, and a new one of a single species ; all of Europe and of Eastern Asia, except one in the Canary Islands. The order Brunoiiiacece^ also by De Candolle, contains a single genus of two Australian species ; both made known by the prince of botanists whose name they bear. Mr. Boissier, the most active and promising botanist of the Genevan school, has elaborated the Plumb aglnacecB. The tribe Statioece comprises six genera, namely, 2Egialitis, R. Br., of the shores of eastern tropical Asia and Australia ; Acantholimon, Boiss., of forty-two Central Asian species, and Goniolimon, Boiss., of seven North- Asian species, — both dis- tinguished from the following by their capitate instead of filiform stigmas : Statice itself, reaching to one hundred and ten species ; Armeria, with fifty-two species, and Limonias- trum, of two Mediterranean species. The Statice of our own coast, S. Caroliniana^ Walt., Mr. Boissier distinguishes from 8. Limonium by its fistulous scape, stricter branches, pyram- idal instead of corymbose panicle, the distant one-flowered spikelets, and the very acute calyx-lobes; the Californian plant he introduces is a new species. The tribe Plumbagem consists of the Siberian Plumbagella, the European and tropi- cal Plumbago, the Abyssinian Yaloradia, and the African and North Indian Yogelia. In the Corrigenda to the volume we notice that Bentham has corrected the orthography of Trichostemma, so printed in the " Genera " of Linnaeus, and by mistake in the " Labiatarum Gen. et Sp.," to Trichostema, as written by Gronvius, by Linnaeus in the " Hortus Cliffortianus," and as the derivation requires. 26 REVIEWS. This half volume (Pars XIII., sectio posterior) has ap- peared very nearly at the date announced for it, last autumn, when the twelfth was published. It is the second part, antici- pating the first, which is to contain the Solcmaccce and the Plantaginacece^ two families which will finish the Monopet- alous series, as this begins the Apetalce or Monochlamydeca. It comprises the Phytolaccaceoe^ Saholacece ( Chenopodece)^ Basellacece, and AmarantacecG^ elaborated by Moquin-Tandon of Toulouse, and the JSfyctaginacece^ by Professor Choisy of Geneva. Of Phytolaccacece we have in the United States, only Petivera alliacea, which grows in Florida (probably not in " Carolina"), Pivina Icevis (to which we are surprised to see P. jjortulacoides, Nutt., joined), and Phytolacca de- candra^ which last is now so widely dispersed over the world that its native country is uncertain. The large family of Saholacece comprises 72 genera, dis- posed nearly as in Tandon's " Chenopodearum Enumeratio," in two suborders and seven tribes, most of which are further di- vided into subtribes. Our genera of the Cyclolohece (those with the embryo nearly annular) are Aphanisma, Nutt., a Calif ornian plant discovered by Mr. Nuttall ; Teloxys aristata, which is credited to us because Linnaeus referred his Cheno- podiumVlrginicum to C. aristatum, but it is doubtful i£ we possess the genus ; Cycloloma (^Salsola 2^latypJiylla.,Michx.^ ; Chenopodium, to which Tandon now reunites the greater part of his Ambrina (C amhrosiodes^ C. anthelminticum, etc.^, leaving in Poubieva only the original species, recently illus- trated in this Journal by Mr. Carey ; Blitum, to wdiich the author now refers, as a section, his former genus, Agathophy- ton QChenoj)odiii'ni Bonus Henricus^Tj.^ ; Monolepis, Schrad. (^Blitum chenopodloides^ Nutt.) ; Atriplex, of which too many of the older species are credited to the United States ; Obione, Gagrtn., of which nine species are North American, including (apparently with sufficient reason) the Pterochiton, Torr., Grayia, Hook, and Am., of a single species ; Eurotia ; a doubtful Kochia ; a Corispermum (which Tandon seems not to know as also a native of this country) ; Salicornia, in which we have S. herbacea f S. Peimviana (Caro. Eraser), and S. BE CANDOLLE'S PRODROMUS, 27 Virginica^ to which last he evidently would refer S. mucro- nata, Bigelow, a name unknown to him (and he has also dropped, apparently by accident, the homonym of Lagasca, so that the point in which we are interested is not elucidated) ; Ar- throcnemnm. (^A,f ambiguum= Salicornia amhigica, Michx.) being- still kept distinct. Of the Spirolobece we have in North America, Chenopodina, a genus newly founded for the Chenopodium maritimumy L., which was formerly referred to Suagda, besides which species Tandon also gives us C. line- aris, EIL, which, however, he thinks may be a variety of C. prostrata, which again he thinks may not prove distinct from C. maritima, and C. depressa (Salsola depressa, Pursh) ; of Shoberia, we have S. calceoUformis ( Chenopodium ccdceoli- forme. Hook.), which is stated also to be found " near New York " ; of Salsola, we have S. hali only. The singular genus Sarcobatus of Nees (the Fremontia of Torrey in the reports of Fremont's first and second journeys) is enumerated among the Genera exclusa, and said to be " dubise sedis." Prob- ably the author had not seen the figure of the fertile plant published by Dr. Torrey. Acnida, following the aspect and inflorescence, is here referred to the Amarantacece. The order Basellacca^, familiar to us only by the Bous- singaultia baselloides, which is cultivated as an ornamen- tal climbing plant, contains six genera, entirely of tropical plants. The order Amarantacem includes forty-five genera, ar- ranged under three tribes. There are credited to this coun- try : Celosia, one Californian species ; Amaranthus, about nine species ; Mengea of Schauer, a Californian species which has much the aspect of Amarantus Blltimi ; one or more species of Euxolus, Raf. (^Amarantus lividus, L. etc.) ; Acnida, in which A. rusocarpa appears to be mixed up, in a manner that requires much investigation to unravel, with Amarantus tamariscimis, Nutt., wdiich again, though entirely distinct from Acnida itself, nearly accords in character with Moquin-Tandon's section Montelia ; Banalia, a new genus, one section of which includes an Oregon species (^Halomoene- mis occidentalism Nutt., ined.) ; an obscure Polj^cneraura ; 28 REVIEWS. Gossypiantlius, Hook., two Texan species ; Iresine, two spe- cies ; Alternanthera, one species (^Achyranthus repens^ L.) ; besides the A. (Cladotlirix, Nutt.) lanuginosa^ which Lind- heimer and Wright find abundantly in Texas, and which will certainly stand as a separate genus, if a striking peculiarity in respect to its fruit, observed by Dr. Torrey, proves to be a normal condition. Telanthera ficoidea and T. polygonoides appear to be only introduced plants along our southern coast. Frgelichia (Oplotheca, Nutt.) has three North American species. Phyllepidium of Rafinesque is not identified and probably never will be. The remaining family, Nyctaginacece, includes eighteen genera, in three tribes. Of Mirablis, though no sjiecies are credited to us, we have one or more in Texas, as well as the three species of Nyctaginia, Choisy. Of Oxybaphus, six North American species are indicated ; and the Peruvian Alliona incarnata comes also from California. Four species of Abonia are described, besides A. f (Tripterocalyx) mi- crantha^ Torr., which Dr. Torrey has since raised to the rank of a genus. Pisonia aculeata is found on Key West. Boer- haavia furnishes us three or four species ; and there shall re- main some nndescribed Texan representatives of the family. The second part of the fourteenth volume of the " Prodro- mus " contains the Thymelceacece by Meisner, the Elceagnacece by von Schlechtendal, the(9'7'z«55iaceoB by De Candolle, resting merely on one of those outlying or anomalous genera which there is too great tendency to raise to ordinal importance, merely because the author knows not what to do with them, — and SantalacecB by De Candolle. Of the first order we have only Dirca, peculiar to this country, and with no congener known. There is nothing to add respecting our three species of Eloeagnaceoe. As to our few Santalacece., it is interesting to remark that one of our characteristic genera, Pyrularia (the Oilnut), is found to have two representatives in the Himalayas (Sphserocarya, Wall.), and apparently two more in southern India (Scleropyrum, Arn.). Also that a Euro- pean species is introduced into our Comandra (the Thesium elegans of Rochel), and the genus itself shown to be hardly DE CANDOLLE'S PRODROMUS. 29 distinct from Thesium. And Darbya, Gray, published in this Journal twelve years ago, is reduced to a subgenus of Comandra, to which we are not disposed to object. But we take the new species of true Comandra ( C. ijallida) to be a mere variety of C. umbellata ; which, by the way, we did not state to be eight or ten feet, but only as many inches in height. De Candolle thinks that the hairs which connect the anthers of Comandra, and of most Thesia also, with the per- ianth, belong to the latter, not to the former, as the generic name implies. Our own observations, and especially some made by Mr. H. J. Clark upon very young flower-buds, con- firm this view. The discovery, announced in this Journal in 1854, that the striking genus Buckleya, Torr., is truly dichla- mydeous in the female flowers, proves a capital fact for M. De Candolle ; who draws from it the confident inference that the floral envelope which in all other plants of the order occurs alone, and has the stamens opposite its lobes, is corolla and not calyx, and consequently so in the Loranthacece and ProteacecB also. Our author's views are presented in detail in an article, " Sur la Famille des Santalacese," in the " Biblio- theque Universelle," published last autumn, and they appear wellnigh convincing. An analogous case is found in Zan- thoxylum (only here the suppression is the rare case), Z. Americanmn plainly wanting that which in Z. Carolinianum is the corolla (" Genera Illustr." 2, p. 148). Nyssa offers a good instance of the limb of a calyx so reduced as to have escaped notice until four years ago. For what to De Candolle seem to be petals (p. 622, note in char, of order Santalacece}, were seen to be so, and the observations recorded in the fifth volume of the " Memoirs of the American Academy," p. 336, and afterwards extended in the " Manual Bot. U. S.," ed. 2, p. 162 (1856). It is singular that De Candolle should re- main so uncertain of the place of Nyssa in the Natural Sys- tem. If he will compare it and Mastixia with Cornus he will surely be convinced that Nyssa is a true Cornaceous genus. So of Cevallia, the true place of which our author seems not to know, although given in the " Flora of North America " many years ago, under the sanction (we may add) of the very 30 REVIEWS. highest authority. Indeed, so plain is its relationship to Gronovia that Fenzl soon saw and corrected his mistake in referring the genus to Calycerece. And if at this date any should doubt that these are Loasaceous plants, let them turn to the characters of Petalonyx, in the "Memoirs of the American Academy," v., p. 319. Leaving these details, let us consider our pleasing prospect in respect to the continuation (at least through the Dicoty- ledoneoe) of the great work upon which the De Candolles, father and son, and other excellent botanists, have bestowed so much labor and talent. The great order of Lauracece was to have been included in the present volume. It would have extended the volume unduly. But, unfortunately, or fortu- nately, as the case may be. Professor De Yriese has gone to Java on a government mission without finishing the work ; and the indefatigable Meisner now takes it in hand. It is to form the leading part of volume xv., the Begoniacece by De Candolle himself, and the Aristolochiacece by Duchartre being appended, and perhaps the £Juphorbiaceoe, also by De Candolle, except the genus Euphorbia which Boissier under- takes. The sixteenth volume is intended to commence with the Urticacece proper, by Weddell, or the 3Ionimiacem by Tulasne. We are pleased to learn that Professor Anderson of Stockholm is to elaborate the Salicineo^. The other section of the sixteenth volume of De Candolle' s " Prodromus " has just been issued. The two parts form in- deed independent volumes, and are paged and indexed as such, so that for all time botanists will have to quote D C. Prodr. xvi. (I), p., etc., which is to be regretted, but there is no help for it. The present (prior) part, of 450 pages besides 65 intercalated ones, contains the Buxacem and some other plants excluded from the Uuphorhiacece, by Dr. Miiller ; the Empe- tracecE by Alphonse De Candolle himself (Empetrum reduced to one species, Corema of two, and a Ceratiola) ; Cannahinem by the same (the JJlmacece and Artocarjyem postponed not being ready), the Urticacece (i. e. the Urticece) by Weddell ; Piperacece by Casimir De Candolle (the Saururem made a mere tribe, and the PipereoR mainly included under Piper of DE CANDOLLE'S PRODROMUS. 31 635 species, Peperomia of 389 !) ; CJdoranthacece by Solms- Laubacli of Halle, ledueed to three genera ; and finally, Gar- ryacece by the editor, comprising nine species of Garrya. It appears that the latter end of the volume was printed first, which explains the omission of G. huxifolia^ a species discov- ered in northern California by Bolander, and published a year and a half ago. If one or two collaborators will now bring up their arrears, the editor may very soon have the great satisfaction of announcing the completion of the great Dicotyledonous series. The seventeenth volume of De Candolle's "Prodromus" contains, in the first place, certain small outlying orders, or some of them genera that have to do duty as orders, which on various accounts have been left out of the " Prodromus " as it went on, namely, Sarraceniaceo3^ Phytocreneoe^ Car- diopteridece^ Salvador acece^ Cynocramhe^ Satidacece (one Ba- tis), Lennoacece (by Solms-Laubach), Podostemacece (by Weddell). Then Nepenthacea;^ by Dr. Hooker ; Cytinacea', by the same ; Balanophoracem^ by Eichler ; Vlmaceoe. by Planchon ; Moracec^^ by Bureau ; and a synopsis of the genera of Artocarpece. For the complete elaboration of the last family the volume has been a good while kept back, and has at length been issued, and (sad to say) the work concluded without it. It is to be hoped that whenever M. Bureau finishes his undertaking, the publishers of the " Prodromus " may print the Artocarj)Cce uniformly with the rest, so that it may be appended. Then follow a few pages of Genera omissa, with brief references, indicating that they are, or may be, so far as has been made out or conjectured ; and finally, the wearied editor appends his "Prodromi Historia, Numeri, Conclusio." It is a terse and highly interesting account of this work, which (including the two preceding volumes of " Systema") occupied his celebrated and indefatigable father from the year 1816, or earlier, down to the end of his life, in 1841, and himself to the close of the past year : an enumeration of the contributors who have worked up particular families or genera ; an enu- meration of the orders, specifying the volume which contains 82 REVIEWS. each, and the number of genera and species described, 5134 genera and 58,975 species, which the missing Artocarpece, it is estimated, would bring up to 5163 genera and about 60,000 species of Dicotyledoues. Among a few statistical data which are given, the ten orders are enumerated wdiich contain the greatest number of genera, beginning with Com- positce and ending with Cruclferce. Then the ten which most abound in species, which begin with Comj^ositm (8561 species) and end with Umhelliferce (1016). Leguminoseoe are the second in both lists, and next Ituhiaceos in the former and JEuphorhiacece in the latter. But the long interval be- tween the publication of many orders, say between Crucifercv. and Eupliorhiaceoe^ much diminishes the value of such com- parisons. The reasons which have prevented a more rapid publication of the volumes of the " Prodromus," especially since the work has been largely distributed among collabora- tors, are hinted at ; and finally the regrettable announcement is made that the publication is now relinquished, at the close of the Dicotyledones. A full Index, down to genera and their sections, filling 170 pages, closes this great work. We sincerely congratulate the editor upon the successful completion of this great undertaking at the limits he felt obliged to prescribe, and thank him heartily for his long and faithful service and many sacrifices. As it may be hoped that he has still years of good work in him, all will regret that he could not bear this burden through a few of them, while a half dozen collaborators, who might be named, elaborate the Monocotyledonous orders. But, as he declares that he should doubtless perish under it, we prefer the living botanist to the completed " Prodromus." We may expect from him origi- nal work instead of editorial drudgery, perhaps a new edi- tion of his Geographical Botany, or new researches upon the same subject, investigated with his impartial judgment, under the new light which was just dawning when that comprehen- sive treatise was published. Since these remarks were written we have received an in- teresting pamphlet, separately issued from the " Archives des Sciences " of the '' Bibliotheque Universelle " for November, ENDLICHER'S GENERA PLANTARUM. 33 entitled, " Reflexions siir les Ouvrages Gen^raux de Bota- nlque Descrij)tive." In this M. De Candolle gives the his- tory of the " Prodromus " and its forerunner with consider- able fullness, explains more particularly his editorial trials and burdens, and the reasons why the work could not be made to get on faster, and gives his views as to the most practicable method of combining the labors of the botanist of another generation in the production of the new " Systema Vegetabiliuni " which will be demanded. An estimate is made of the time it must needs require, even with all the available monographers of the day enlisted in the service. The in- creased difficulties, or at least the augmented labor, of sys- tematic botanical work, under the present demands of the science, are indicated. It appears that while in his father's time one could elaborate at the rate of about ten species a day, a faithful monographer now, under the modern require- ments, can seldom exceed three or four hundred species per annum, that is, about a species a day! We suppose that the case on the whole is not overstated. ENDLICHER'S GENERA PLANTARUM. This is one of the most iniDortant works ^ of the asfe : and we are anxious to make it more generally known to the bota- nists of this country. It is not too much to say, that without this, and Lindley's introduction to the Natural System (or some equivalent work), no person who does not possess the advantage of a large library and an extensive general collec- tion of plants, can obtain any correct idea of tlie present state of systematic botany. The work is published in parts, of eighty pages each, in an imperial octavo or a kind of oblong quarto form, closely printed in double columns. The eleventh fasciculus, which is the last we have received, reaches to page 880; but probably two or more additional numbers have by 1 Endliclier's Genera Plantarum secundum Ordines Naturales disposita. Vienna, 1836-40. (American Journal of Science and Arts, xxxix. 176.) 34 REVIEWS. this time appeared. It is stated in the original announcement that the work will not exceed ten or twelve numbers ; we imaGfine, however, that four or five additional numbers will be required for its completion. It commences, like the " Genera Plantarum" of Jussieu, with the plants of the simplest or lowest organization (Thallophyta, Endl.) ; a plan which is now the most common and perhaps the most philosophical, but which is attended with many practical inconveniences to the tyro. The first edition of the '' Genera Plantarum " by Linnaeus was published at Leyden in the year 1737 ; the second and third w^ere published at the same place, the one in 1742, the other in 1752 ; the fourth and fifth were published at Stock- holm ; the latter (termed the sixth in our copy) in the year 1764, which is the last by Linnseus himself, is the edition gen- erally cited, and was reprinted at Vienna in 1767. This last Stockholm edition forms the excellent model of all the suc- ceeding editions, as they are termed, edited by various au- thors. It comprises 1239 genera, which in an appendix are reduced as far as possible to their proper natural orders. The first edition after the death of Linnseus is, w^e believe, that of Reichard, published at Frankfort in 1778, about the same time with the edition of the " Systema Plantarum " by the same author. To this succeeded the edition by Schreber (published also at Frankfort, 1789-1791, in two volumes), who is chiefly famous for having in this work changed all the unclassical names of Aublet and others for new ones made according to the Linnaean canons. Succeeding authors in plu(?king these borrowed plumes have despoiled him of some riglitf Lil feathers ; as in the case of the genus Brasenia, for which most botanists have retained Michaux's name, Hydro- peltis, which was published a dozen years later. The number of .)," were the innovation proj)osed in his day, and have written T. cinni- tus, Swains., had he thought best to approve the division of his genus Muscicapa. On the other hand, if he disapproved the division, we may add, he would not have thanked a con- temporary for making him seem to adopt it. The hardship is still greater when the question is not of the division of an old genus, but of the proper place of a species among ad- mitted genera, when it is surely improper to cite an author as referring to one genus, while he expressly maintains that it belongs to another. In fact, the remedy is much worse tlian the disease which the English doctors would cure. Linnaeus maintains that he is the true naturalist who understands 50 REVIEWS. genera ; but from the new practice it will inevitably follow, as Professor Agassiz asserts, that the proper establishment and definition of genera which demand the highest powers of the naturalist, will be less esteemed than the mere dis- tinguishing of species ; a result which, far from promoting science, will especially retard the progress of that part of zoology in which there is most to be done, and in which the science of the animal is still far behind that of the vegetable kingdom. It is of the greatest importance that we should be able to thread our way back through entangled synonymy and mistaken references to the original sources. Here our difficulties would be greatly multiplied, unless two sorts of synonyms are used. For who, as Mr. Agassiz says, can find out what Linnaeus has said of lluscicapa crinita, without a direct reference to the genus in which Linnaeus himself placed it ? And when, as often happens, the Linnaean species is mistaken, so that the Tyrannus crinitus^ Linn, (sp.) accord- ing to Swainson, is not the T. criiiitus, Linn, (sp.) accord- ing to some other author, the confusion becomes inextricable, unless we encumber ourselves with two modes of annotation, the old for expressing synonymy, and the new for the names really adopted. " Then the two modes will not agree with each other, nor can one know whither to turn himself. Surely the authors of this new rule cannot have considered these inconveniences, else they would have themselves discarded it. Therefore I entreat and pray them, by all the interests of the science they wish to promote, to abandon their proposi- tion, and not to introduce a new schism into natural history, but to return again to the system of Linnaeus, the most simple of all, and the least likely to errors and Babylonish confusion in nomenclature." The Committee of the American Association more wisely adopted the mode, afterwards employed by Mr. Dana in his great work on Zoophytes, namely, that of appending to the specific name the original authority for the species in brackets, and adding without brackets the name of the author who first described the species under the later received genus. To this plan there can be no objection, except that VON MOHUS VEGETABLE CELL. 51 it is rather cumbrous, if it is to be used in every brief mention of the sj^ecies, and, in our opinion, quite superflu- ous in a systematic treatise, where the synonymy is given in proper historical order. The recommendation to make sub- generic names agree in gender with that of the genus. Pro- fessor Agassiz thinks is of no consequence, unless the new annotation, just animadverted upon, should come into use. Besides, it would often interfere with the rule of priority, which requires synonyms, when they exist, to be adopted for sectional names. But he strongly commends the rule, that the etymology of names should always be stated by the pro- poser. Justly does Mr. Agassiz condemn the practice of those who change the authority of a genus when they extend or narrow its bounds, or correct a faulty orthography. Thus he would write Lepidosteus, Lacep., although Lacepede wrote Lepi- sosteus ; and especially would he write Perca, Linn. (Cuv.), not Perca, Cuv. Hearty and just, also, is his censure of the custom of those French zoologists who use vernacular appellations in scien- tific works, either to the exclusion of the systematic name, or in precedence of it. Our author closes this part of his preface with some ex- cellent reflections on the study of genera in the animal king- dom, and the need of a thorough reinvestigation of the grounds upon which natural families are constituted; re- marks which we would gladly copy, if our limits allowed. VON MOHL'S VEGETABLE CELL. We desire in a special manner to commend this condensed treatise ^ not only to botanists, but to animal physiologists, to ^ Grundziige der Anatomie und Physiologie der vegetahilischen Zelle. Hugo Von Mohl : Braunschweig, 1851. English translation {Principles of the Anatomy and Physiology of the Vegetable Cell) by Arthur Henfrey, London, 1852. (American Journal of Science and Arts, 2 ser., xv. 451.) 52 REVIEWS. medical students, and to all who would obtain a clear view of the present state of vegetable anatomy and physiology, — a knowledge of which, most interesting in itself, is almost indis- pensable to the correct understanding of the minute anatomy and physiology of animals. Professor Mohl is, without ques- tion, the first of vegetable anatomists, and his statements carry with them the highest authority on this class of subjects. We copy the short preface which he has contributed to the Eng- lish translation, as it gives a clear view of the nature and scope of the work. " Mr. Arthur Henfrey having informed me that he intends publishing an English translation of the present treatise, I take this opportunity of making known to the English reader the purpose I had in view in the preparation of the book. The following pages were not originally intended to appear as an independent work, or to give a summary of the wide subject of the Anatomy and Physiology of Plants, but appeared as an article in the ' Cyclopedia of Physiology,' published by Dr. Rudolph Wagner of Gottingen, drawn up to furnish students of Animal Physiology, and more particularly the medical profession, with a review of the anatomical and physiological conditions of vegetables (of the cell), in order to enable them to form a definite judgment upon the analogies which might be drawn between the structure and vital functions of animals and plants. This intention, together with the circumstance that I was compelled to crowd the whole exposition into the space of a few sheets, rendered it necessary to direct especial attention to the individual cell, as the fundamental organ of the vegetable organism. Since, however, the cell only pre- sents itself in anatomical and physiological independence in the lowest plants, and since, in the more highly organized plants, both the structure and the physiological functions of the individual cells become subject to greater dependence upon the other parts of the plant, in proportion as the collec- tive organization of the vegetable is more complex ; moreover, since functions then present themselves, of which no trace can be found in the lower plants, it became requisite to take ac- count of the plants of higher rank, and of the various organs VON MOHUS VEGETABLE CELL. 53 which these possess. The treatise, therefore, contains, if an imperfect, still in many respects a more extensive resume of Vegetable Physiology than might have been conjectured from the title. " Unhappily, the Physiology of Plants is a science which yet lies in its earliest infancy. Few of its dogmas can be re- garded as settled beyond doubt ; at every step we meet wath imj^erfeet observations, and consequently with the most con- tradictory views ; thus, for example, opinions are still quite divided regarding the doctrines of the development of the cell, of the origin of the embryo, and of the existence of an impregnation in the higher Cryptogams. Both in these and in other cases, the small compass of the present treatise for- bids a more extensive detail of the researches upon w^hieh the opposing views are founded ; I hope, however, that I have succeeded in making clearly prominent the chief points upon which these contests turn, and thus in facilitating the forma- tion of a judgment by the reader; and I have never neglected to indicate the literature from which further instruction is to be derived." It may be well to notice the views of so excellent an ob- server upon sundry points which have been more or less mat- ters of controversy. As to the milk- vessels, or vessels of the latex, Mohl inclines to adopt the view that considers them as intercellular passages which have acquired membranous lin- ings (p. 2). He denies that the membrane of nascent cells is soluble in water, as Schleiden states (p. 9). He briefly states the ground on which, in his controversy with Harting and Mulder, he successfully maintains that the primary cell- membrane is thickened by successive concentric layers of cellulose deposited on its inner face. The combination of spiral markings and pits on the wood-cells of Taxus and Tor- reya, as also in the Linden, is explained by considering the former to belong to a second layer or deposition within that to which the pits belong. This tertiary membrane or deposit forms the spiral fibre or band in the cells of the seed-coat of CoUomia, the hairs of the achenium of Senecio, etc. (p. 18). -The whole subject of spiral and other markings, rings, dots, 64 REVIEWS. slits, reticulations, etc., on the walls of cells is expounded in a masterly and convincing way. Mohl maintains (p. 28), as he has done in the " Botanische Zeitung," that cellulose forms the basis of all vegetable membranes in the higher plants, and that what Mulder regarded as peculiar compounds are com- binations of cellulose with foreign infiltrated deposits, which interfere with the chemical reactions of cellulose, but which may be removed by previous maceration in caustic potash and nitric acid. He now maintains, in opposition to his early views (still defended by Schleiden), that the intercellular sub- stance is a product or secretion of the cell, and not a univer- sally distributed mass in which the cells are imbedded (p. 33). He shows that the thickened " cuticle '* of Unger, Mulder, Harting, etc., consists of secondary layers of cell-membrane, deposited from the inside, and infiltrated with some substance that is colored brown by iodine ; with the exception of an extremely thin external pellicle, the real cuticle of Brongniart, which is probably a secretion from the surface of the cell, like that which forms the outer coat of pollen-grains (p. 35). He insists that the layer of protoplasm, his primordial utricle, lining the cell is a soft and delicate membrane, and not a mere layer of mucilage (p. 37) ; the mode in which this is con- structed and a partition formed by an inward growth at the fold, in the multiplication of cells by division, is very clearly explained (p. 50-53). Free cell-formation is said to occur, in Phagnogamous plants, only in the embryo-sac, in which both the embryonal vesicle, or rudiment of the embryo, and the cells of the albumen, originate in this manner ; in the Crypto- gamia, only in the formation of spores in certain cases, as in the Lichens (p. 58) ; contrary to the view of Schleiden, who long maintained this to be the universal, and lately a general. Diode of cell-production. Schleiden's original account of the process of the formation of a free cell from a nucleus is directly controverted in all essential points ; the nucleus, ac- cording to Mohl, being always central, and at no time con- nected with the cell-membrane, but always enclosed in the primordial utricle. A nucleus, or mass of nitrogenous sub- stance, first has the primordial utricle, or nitrogenous mem- VON MOHVS VEGETABLE CELL. 55 brane, developed over its surface ; and then the cell-membrane (of cellulose) is deposited upon this. In Cryptogamous plants, such masses, of larger or smaller size, may become coated with a cell-membrane without any proper nucleus ap- pearing (pp. 57-60). Thus much for what relates to the ana- tomical condition of the cell. Under the second head, the Physiological Condition of "the Cell, our author treats, first, of tlie cell as an organ of nutrition, next, as an organ of propagation, and finally, as an organ of motion. He pronounces against the Knightian doctrine, that plants ultimately degenerate and perish wdien propagated for generation after generation by division (from the bud) (p. 64). That the crude sap, though absorbed by the parenchymatous tissue of the root, ascends through the woody tissue, and that the assimilated sap returns through the bark, and thence more or less into the wood by means of the medullary rays, is very neatly shown. " A few simple ex- periments leave no doubt about this. . . . If the bark of a plant, best of a tree, is cut through in a ring down to the wood, there is no interruption in the flow of sap to the parts situated above the wound ; but if the wood is cut through, the greatest care being taken to avoid injuring the bark, that por- tion of the plant above the wound dries up at once. From the wood of the stem and branches the sap flows onward into the leaves, as is proved by the powerful expiration of watery vapor from them. Before the sap has reached the leaves it is incapable of being applied to nutrition ; consequently the vegetation of a plant comes to a stand-still when it is deprived of its leaves. The sap ascending from the root to the leaves is thence termed the crude sap. It undergoes a chemical change in the leaves, rendering it fit to be applied to the nutrition of the plant. To this end the sap flows backwards from the leaves through the bark to the lower parts, as the following circum- stances testify. If the bark is cut off the stem in a ring, the growth of the portion below the wound stands as it were still ; the stem becomes no thicker ; in the potato plant no tubers are produced, etc. ; but, on the other hand, the growth above the wound is increased beyond the usual measure, thicker 66 REVIEWS. layers of wood are deposited, more fruit is perfected, tliis ripens sooner, etc. The deposition of starch which occurs in the cells of the medullary rays in autumn, goes to prove that the portion of assimilated sap which is not used for nutrition on the way to the root, runs back to the wood through these horizontal medullary rays ; and thus the sap describes a kind of circle, not, indeed, in determinate vessels, but in a definite path leading through the different parts of the plant. It is difficult to see how in recent times the results of these experiments could have been so questioned, and the existence of the de- scending current in the bark denied. Certainly it is no im- provement on the theory cast aside when the increased growth above the annular wound is explained by the artificial inter- ruption of the upward current of crude sap, in consequence of which the sap contained in the upper part of the plant must soon become greatly concentrated and potential for development (Schleiden, ' Grundziige,' 2d ed., ii. 513). When we can succeed in fattening an animal by depriving it of a portion of its accustomed food, this explanation may be received as satis- factory " (p. 70, 71). After some excellent points of criti- cism, Mohl concludes that the discovery of endosmose has not fully solved the problem of the movement of the sap in plants, although in all probability it does play an important and j^er- haps the principal part in its absorption and conve3^ance (p. 77). To the question, whether plants live on inorganic food alone or take in also organic matters, Mohl gives a sen- sible answer, rejecting the extreme view of Liebig, while still fully recognizing the great office and result of vegetation (p. 78). According to Mohl, however, it is proved that plants do not absorb the carbonic acid dissolved in water with the latter by means of their roots (p. 81) ; but this seems hardly reconcilable with several facts stated on the next page, from which it is justly concluded that carbonic acid is carried up with the ascending sap into the leaves. From the fact that plants perish so soon in air deprived of all oxygen gas, that sensitive leaves lose their irritability under such circum- stances, etc., Mohl concludes, apparently with good reason, that the absorption of oxygen and the exhalation of carbonic VON MOHUS VEGETABLE CELL. 57 acid in plants is a true physiological function, intimately con- nected with the life of the plant ; and that this (rather than the opposite and predominant nutrient process, through which carbonic acid is decomposed and oxygen evolved) should be considered as the respiration of plants, if we use the term at all (p. 8G). "The roots of plants, and not the leaves, take up the substances which furnish plants with nitrogen, while, on the contrary, the leaves play the essentially active part in the absorption of carbonic acid " (p. 88). The analogy of the milky juice of plants to the blood of animals, as pro- pounded by Schultz, is thoroughly refuted by our author, and the flowing movement in the milli-vessels described by Schultz is positively denied to take place in an uninjured plant, except as produced by mechanical causes. That the milky juice is not a nutrient material, still less the nutrient juice, is also manifest (p. 96). The cell as an organ of propagation is treated, first, as respects the multiplication of plants by division ; second, by spores; and third, by seeds. The conjugation of certain Confervoid Algce, such as Zygnema, is said to bear no anal- ogy to sexual reproduction (p. 113) ; a conclusion which may be questioned. A good summary is given of the facts known respecting the free and spontaneous movements of the spores of the lower Algm (p. 115) ; and also of the recent discoveries respecting the bisexual reproduction of the higher Crypto- gamia. The reprint of Henfrey's Report, in the January and March numbers of this Journal, has placed our readers au courant with the present state of knowledge on this interesting subject. It should be noticed that Mohl denies the existence of antheridia in lower Cryptogamia, or Thdlopliytes ; but maintains that the small bodies, moving by two cilia, discov- ered by Decaisne and Thuret in the Fumcece^ are more prop- erly a second kind of spores, analogous to the small spores of the Floridece, than of the nature of the seminal filaments of Ferns, Mosses, etc. (p. 117). The latter researches of Itzigsohn, Thuret, Tulasne, etc., however, lead rather to the conclusion that the lower Cryptogamia (except the very low- 58 REVIEWS. est) are likewise bisexual.^ Under the head of Eeproduction by Seeds, Mohl gives an interesting and critical account of the development and structure of the pollen and the ovule, and of the origin of the embryo. The latter arises by cell-multipli- cation of the germinal vesicle, a cell produced by free cell- formation in the embryo-sac usually before the pollen-tube has reached the latter. The germinal vesicle and the extremity of the pollen-tube are separated by the thickness of the parie- ties of the embryo-sac. The penetration of the pollen-tube into the latter or into an introverted portion of it, and the formation of the embryo from the apex of the pollen-tube itself, as taught by Schleiden, are wholl}^ repudiated ; and indeed the Schleidenian doctrine may now be considered as thoroughly demolished, by the direct observations of Amici, Mohl, Muller, Hofmeister, Unger, Henfrey, and Tulasne. The Cell as an Organ of Motion is considered as respects move- ments of individual cells through the agency of vibratile cilia, as respects the directions and curvature assumed by organs, and as respects movements by irritation of stimuli, etc., giving an excellent summary of our knowledge on these points, with much admirable criticism ; which want of space prevents us from noticing in detail. 1 Itzigsohn in Botanische Zeitung, May, 1850. — Here it is announced that the hlack clots on the surface of the frond of Burrea ciliaris contain antheridia, that is, cells from which escape animalcular-like corpuscles that move freely in water, and are similar to those of Mosses and Liver- worts. Later, after stating that others had failed to detect those move- ments, he announces that they had been observed by Rabenhorst, after many ineffectual trials. He also (December, 1850, February, 1851) states that these " spermatozoids " do not manifest viral movements until after the maceration of the Lichen in water for several days. Tulasne {L. R.) Memoire pour servir a VHistoire Organographique et Physiologique des Lichenes ; in Ann. Sciences Naturelles, 3 ser., xviii., No. 1, 2, 3, 4 (1852), with 16 plates. — A most admirable and complete memoir, elucidating in an unequalled manner the whole structure and morphology of the Lichens. It is to be hoped that the author will publish it in a separate form, as it introduces a new era in Lichenography. On the subject of the so-called antheridia (which alone we can here notice), M. Tulasne has recognized the universal occurrence of these bodies in Lichens, has ascertained their structure and development ; but he has never detected any free movement of the corpuscles, except the general GROWTH OF PLANTS IN GLAZED CASES. 59 ON THE GROWTH OF PLANTS IN GLAZED CASES. The first edition of this little treatise/ published in 1842, is doubtless well known to many of our readers ; and some may remember Mr. Ward's original account of his interest- ing discovery of a method of growing every sort of plant in the dun atmosphere of the smokiest part of London, pub- molecular or Brownian motion common to all minute particles. He therefore gives to the so-called antheridia the name of spermogonia, and to the contained corpuscles the name of spermatia. He unhesitatingly recognizes in them an apparatus of reproduction, doubtless analogous, at least in function, to those of the Florideous AlgcEy in which the corpuscles are equally motionless, and of certain Fungi, and therefore probably representing male organs. Tulasne likewise calls attention to the fact that these dark tubercles or dots were particularly noticed by Dillenius, more than a century ago, in Borrera ciliaris ; and that Hedwig, in 1784, expressed the opinion that they constituted the male apparatus of Lichens. Decaisne & Thuret, Recherches sur les Antkerides et les Spores de quelques Fucus ; in Ann. ScL NaL, 3 sen, iii. p. 5. — Here the corpuscles known to the earlier Algologists, and considered by Agardh and Montague as a second kind of spores (a view which Mohl adopts), are announced to be the spermatozoids of the antheridia of Fucacece ; their active movements are described, and the discovery of the two cilia is announced by whose vibration the movement is effected. Thuret, Recherches sur les zoospores des Algees et les Antherides des Cryptogames. — These researches were communicated to the Academy of Sciences, Paris, and were rewarded by the great prize for natural sciences, in 1847. A copious abstract has been published in the Ann. Sci. NaL, 3 ser., xiv. and xvi. (1850, 1851), with 30 plates. As to anther- idia, bodies like the free moving corpuscles of the Fucacece are shown likewise to occur in all the Floridece, except that they do not exhibit spon- taneous movements ; nevertheless, M. Thuret does not hesitate to attrib- ute to both the same functions as those which the seminal filaments of the higher Cryptogamia fulfil. The Antheridia of Chara (in which Thuret first discovered the cilia by whose vibration the coiled filaments are moved), of the Liverworts, Mosses, and Ferns, are also admirably illus- trated ; but nothing of consequence is added to the facts mentioned in Henfrey's Report. L^veille, in Ann. Sci. Nat, 3 ser., xv. p. 119, has indicated the prob- able existence of the analogues of antheridia in Fungi. 1 On the Growth of Plants in closely glazed Cases. By N. B. Ward. London, 1852. (American Journal of Science and Arts, 2 ser., xvi. 132.) 60 REVIEWS. lished in the " Companion to the Botanical Magazine " in 1836. This new edition if reduced in size is increased in interest, and is embellished with tasteful illustrations on wood, several of them exhibiting approved forms of those glazed cases with which the name of our author is inseparably connected. The first chaj^ter, on the natural conditions of plants, their rela- tions to heat, light, and moisture, and the necessity of attend- ing to the particular conditions, or combinations of circum- stances, under which each species flourishes, is illustrated by ingenious and often novel observations. The second chapter treats of the causes which interfere with the natural condi- tions of plants in large towns, and gives some idea of the obstacles which prevent the cultivation of even ordinary plants in the open air of London, and to some extent in other large British towns. The third, on the imitation of the nat- ural conditions of plants in closely glazed cases, tells us how a simple incident (the accidental growth of a seedling fern and a grass in a glass bottle, in which the chrysalis of a Sphinx had been buried in some moist mould), carefully and wisely reflected on, taught Mr. AA'^ard how to overcome these obstacles, and thus to surround himself with his favorite plants, in beautiful vegetation, while living in one of the murkiest parts of London, and even to grow with complete success such ferns as the THchomanes radicans^ which is ut- terly uncul Livable in any other way. A fourth chapter treats of the conveyance of living plants on shipboard ; which brings to view one of the most important practical applica- tions of Mr. Ward's discovery. Sir William Hooker states that " the Wardian Cases have been the means, in the last fifteen years, of introducing more new and valuable plants to our gardens than were imported during the preceding century ; and in the character of Domes- tic Greenhouses, i. e., as a means of cultivating plants with success in our parlors, our halls, and our drawing-rooms, they have constituted a new era in horticulture." Formerly only one plant in a thousand survived the voyage from China to England. Eecently, availing himself of our author's discovery, Mr. Fortune planted 250 species of plants in these cases in GROWTH OF PLANTS IN GLAZED CASES. 61 China, and landed 215 of them in England alive and healthful. The same person lately conveyed in this way 20,000 growing tea-plants, in safety and high health, from Shanghai to the Himalayas. In fact, this mode of conveyance is now univer- sally adopted, and has proved so successful, whenever prop- erly managed, that it is no exaggeration to say that, prob- ably, " there is not a single portion of the civilized world which has not been more or less benefited by the invention." An indispensable requisite to success in the transmission of living plants by this method is, that the glazed cases should be freely exposed to the light. Where this cannot be done, we must be content with the former method, of conveying plants in a passive condition, closely packed in peat-moss, — a plan, however, which is only partially successful in pro- tracted voyages. Two additional and highly interesting chapters treat of the application of the " closed " plan in im- proving the condition of the poor ; and on its probable future applications in comparative researches in vegetable phys- iology, and even in the treatment of diseases. To these, as to the other topics of the work, no justice can be rendered to our author's suggestions except by lengthened quotations, which the nature of this notice does not admit of. It must suffice to direct attention to this fascinating little volume. Those who read it and who have a true fondness for gTOwing plants will scarcely be contented without a Ward case, of more or less pretension ; which they will find an unfailing source of interest, especially during the long and total suspension of vegetation in our protracted winters. With proper manage- ment, and wdth the requisite amount of light, any plant may thus be cultivated. But we particularly recommend Ferns and Lycopodia, of the most delicate kinds, as requiring least care, and as making the prettiest appearance at all seasons. Most of these require little light ; although our clear skies afford us this in abundance. So little bituminous coal is consumed, even in our largest cities, that the " fuliginous matter " with which all British towns are begrimed and rendered noxious to vegetation, here interposes no obstacle to rearing plants. Quite unlike England, the principal obstacle to the growth of 62 REVIEWS. delicate plants in our houses in winter, and in our grounds in summer, comes from the dryness of the air. For this, the Ward case affords a perfect remedy; as nothing is easier than to furnish a saturated atmosphere for those plants that require it, or to supply and retain the degree of moisture which suits any particular species. HOOKER AND THOMSON'S INDIAN FLORA. One half of this volume ^ is occupied by the Introductory Essay, in which a series of important general topics, akin to those discussed in the introduction to Dr. Hooker's New Zea- land Flora, are treated with equal boldness and judgment, and with the same freshness and originality of illustration. These are arranged under six general heads, namely: 1. The object, scope, and design of the " Flora Indica." 2. General considerations connected with the study of systematic botany. 3. The variation and origin of species, the effects of hybridi- zation, and the geographical distribution of species. 4. Sum- mary of the labors of Indian botanists. 5. Sketch of the meteorology of India. 6. Sketch of the physical features and vegetation of the provinces of India. To which two maps are added : one of monthly isotherms, from Dove ; the other a large and original map illustrating the physical geography of India and its botanical provinces. A complete alphabetical index to this part of the work is appended, as well as a detailed table of contents. To enumerate even the principal points which are discussed would require a space which we are unable now to devote to this subject. Some of them we may hope to consider here- after in other connections. Among the conclusions or sug- gestions that strike us as most true and timely are : The great ^ Flora Indica : being a systematic account of the Plants of British India. Vol. I. By J. D. Hooker and Thomas Thomson. (Ranu7iculacecB to Fumariacece.) With an Introductory Essay. London, 1855. (American Journal of Science and Arts, 2 ser., xxi. 134.) HOOKER AND THOMSON'S INDIAN FLORA. 63 want on the part of many naturalists of clear and logical views in respect to classification and system ; — " the prevail- ing tendency on the part of students of all branches of nat- ural history to exaggerate the number of species, and to separate accidental forms by trifling characters;" — the un- philosophical and detrimental character of " the modern sj^s- tem of elevating every minor group, however trifling the peculiarities by which it is distinguished, to the rank of a genus ; " in other words, of considering every group of species to form a genus, — evincing a want of appreciation of the true value and nature of classification ; — the fact that in the vege- table kingdom we do not discover that close and obvious con- nection between structure and function which is almost uni- versally apparent in the animal kingdom, giving to physiol- ogy a greater influence over classification in zoology than in botany, and offering a guide to determining the relative value of structural characters in the one kingdom which is com- paratively little available in the other, but yet may not safely be neglected. Our authors assume, as most accordant with known facts on the whole, that species are distinct creations, and not arbi- trary assumptions of the systematists ; and they adopt that idea of species which alone appears to give them a perfectly clear and intelligible, distinct, objective existence in nature, namely, that they consist of individuals which have originated each from a common stock. They assume not only their original, but their continued definiteness in nature ; but their variations, surprising as they often are, are restricted within certain limits, to which we may add that these limits are not a priori determinable. Among the causes inducing variation, or tending to produce a blended series of individual forms, if such did not exist from the beginning, they first consider the effects of hybridization ; and remark that recent experiments have led to the following results : " 1. It is a much more difficult operation to produce hy- brids, even under every advantage, than is usually supposed. The number of species capable of being impregnated, even by skillful management, is very few ; and in nature the stigma 64 REVIEWS. exerts a specific action, which not only favors and quickens the operation of the pollen of its own sj)ecies, but resists and retards the action of that of another ; so that the artist has not only to forestall the natural operation, but to experience opposition to his conducting the artificial one. " 2. Even when the impregnation is once effected, very few seeds are produced ; still fewer of these ripen ; and fewest of all become healthy plants, capable of maintaining an inde- j)endent existence. " 3. The offspring of a hybrid has never yet been known to possess a character foreign to those of its parents ; but it blends those of each; — whence hybridization must be re- garded as a means of obliterating, not creating, species. " 4. The offspring of hybrids are almost invariably abso- lutely barren, nor do we know an authenticated instance of the second generation maturing its seeds. " 5. In the animal kingdom hybrids are still rarer in an artificial state, are all but unknown in a natural one, and are almost invariably barren." Perhaps some of these dicta are too unqualifiedly stated ; indeed they are manifestly intended to affirm the results to which the whole evidence points, rather than those which can be said to be thoroughly verified. The third proposition, however, is absolutely true ; and, in connection with it, well do our authors say, that all we could legitimately conclude is, that were hybrids of the general occurrence which some botanists imagine, they would long ago have obliterated all traces of species as definite creations ; whereas, exceptional in art, and not proven if not almost im- possible in nature, they cannot be assumed to have produced any appreciable result. There is one point, however, which our authors do not take into consideration, but which should not be overlooked, namely, what is generally admitted as a fact, that a hybrid may readily be fertilized by the pollen of either of its parents ; and that if hj^brid plants are occasion- ally produced in nature, they would ordinarily stand a very good chance of being fertilized in this way. In such cases they are said to revert to the type of the species of the im- HOOKER AND THOMSON'S INDIAN FLORA, 65 pregnating parent; but would tliey return exactly to that type, inheriting as they do a portion of the blood of a cognate species ? And where — as not unf requently occurs — two or more generally well-marked forms in nature are connected by certain occasional individuals of intermediate character, is it not very supposable that two species may have partially blended in this way ? At any rate there is a vera causa,, or what passes as such, which requires to be taken into account, as has not yet been done, so far as we know. This doubtless has oper- ated in the case of cultivated plants, and contributed, along with other causes, to the inextricable blending of certain species. But we are not disposed to exaggerate its influence in nature ; since we suppose, with Dr. Hooker, that wild plants rarely hybridize. Yet the possibility and even the probability of the occurrence must not be overlooked in a thorough discussion of the general question of the limitation and permanence of species. However it may be as a blending influence, hybridization is far from being a considerable, or the most potent cause of the variation of species, since " the offsj^ring of a hybrid has never yet been known to possess a character foreign to those of its parents." And we equally agree with our authors that the known facts of the case " especially warn us not to con- sider the influence of climate as paramount in determining the distribution of species or the prevalence of forms," or even as the most efficient cause of variation. What the cause is that the legitimate offspring does occasionally possess a character foreign to those of its parents we are wholly unable to say ; but the fact is undoubted, and perhaps of more fre- quent occurrence than is generally supposed. It is usual to say that the abnormal forms originate only in cultivated or domesticated individuals : it were perhaps better to say that they were perpetuated, or are favorably situated for continua- tion and full development, only under these circumstances, on account of the greater segregation ; for of the very various species of plants which are cultivated none are free from the tendency to "sport" into races, whether of ancient or of recent introduction. Why their existence is so transitory in 66 REVIEWS. nature, and so capable of being continued and further devel- oped in domestication, it is not difficult to imagine. Our authors perhaps, in common with naturalists generally, do not sufficiently recognize the natural tendency to perpetuation of individual characteristics. As regards ordinary variation between different individuals of the same species, the want of due consideration of what every good observer knows to be true, has indeed " mainly contributed to such an undue multiplication of species in the vegetable kingdom as botanists unfamiliar with large herbaria and exotic plants are slow to believe, and to the exaggerated estimates of the supposed known extent of the vegetable crea- tion that gain common credence." Our authors believe that the number is swelled one third beyond its due extent by the introduction of bad species founded on habit, and on acci- dental variations produced by soil, exposure, etc. ; and, we would add, on the imperfection of the materials from which the greater part of the species that crowd our books were originally described, most of them without due elaboration of already published species, and drawing after them an ever lengthening train of nominal species, founded on mere guesses at supposed differences from vague and incomplete descrip- tions, without any collection of specimens. We have already exceeded our limits, while yet at the beginning of Drs. Hooker and Thomson's interesting and suggestive volume. AYe regret that we must omit all notice of their remarks upon habit as indicating specific difference, which, contrary to the general view, they regard as " most deceptive," and must pass over their important section upon geographical distribution in general, and its dependence upon specific centres. We only add, that whoever would attain a clear comprehension of the configuration, the diverse climates, and the general botanical geography of those exten- sive and widely varied regions which are comprised, and in most minds confused, under the general name of India, has only to study the admirable sections on the meteorology of India, and on the physical features and vegetation of its provinces, which occupy a large portion of the Introductory DE CANDOLLE'S GEOGRAPHIE BOTANIQUE. 67 Essay. The present commencement of the Flora itself, although comprising only fifteen natural orders, is also an inviting subject for extended comment and almost unqualified commendation. DE CANDOLLE'S GEOGRAPHIE BOTANIQUE. The "Geographic Botanique" ^ of De CandoUe is not only one of the most important works of our day, but one which addresses, and will greatly interest, a much broader circle of scientific readers than any other modern production of a botanical author. It is, and probably long will be, the stand- ard treatise upon a wide class of questions, highly and almost equally interesting to the botanist, the zoologist, the geologist, the ethnologist, and the student of general terrestrial physics. To its production the author has devoted no small portion of the best years of his life ; and it bears throughout the marks of untiring labor, directed by a remarkably sound, conscien- tious, and thoroughly systematic mind. Along with the admi- rable methodical spirit which is his by rightful inheritance, the j^ounger De CandoUe brings to these investigations a par- ticular aptitude for numerical and exact forms, an intimate acquaintance with general physical science, and considerable ethnological and philological learning ; which last is turned to good account in his chapters on the history of cultivated and naturalized plants. The result in the work before us — even if there were no other claims to the distinction — may fairly be said to go far toward inscribing the name of De CandoUe anew in that select list of philosophical naturalists in which his father holds so eminent a position. To give some idea of the topics considered in these volumes, and of the order of investigation (which proceeds in an ad- mirable course, from the more simple, general, and better ^ Geographie Botanique raisonnc'e, ou Exposition des Faites principaux et des Lois concernant la Distribution Geographique des Plantes de VEurope ActueUe. Alphonse De CandoUe. Paris and Geneva, 1855. (American Journal of Science and Arts, 2 ser., xxii. 429.) 68 REVIEWS. known facts and principles towards the more complex, hypo- thetical and obscure), we will copy the titles of the chapters, twenty-seven in number ; which are arranged in four books, and subdivided into articles, and these again into sections, to such an extent as to fill eight closely printed pages with the bare enumeration. Indeed, this repeated subdivision gives a rigid and rather tedious asjDcct to some parts of the work, and involves occasional repetitions ; but it would not be easy to collocate well and clearly so vast an amount of material in any better way. The First Book is occupied with some preliminary consid- erations upon the way in which temperature, light, and mois- ture act upon plants. Its three chapters treat of the relations of plants to surrounding physical conditions, and especially to heat and light ; and contain the author's happy distinction be- tween the temperatures actually operative in vegetation, and those which (being below the freezing point, etc.) are alto- gether null for vegetation, and ought to be eliminated from the tables of mean teoiperature, when these are viewed in relation to the northern and southern geographical range of species. The Second Book is devoted to Geographical Botany, or the study of species, genera, and families, from a geographical point of view. Chapter iv. relates to the limitation of species upon plains and ujjon mountains, and the probable causes of their actual limits, applied both to spontaneous and cultivated plants ; and there is good endeavor to show that the northern limit of species is fixed rather by the sum of heat available for vegetation during the growing season, than by the mean tem- perature of the year. Chapter v. treats of the shape of the area occupied by a species, a very curious point ; and it seems that the area of species inclines to be circular or elliptical. Chapter vi. treats of the associations or disjunction of the in- dividuals of a species in its area. Chapter vii. treats of the area of species as to extent of surface, considered as to the families they belong to, as to stations, as to size and duration of the plant, and as to the character of the fruit and seed, whether affording facilities to dispersion or not. Chapter viii. DE CANDOLLE^S GEOGRAPHIE BOTANIQUE, 69 considers the changes which may have taken place in the hab- itation of species, and discusses with great fullness the whole subject of naturalization, the obstacles in the way, the causes and means of transport, and the interchanges which have been efeected between the New and the Old Worlds. Chapter ix. is a very long and interesting one, on the geographical origin of the principal cultivated plants, not only those intentionally, but also those unintentionally cultivated by man, — a chapter full of valuable matter, carefully collected and well discussed.^ Chapter x. treats of disjoined species, — those occupying two or more widely separated areas, and not in intermediate stations. Chapter xi. discourses of the early condition and probable origin of the existing species ; and brings out the various facts which go far to prove the geological antiquity of the greater part of existing species ; and that their creation was probably successive. Chapter xii. treats of genera and their geographical distribution, and maintains the view (in which we by no means coincide) that genera are truly natu- rally-limited groups, even more so than species. Chapter xiii. is devoted to the distribution of the species of a genus within its area. Chapter xiv. treats of the extent of surface occupied by genera. Chapter xv. discourses of the origin and duration of genera. Chapters xvi.-xix. treat of families, as to their area, geographical limits, the distribution of species within the area of the family, etc. The Third Book is devoted to Geographical Botany, or the characters of different countries considered as to their veg-eta- tion. Chapter xx., of the characters of the vegetation of a country ; considered, in Chapter xxi., as to the relative num- bers in the great classes respectively. Chapter xxii., compari- son of different countries in respect to those natural orders 1 It is singular that M. De Candolle should be so slow to abandon the idea that the aborigines of Carolina, or any other part of North America, cultivated or knew anything of the Potato, which, if Raleigh obtained them in Carolina, were certainly imported thither. But, though our aborigines had no Potatoes, they had Pumpkins or Squashes and Beans, which all writers upon the history of cultivated plants have overlooked except the late Dr. Harris. 70 REVIEWS. which abound most in species ; and Chapter xxiii., as regards their most characteristic natural families. Chapter xxiv., on the variety of vegetable forms in different countries and in the world at large, i. e., the probable number of species, the pro- portion of genera to species, and of orders of genera and species. Chapter xxv., the division of the earth's surface into natural botanical regions. Chapter xxvi., sketch of the vegetation of the different countries in respect to the probable origin of their existing species, etc. The Fourth Book, of a single brief chapter, consists merely of a summary of the author's general conclusions. We give these entire, for convenience availing ourselves of a translation in Hooker's " Journal of Botany." " The plants now inhabiting the globe have survived many changes, geological, geographical, and, latterly, historical. The history of their distribution is hence intimately connected with that of the whole vegetable kingdom. " To explain existing facts, it is fortunately unnecessary to adopt any conclusion upon the most obscure hypotheses of cosmogony and palaeontology, or on the mode of creation of species, the number originally created, and their primitive distribution. Botanical geography can indicate certain prob- abilities, certain theories, but the principal facts in distribu- tion depend upon more recent and less obscure causes. It suffices to understand and to allow certain facts and theories, which appear probable, namely, that groups of organized be- ings under different hereditary forms (classes, orders, genera, species, and races) have appeared in different places and at different times ; the more simple perhaps at first, the more complicated afterwards ; that each of these groups has had a primitive centre of creation of greater or less extent ; that they have, during the period of their existence, been able to become more rare or common, to spread more or less widely, according to the nature of the plants composing them, the means of propagation and diffusion they are possessed of, the absence or presence of animals noxious to them, the form and extent of the area they inhabit, the nature of the successive climates of each country, and the means of transport that the DE CANDOLLE'S GJ^OGRAPHIE BOTANIQUE. 71 relative positions of land and sea may afford ; that many of tliese groups have become extinct, whilst others have increased, at least so far as can be judged by comparing existing epochs with preceding ones ; and lastly, that the latest geological epoch, the Quaternary (that which preceded the existence of man in Europe, and which followed the last elevation of the Alps), has lasted many thousand years, during which impor- tant geographical and physical changes have affected Europe and some neighboring countries, whilst other regions of the globe have suffered no change, or have been exposed to a dif- ferent series of changes. " Thus the principal facts of geology and palaeontology, re- duced to the most general and incontestable, suffice to explain the facts of Botanical geography, or at least to indicate the nature of the explanation, which it requires the progress of many sciences to complete. *' The most numerous, the most important, and often the most anomalous facts in the existing distribution of plants, are explained by the operation of causes anterior to those now in operation, or by the joint operation of these and of still more ancient causes, sometimes of such as are primitive (con- nected with the earliest condition of the planet). The geo- graphical and physical oj^erations of our own epoch play but a secondary part. I have shown that in starting from an original fact, which it is impossible to understand, of the crea- tion of a certain form, in a certain country, and at a certain time, we ought to be able, and sometimes are able to explain the following facts, chiefly by causes that operated previous to our own epoch : 1, the very unequal areas occupied by natural orders, genera, and species ; 2, the disconnection of the areas that some of the species inhabit ; 3, the distribution of the species of a genus or family in the area occupied by the genus or family ; 4, the differences between the vegetations of coun- tries that have analogous climates and that are not far apart, and the resemblance between the vegetation of the countries that are apart, but between which an interchange of i)lants is now impossible. "The only phenomena explainable by existing circum- 72 REVIEWS. stances, are : 1, the limitation of species, and consequently of genera and families, in every country where they now appear ; 2, the distribution of the individuals of a species in the country it inhabits ; 3, the geographical origin and extension of cultivated species ; 4, the naturalization of species and the opposite phenomenon of their increasing rarity ; 5, the disap- pearance of species contemporaneous with man. " In all this we observe proofs of the greater influence of primitive causes, and of those anterior to our epoch ; but the growing activity of man is daily effacing these, and it is no small advantage of our progressing civilization that it enables us to collect a multitude of facts of which our successors will have no visible and tangible proof." An Appendix, indicating the researches now needed for the advancement of Geographical-botanical science, under several heads, addressed respectively to physicists and meteorologists, to geographers, to geologists, to vegetable physiologists, de- scriptive and traveling botanists, and to philologists, brings these most interesting volumes to a conclusion. Our present object is to call the attention of American nat- uralists and natural philosophers to this work, not to criti- cise it. That would require much consideration and a wider range of knowledge than we can pretend to. There are, how- ever, several topics upon which w^e are inclined to venture a few remarks, as fitting opportunities occur. HENFREY'S BOTANY.^ This is a well-planned, compact, and comprehensive work, in w^hich we may say, that the author has fairly accom- plished his purpose, namely : — "to produce a good working text-book for the student, from which may be obtained a groundwork of knowledge in all branches of the science, 1 An Elementary Course of Botany ; Structural, Physiological, and Sys- tematic ; icith a brief Outline of the Geographical and Geological Distribu- tion of Plants. By Arthur Henfrey. London, 1857. (American Journal of Science and Arts, 2 ser., xxiv. 434.) HENFRETS BOTANY. 73 without the attention being diverted from the more strik- ing features of the subject by details comparatively unim- portant." The work is divided into four parts. I. Morphology or Comparative Anatomy ; treating, in successive chapters, 1st, of General Morphology ; 2d, of the Phanerogamia, or the j^arts of Flowering plants and their modifications, and the laws which regulate them ; 3d, Morphology of the Cryptogamia. Part II. Systematic Botany ; treating, 1st, of the princi2)les of Classification; 2d, of systems of Classification, and 3d, Sys- tematic Descriptions of the natural orders, followed by an artificial analysis. Part III. Physiology ; comprising, 1st, the physiological Anatomy of plants ; 2d, general considera- tions on the Physiology of plants ; 3d, Physiology of Vege- tation; 4th, the reproduction of plants; 5th, Miscellaneous phenomena, under which are ranked the evolution of heat in plants, luminosity, and movements of plants. Part IV. Geo- graphical and Geological Botany, very summarily disposed of in about forty pages. It seems strange at first to interpose systematic botany between the morphological and the physiological ; but if the anatomy and physiology of plants are to be completely dis- joined from the study of the organs of the plant as a whole, the present arrangement is perhaps as good as any. It is adopted, as the preface shows, for the convenience of instructing medi- cal students, who compose the principal part of classes in Great Britain as well as on the Continent ; — for whom " one short course of lectures is devoted to this science, and three months is commonly all the time allotted to the teacher for laying the foundations and building the superstructure of a knowledge of botany in the minds of his pupils, very few of whom come prepared even with the most rudimentary acquaintance with the science." But the author remarks tliat " if the previous education of medical students prepared them, as it should, with an elementary knowdedge of the natural sciences, we should make physiology the most conspicuous feature of a course of botany in a medical school." While in England botany is scarcely an academical study. 74 REVIEWS, here it pertains to collegiate and academical instruction where it is taught at all. In Europe not even an apothecary can be licensed without passing an examination in botany ; in the whole United States, we believe, it forms no part, at least no regular part, of the medical curriculum ; no medical school has a botanical chair ; and no knowledge whatever of the science of the vegetable kingdom, which supplies the greater part of the jnateria medica^ is required for the degree of Doctor in Medicine ! Professor Henfrey is chiefly known, and most highly es- teemed, as a vegetable anatomist. Upon this subject he may speak with an authority which as a systematist, or even as a morphologist, he w^ould not pretend to. We shall offer no apology, therefore, for making an occasional criticism, and for pointing out several errors in matters of detail. These are not intended to disparage the work, for if we had not formed a high opinion of it on the whole, w^e should not take this trouble. As respects the first point noticed, our author, if wTong, is not alone. Still, we hardly expected him to teach that the radicle of the embryo is the true root ; and we cannot let pass unchallenged his reiterated statement that in Monocoty- ledons, the radicle, or its inferior extremit}^, is never devel- oped into a root in germination, but is abortive (pp. 14, 16, 18, 391, 537). Any one who will examine the germination of the seed of an Iris, an Onion, or even of a grain of Indian Corn, cannot fail to perceive that a primary root is developed, and that this is a direct prolongation of the extremity of the radicle. This, indeed, does not continue as a tap-root; neither does it in a great many Dicotyledons. In Squashes, Pumpkins, etc., there is no one primary root, but a cluster of rootlets from the first, all springing from the base of the stout radicle. In fact, this distinction between Monocotyledons and Dicotyledons is null. A character of certain monocot}^- ledonous embryos, neither strictly peculiar to the class, nor by any means universal in it, should not be assumed as distinctive. As to the morphology of the radicle itself, we suppose that the germination of any of the larger Cucurhitacece, or of a bean. HENFRETS BOTANY. 75 would suffice to convince any observer that the radicle is simply the first internode of the stem, giving birth to the primary root from its inferior extremity, usually, — and indeed, from the exceptional cases where it does not we should draw additional proof of its cauline nature. In fact, we know of no char- acter in which a root differs from an internode of a stem in which it does not also differ from the radicle, excepting its tendency to direct its inferior extremity downwards. Again, should the statement, that " the radicle of a monocotyledonous embryo is never developed " be held to mean that the radicle never lengthens, we remark, no more does it in the Pea and some other hypogaeous Dicotyledons ; and we are not quite sure that the statement is absolutely true of all Monocoty- ledons. Koot-hairs or fihrillm are mentioned (p. 19) as " often " occurring on young roots. Do they not always occur ? Surely it cannot be true that : " the branches of the axial root are originally growths from the apex of the root thrown off to the side," (p. 538). By some slip of the pen, Myrica Gale is adduced as an instance of whorled leaves (p. 45). On p. 49 the expression " over the petiole," instead of above or within it, would lead to a misconception. Something more might be said about the tendrils of Cucur- hitacece (which, besides, are not always single) ; but are the students of King's College really taught that, " tendrils of the vine are metamorphosed flowering branches arising in the axils of the leaves " ? (p. 62.) •' In all seeds except in those of the few orders which present an incomplete or acotyledonous embryo, we do not find the young plant possessed ... of a plumule " (j>. QQ^. Even some much developed embryos, such as those of Maple and Morning Glory, do not show the plumule until after the full development of the cotyledons. It may be said, indeed, that the plumule is hi posse when not in esse, but so it is no less in the cases excepted from the statement. Very singular is the statement (on p. 68) that in England " the terminal bud of the Lilac is generally killed by the frost in the winter ; " since in our much colder winter it is as 76 REVIEWS. completely hardy as tlie other buds whenever it happens to be formed, and, like them, is well developed before summer is over. As a general rule here, and we presume in England also, no terminal winter bud appears during the growing season, and so there is none to be killed by the frost of the following winter. The deeply alveolate receptacle of the Cotton-Thistle is figured (on p. 78) as an illustration of a ^^aZeaceows re- ceptacle. Truly terminal flowers are said to be rare (p. 86) : we do not quite understand this. The interesting questions relating to the phyllotaxy and symmetry of the flower are clearly stated, but no new light is brought to bear upon them, — nor all of the old. The opposi- tion of the stamens of Rhamnacece to the petals is, as usual, attributed to the probable suppression of an outer stamineal circle, although there is nothing in the blossom (as there is in Geraniacece^ etc.) to base the supposition upon. And our author has overlooked the most natural of explanations for this and strictly like cases, the one moreover which tells directly against the doctrine of transverse chorisis, — namely, that in these cases of ante-position there is a return to normal phyllotaxy, i. e., to the superposition of the corres- ponding elements of successive whorls, — a view first sug- gested, we believe, by Lestibudois. " Eeal cases of collateral multiplication may probably be explained by comparison of a primary staminal leaf with an ordinary compound stem-leaf, and supposing the filament to subdivide like the petiole does [sic] in such cases." This is certainly the way we regard it ; and as respects the application of this hypothesis to the stamens of Cruciferca^ we do not see what argument Megacarpcea "polyandra brings against it ; as the increase in the number of stamens is quite as explicable upon this as upon the ordinary theory. Indeed, our author's view that the glands represent suppressed stamens would seem to be negatived by this very case, since the glands have not dis- appeared with the increase of the stamens, but the contrary. The abnormal fertile flowers of Viola and Impatiens are HENFRETS BOTANY. 77 not " achlamydeous," as our author states them to be (p. 90) ; generally they are not even apetalous. In the botanical sense of the word, and as it is employed in the same sentence (p. 93), the petals of the vine cannot be said " to cohere above." The valvate petals are merely cadu- cous for the most part before expanding, just as is more de- cidedly the case in many Araliacece. In passing, we remark that a valvate aestivation of the corolla in the latter is much less distinctive than our author supposes (p. 311). Aralia itself has the petals imbricated in the bud. It is becoming common to regard the tube of a so-called superior calyx as a cup-like receptacle ; and there appears to be reason for it in CactacecB and some other cases. Professor Henfrey would seem to apply this view universally ; " for ex- ample, in Rosacem^ Umhelliferoe^ Cuciirhitacece^ Compositcc^'^ [!] etc. But if applied to Rosa, why not to the Sangidsorheoe and to other Bosacece with a calyx-tube lined with a disk bearing the stamens, etc. ? And is the cup a receptacle in those Melastomacece which have an adnate ovary, but a calyx when the ovary is free ? And how is it when the ovary and cup cohere only by the nerves of the latter ? For palece Professor Henfrey coins an English word, " pales " (p. 110), of which the singular would probably be " pale." AVe would propose to call them " palets." There are convincing reasons why the perigynium of Carex cannot be regarded as a perianth, as our author takes it to be (p. 111). It is not correct to say that the false dissepiments of Datura are formed " while the seeds are ripening " (p. 124) ; they equally exist in the ovary. And we doubt if the transverse false septa in Cathartocarpus and other Leguminosce are " placental developments." We are pleased to find that our author prefers to consider placentse as belonging to the carpels rather than to the axis, although the close of paragraph 226 appears to imply the contrary. We cannot agree that, "externally the campylotropous ovule resembles the anatropous, except that there is no 78 BE VIEWS. rhai^lie " (p. 130). No attentive student could fail to recog- nize the difference, especially in the families cited ( Cruciferce and Caryoj)hyllacece). Eipening must be regarded in a remarkably broad sense when it is stated with emphasis, " that the distinction between endocarps and epicarps, in the common stone-fruits, arises entirely during the ripening of fruit." Also : " it is well known that the easy separation of the pulj) from the stone is a sign of ripeness." When are cling-stone peaches ripe ? Again : " In Taxus . . . during the ripening of the seed a succulent cup-like envelope grows up around it" (p. 136). Is ripening synonymous wdth the formation and growth, as well as the maturing of the fruit ? Lindley's system of the classification and nomenclature of fruits is adopted, with some modifications. It is well to have such a system, as an analysis of the diversities of structure ; but of the thirty-six kinds so carefully defined and named only fifteen or sixteen are ever used in descriptive botany, or ever will be, it is devoutly hoped. There is much inconven- ience in practice, and little advantage in designating every possible modification of the same organ or set of organs by a distinct substantive name, or in distinguishing by separate technical names fruits formed of a simj^le ovary from those of a compound ovary, or fruits with an adherent from those with a free ovary. Why not call the gooseberry and the grape equally a berry, instead of restricting this name to the former and naming the latter a nuculanium ; and why name the pod of an Iris a dii^lotegia^ while that of a Lily is called a capsule f And while we term the pod of Saxifraga steU laris a capsule^ and that of S. tridactylites a diplotegia, what name are we to apply to that of S. aizoides, which is only half-superior ? Probably a wrong example is adduced on p. 148, for we cannot believe that any species of Ranunculus has the rhaphe averse from the placenta in the ripe fruit. By an oversight, on the same page, the fruits of Lahiatm are spoken of as seeds. As respects the systematic part, the chapters on the prin- HENFRETS BOTANY. 79 ciples of classification, nomenclature, etc., strike us as sound and good throughout ; and in the account of the natural or- ders a great amount of information, such as the medical stu- dent needs, is given in a comparatively small space. Errors or misconceptions will necessarily occur in the compilation of such an amount of materials, treating of structure, affinities, distribution, sensible properties, and medicinal or economical uses. They are not more numerous than was to be ex- pected, and we are not disposed to make them the subject of criticism. We may remark, in passing, that, as respects the morphol- ogy of the androecium in Fumariacece^ the name of the writer of the present notice is referred to, by some misconception, as adopting Lindley's well-known hypothesis of the splitting of two stamens into halves ; whereas he has maintained a very different view. And then this is mentioned as " offer- ing a phenomenon of chorisis," which in that view is quite incomprehensible to us. We w^ere surprised at the statement that the bark and leaves of Hamamdis, Virr/mica " are astringent and contain an acrid volatile oil " (p. 207). We trace it back to Lindley's "Vegetable Kingdom" (p. 784), and find, " The kernels of Hamamelis Virginica are oily and eatable. The leaves and bark are very astringent, and also contain a peculiar acrid essential oil ; " and this, we find, comes from Endlicher's " Enchiridion." How did this bland and inert plant acquire such a reputation ? Dr. Barton, who has figured it, says nothing of its possessing any sensible properties or useful qualities at all, except its use for divining-rods ; nor do Pursh, Bigelow, Elliott, Darlington, etc., allude to any pop- ular reputation of such qualities. No sign of any essential oil is to be detected in the foliage, and prolonged mastication of the leaves and bark while we write yields not the slightest trace of acridity and hardly any of astringency ; no more, certainly, than a Beech leaf. We never heard of the seeds being eaten ; and as they are " about the size of a grain of barley," or not much larger, and have a thick bony coat, they are not likely to become an important article of diet. After 80 REVIEWS. some search, we find tlie source of these extraordmary state- ments in the " Medical Flora " of the eccentric Rafinesque. He says the seeds are called Pistachio nuts in the Southern States, are rather oily and palatable, etc., but he neglects to mention their size. He adds, " The bark and leaves are some- what bitter, very astringent, leaving a sweetish pungent taste. The smell is not unpleasant. It has not been analyzed as yet, but probably contains tannin, a marine, extractive, and an essential oil." To all this, Endlicher, on the strength of " the sweetish pungent taste," has added the acridity ; and so one of the blandest and most useless of shrubs gets a world-wide and wholly factitious reputation for active medical qualities and esculent seeds ; and even Dr. Griffith, who must have known the shrub, has been induced to give it a place in his " Medical Botany." Our remaining remark relates to the random way in which mere analogies are mixed up with affinities in estimating or expressing the relationship of orders, etc., in this as in some other more notable works. It is, or at least ought to be, well understood, that mere analogy, i. c, likeness in some one re- spect only, however striking the imitation, is no indication of relationship, but that relationship rests upon affinity, i. e., upon agreement or similarity in the whole plan of structure, and especially of floral structure, whether general or particular, as the case may be. To speak, therefore, of " evident " and "most distinct" affinities between Coniferce and Lycopo- d'lacem is an example of this prevalent misconception of what affinity is. This is more intelligible, however, than the " ap- proach " suggested of Aquifoliacece to Loganiacem and Apo- cynacea?^ while their resemblance to CdastracecB is thought to be of small account ; or that of UmhelUferce to I^uhiacece, Saxifragacece, and even to Geraniacece, to which the resem- blances do indeed " seem rather superficial." Again, jLcm- thoxylacece (i. e., Rutaceai) are said to have considerable affinity to Oleacece^ because Ptelea, in the former, has a samaroid fruit, as has Fraxinus in the latter. May we add, as quite as much to the purpose, that the common Xanthoxy- lums have pinnate leaves, and are popularly called Prickly Ash? HENFREY'S BOTANY. 81 The study of affinities is neither guess-work nor divination, but a matter of logical deduction from structure, based upon scientific principles, — principles recognized and acted upon by sound botanists with considerable unanimity, although they have never been reduced to a system, nor expounded in detail, so as to make them matters of elementary instruction. Until this desideratum is supplied, the young botanist can do no better than to take as models the writings of Brown, and of those botanists who, according to their ability, have most closely followed the footsteps of this master in science. Having continued this review far beyond our intention at the outset, we have small space left for noticing the best part of Professor Henfrey's treatise, namely, the third or Physio- logical part. Suffice it to say, that, in the important chapter on the physiological anatomy of plants, our author w^rites from the fullness of his acquaintance with the writings and doings of all the continental phytotomists, and also with the authority of an experienced original investigator. And, so far as we know, it comprises much the best resume of vege- table anatomy and development now extant in the English language, at once succinct, clear, trustworthy, and well brought up to the present state of the science. Perhaps the succeed- ing chapters, on the Physiology of Plants generally, the Phys- iology of Vegetation, and on Reproduction, are equally com- mendable in their way ; but we have as yet barely glanced over the pages. We like the following definition, and the ensuing paragraph upon the role of vitality in plants. " The physiology of plants is that department of botany in which we investigate the phenomena of the life of plants, manifested in a series of changes taking place in the diverse parts of which each plant is composed " — (p. 475). " The physiological phenomena which indicate vitality are always of more or less complex nature, and admit of being analyzed into a number of factors, of which a large propor- tion are found to be purely physical or chemical. A very considerable part of the changes which accompany the process of organization are the results of the action of physical and chemical forces, [and] capable of being explained up to a 82 REVIEWS. certain point, by the known laws of those forces. But in every case, after referring* all the chemical and physical phenomena to their respective places, there remains a residual phenomenon to be accounted for, which is precisely the most important of all, — namely, that in living organic structures . . . the laws of inorganic matter are subdued under a higher influence, and caused to undergo modifications never occur- ring except in the presence of living matter , while — most important of all — the peculiar compounds of matter thus produced are not only made to assume forms, according to definite laws, totally unlike any forms of mineral matter, but [to] constitute bodies manifesting a continued interchange of material with the surrounding media, which, instead of re- sulting in decomposition, as in mineral bodies, effects a repro- duction and increase of the already existing [organized] matter"— (p. 542). In the paragraph on the longevity of trees (p. 549), we find renewed occasion to notice the longevity of unfounded statements, copied from one book into another long after the error has been pointed out. Here again the Adansonia of Senegal and the Wellingtonia or Sequoia of California figure as trees " whose age, deduced from the rings of growth of the stems, would amount to upwards of 3000 years." There is really no evidence to prove that the famous Baobabs de- scribed by Adanson are of such an age ; and as to the Wel- lingtonia in question, an actual counting of the rings has shown that the tree was not half so old as it was vaguely computed to be. The chapter on Reproduction appears to be excellent, as indeed w^e should expect. The geographical and geological part is necessarily very briefly treated. NAUDIN ON THE GENUS CUCURBIT A. 83 NAUDIN ON THE GENUS CUCURBITA. Naudin's "Researches into the Specific Characters and the Varieties of the Genus Cucurbita " i are published in the 6th volume (4th series) of the "Annalesdes Sciences Naturelles," and are of no small interest, being founded upon a very con- scientious investigation of nearly all the known forms, collected for the purpose, and cultivated under the author's eye at the Jardin des Plantes. These forms our author reduces to six species, and the alimentary sorts in cultivation to three, namely, Cucurbita maxima, Q, Pepo, and C. moschata. The remain' ing three species are C. melanosperma of Braun, newly in- troduced from eastern Asia, and the two perennial and tu- berous-rooted species, a peremiis and C, digitata, Gray, na- tives of our southwestern borders, the fruits of which are not esculent. Indeed, the Pumpkins and Squashes cultivated in Northern Europe, and with us, as now understood, belong to only two species, since the third, C. moschata, hardly comes to perfection north of the Mediterranean region. Of these, C. maxima is made to include C. Meloiiepo ; and C, Pepo, comprising our Pumpkins and a large part of our Squashes, IS made to include C. ovifera, aurantia, verrucosa, etc., and the species are defined by botanical characters, which ap- parently may be relied upon. The varieties of C. maxima fall into two main groups, characterized by their fruits, namely, the " Turbans," having crowned fruits, that is, the summit projecting beyond the adnate calyx-tube, a peculiar- ity found in no other species, and the crownless sorts, in which this peculiarity is not manifest. The innumerable varieties of C. Pepo are arranged in seven groups, according to the configuration of their fruits. M. Naudin has not undertaken to discuss the questions respecting the birth^Dlace of these plants. He remarks that C. maxima and C, moschata have been known in European gardens scarcely above two centuries ; but that C. Pcpo was perhaps known to the Greeks and the Romans in the time of Pliny. ^ American Journal of Science and Arts, 2 ser., xxiv. 440. 84 REVIEWS. The younger De CandoUe, in his discussion of the history and origin of the principal cultivated plants, which forms a most interesting chapter of his " Geographic Botanique," although he is unable to assign them to any country as their home, confidently (perhaps too confidently) refers all the squashes and pumpkins to the Old World ; but not to India, because they have no Sanscrit name. He will not believe that any of them came from America, and appears to think little of the current statements that squashes or pumpkins were in cultivation by our aborigines before the European settlement of the country. On the other hand, our lamented Dr. Harris — who, during the later years of his life, assid- uously studied this question, and who was very cautious in drawino- conclusions — had become satisfied that the North American Indians, as far north even as to Canada, cultivated squashes and pumpkins, one or both, along with their maize, before the whites were established here. We are unable at this moment to refer to his manuscripts, or to what he had too imperfectly published upon this subject. But we well remember his laying much stress upon the narrative of Champlain ; and with good reason, as it appears to us on turning casually to the pages of " Les Voyages du Sieur de Champlain . . . ou Journal tres-fidele des Observations faites et Decouvertes de la Nouvelle France," etc., etc., edition of Jean Berjon, Paris, 1613, 4to ; also " Voyages et Decouvertes faites en la Nouvelle France depuis I'annee 1615, jusques a la fin de I'annee 1618," — second edition, published by Collet in 1627, small 12mo, — to which volumes we desire to direct M. De Candolle's attention. In Champlain's narrative of his own voyage along the coast of what is now the State of Maine, in the year 1604, and the two voyages of Le Sieur de Mons along the coast of New England in 1605 and 1606, Citrouilles and Coiirges are rejoeatedly mentioned, along with maize (^Bled cVInde) and beans ; e. g. : " Nous y vismes force citrouilles, courges & petum, qu'ils cultiuet aussi. . . . Pour les febues elles come^oiet a entrer en fleur, come fay soy et les courges et citrouilles " (p. 68). " Ceux que nous anions enuoyes deuers eux, nous appor- NAUDIN ON THE GENUS CUCURBIT A. 85 terent des petites citrouilles de la grosseur du poing, que nous mangeasmes en sallade comme coucombres, qui sont tres- bonnes " (p. 77). See also j^p. 83, 115, 116. Of course it does not follow that these esculents were natives of New England, any more than maize ; but both may probably have been carried north- ward together. Whatever their origin, our Indians were found cultivating them together at this early date as well as in later times. According to Nuttall, the Indians along the whole Upper Missouri half a century ago were cultivating Ciicurhita verrucosa. This common squash is, according to Naudin, a variety of C. Pepo^ as also is C. aurantia (the C. Texana vel ovifera., Gray, "PI. Lindheimerianse"), which has every appearance of being indigenous in the western part of Texas, on the Rio Colorado and its upper tributaries. At least, this is the opinion of Mr. Lindheimer and of Mr. Charles Wright, two good judges. The latter personally informs us that, from the stations and localities in which alone it is met with, he could not suspect it to be other than an indigenous plant. That the later Greeks and Eomans possessed the bottle gourd or Lagenaria, and also some kind of summer squash, seems pretty clear ; but we see no decisive reason for the opinion that they had any form of Cucarhita Pepo^ as that species is now understood. According to De CandoUe, the earliest figures referable to this species are, one of C ovifera by Lobel in 1576, and one of C. verrucosa by Dalechamp in 1587, namely, about a century after the discovery of America, and long after maize had become well known in the south of Europe ; and we have seen that some forms probably of this very species (undoubtedly originating in a warmer region) had by this time found their way in this country nearly as far north as the climate will permit of their cultivation. So that there appears to be about the same evidence for the American origin of some squashes and of pumpkins that there is for the American origin of maize. A remaining argument brought by De Candolle against this view may also be turned the other way, namely, that no 86 REVIEWS. certain species of the genus is known as indigenous to Amer- ica. He has equally allowed that none is known to be in- digenous to the Old World. Now of the six species recog- nized by Naudin, two only are known in their natural wild state, and these are our southwestern species with perennial roots, namely, C ^9ere?i7iis and C. digitata, to which we add that (7. Pe])o itself (i. e., (7. ovifera or aurantia) grows wild in the same district with C.2^€rennis^ and has the same appear- ance of being indigenous there. We leave the subject with these incidental remarks, as we did not intend here to investi- gate this question, and will briefly allude to another subject, upon which Naudin's investigations have thrown new light. It is generally thought that the cultivated Cucurhitacece^ and especially that the species of Cucurbita, cross-breed with extreme facility. According to Naudin this is true of the races only inter se. A good illustration of the immediate and great variation from this cause in the fruit of C. Pepo is given in Naudin's third plate, where fifteen different forms of the fruit are figured, taken from as many individual plants raised from seeds of one fruit, which had grown in the vicinity of other varieties. It is by no means certain, however, that all these forms originated from direct crossing. But the species themselves strangely refuse to hybridize. Naudin carefully experimented with the five species in cultivation at the Jardin des Planus (namely, all known, except C. digitata) ; and out of seventy distinct trials all but five were utterly ineffectual. In five instances the fruit set, indeed, but in none of these was a single seed containing the vestige of an embryo pro- duced ! What are we to think, then, of the universal belief that squashes are spoiled by pumpkins grown in their vicinity, or pumpkins by squashes ; and even melons (which are of a different genus) by squashes? The fact of some such in- fluence seems to be well authenticated. Dr. Darlington, one of the most trustworthy of observers, speaks of it from his own knowledge, thus : " When growing in the vicinity of squashes the fruit [of the pumpkin] is liable to be converted into a kind of hybrid, of little or no value. I have had a crop of pumpkins totally spoiled by that cause, the fruit be- WEDDELUS MONOGRAPH OF URTICACE.E. 87 coming very hard and warty, unfit for the table and unsafe to give to cattle." — ("Fl. Cestrica," ed. 2, p. 555.) Now that this is not the effect of hybridation is clear from the fact that the result appears in the fruit of the season, not in that of the next year, namely, in a generation origi- nated by the crossing. A clue is perhaps furnished by Nau- din's observations, that the ovary is apt to set and even de- velop into a fruit in consequence of the application of the pollen of another species, although, as the result proves, none of the ovules are fertilized. And he hazards the conjecture that the pollen may exert a specific influence first upon the ovary, inciting its farther development, and then upon the ovules. To test this conjecture he was to examine the action, if any there be, of the pollen of Cucurbita upon the ovary of melons. The past summer — which has been as unusually warm in western Europe as it has been cool in this country — must have favored such researches in Paris ; and we may expect soon to hear of the result. Improbable as such an influence seems to be, it is hardly more so than the now authenticated fact that the graft of a variegated variety of a shrub or tree will slowly infect the stock, so that the varie- gation will at length break out in the foliage of the natural branches ; — an old observation, which, according to the Gard- ner's Chronicle, has recently been verified in several instances. WEDDELL'S MONOGRAPH OF URTICACE^. Dr. Weddell's preliminary studies upon the proper Urti- cacece were published a few years ago in the " Annales des Sciences Naturelles." Since then, botanists, aware from this and his other works that the subject was in most able hands, have been anxiously waiting for his full monograph. This, we understand, is now completed, although the last fasciculus has not yet reached this country. The greater part is before us, and an admirable monograph ^ it is, worthy of a place in 1 Monographie de la Famille des Urticees. Par H. A. Weddell {Archives du Museum, ix., livr. 1-4), 185G-57. (American Journal of Science and Arts, 2 ser., xxv. 109.) 88 REVIEWS. the "Arcliives" wliicli contain that model one on the 3fal- 2)igJilaccce of his lamented botanical master. It illustrates in detail about 470 species, under 40 genera, and is accompanied by tNventy well-filled plates, drawn by the author. It opens with a conspectus of the members of the great group to which the true Urticacece belong (which the author inclines to receive rather as the orders of a class than as suborders of an exten- sive order, fully admitting, however, their close affinity inter se), followed by a brief indication of the principal investiga- tors of these plants, and of the resources at his own command. A general account of the organs of vegetation and reproduc- tion, of the affinities, and the geographical distribution of the plants of the group, and of their properties and uses, conclude the preliminary matter. The body of the work is occupied by their systematic arrangement and description. Apetalce being viewed as degenerations of Polypetalm., our author searches anion o- the latter orders for the nearest rela- tives of the great Urticaceous order of alliance, and finds them in the Tiliacece., that is, in the group of orders of which the 3Ialvacem are the highest development. According to AYeddell's happy illustration, Malvacece crown the summit of a three-sided pyramid, with Sterculiacece., Byttneriacece^ and Tiliacem just below them, one upon each face ; under the Byttneriacem he ranks the Eupliorhiaceos, with the Antides- meoB, and under these, at the very base of the pyramid, the Scepacece, the lowest degradation in this direction of the Mal- vaceous type. On the adjacent face, under the Tiliacece^ and on the same level with the £Jupho7'biacece, he inscribes the Urticacece, with the Cuindiferm perhaps underneath them. Upon this ingenious plan of representation, the apetalous orders throughout may be most conveniently and instructively ranked under their superior types ; — bearing in mind that some types degrade as much within an order (e. g., Euphor- biacece, Onagracece inclusive of Haloragece, Caryoplnjllacem including lllecehrecB), as others do through a series of two or three orders, or even as the same group does {e. g., Caryo- phyllaceai) through a series of orders on the other side of the pyramid. WEDDELUS MONOGRAPH OF URTICACEJ^. 89 The reason why this mode of rej^resentation will exhibit botanical affinities so well is, that (as we have elsewhere remarked) the vegetable kingdom does not culminate, — as the animal kingdom does, — and therefore offers no founda- tion in nature for a lineal arrano'ement even of its jrreat groups. But it would appear that the Dicotyledonous orders might be arranged under a considerable number of short series, in groups converging upon the most fully developed or representative order of each type, so as to exhibit what w^e now know of the system of nature much better than in any other way. We think that Dr. Weddell's idea of the affinity of Urtl- cacece is a good one. The floral and seminal characters, the true criteria of affinity, are not abhorrent, but present some strong points of relationship, as do the organs of vegetation. These, once established, allow us to feel the force of the strik- ing coincidence in the bast-tissue of the bark, so remarkable in all this alliance for the length, fineness, and toughness of the fibres, their union end to end, and their lateral indepen- dence, admirably adapting them for their use as textile mate- rials, in which Urticacece vie with Malvacece and Tiliacece. As to geographical distribution, Europe is very poor in Urticacece^ poorer even than would at first view be supposed, as the author remarks. For as nettles like an enriched soil, the five or six European species of Urtica and Parietaria so abound around habitations that they make up in the multitude of individuals for the paucity of species, and perhaps cover nearly as much ground as the great number of intertropical species ; two or three excepted, which are also weeds in the tropics. Temperate North America is not much richer in species than Europe. The greater part are found in the torrid zone, and in islands rather than continents ; the Malay region, India, Mexico, and the West Indies together possess almost two thirds of the known species. Our remaining remarks shall be restricted to one well- known plant described in the work, and to another of recent discovery, which unfortunately was not communicated in sea- son to find a place in it. 90 REVIEWS. The first is our common Pllea pumila. Dr. "Weddell has overlooked the fact that Kafinesque had founded a genus (Adice or Adike) upon it, although the name is mentioned in the work cited by him, where the plant was first published as a Pilea, and although Dr. Torrey had adopted Rafinesque's genus, and figured the species, in an earlier and more consid- erable work (" Flora of the State of New York "), which, having unfortunately been published by the State, and in a large edition, has in consequence remained almost unknown to science. Considering that the three sepals of the fertile flower in this species are nearly equal and not gibbous, it may be doubted whether the single species of Blume's genus Achu- demia, differing only in having five sepals, should not rather be appended to Pilea. We dare say that Dr. Weddell would have so arranged it, if Blume had not published the genus. Since the appearance of the third part of Weddell's mono- graph, but before it had reached this country. Dr. Torrey has published, in the report on Dr. Bigelow's fine California col- lection made in Lieutenant Whipple's Railroad Survey to the Pacific, a new Nettle allied to Boehmeria but with the peni- cillate stigma of Urtica, namely, his Hesj)eroenide tenella (" Pacific Railroad Reports," iv. p. 139). This little plant, it now appears, comes nearest to Wight's monotypic genus Cha- mahaina of India, of which better details than Wight's as to the female flowers and fruit are figured in the present mono- graph. The stigma is intermediate in character between that of Chamabaina and that of Urtica ; and, moreover, as the se- pals of the male flower want the pointed gibbous tips of the former, the stipules are inconspicuous, and the cotyledons are not only reniform but (which is unnoticed in the published description) pretty strongly emarginate at the summit also, the genus will probably be retained. Great thanks are due to Dr. Weddell for his labors upon this family, which he found in a most unsatisfactory and diffi- cult state, and has left in such condition that Nettles and their allies are easy and inviting objects of study. FECUNDATION IN THE VEGETABLE KINGDOM. 91 RADLKOFER'S PROCESS OF FECUNDATION IN THE VEGETABLE KINGDOM.^ This gives in English, and in an accessible form, a sys- tematic and historical survey of the whole subject of vegetable fecundation, including the recent discoveries of Pringsheim, Colin, Braun, and Bary. As to Fungi and Lichens^ — thanks to the observations of Itzigsohn upon the latter, and the most useful and persevering investigations of Tulasne upon both families, — the analogues of male organs in all probability are discovered, and their general presence recognized ; but the fact of fecundation is not made out. In the lower or green Algce, fecundation was first demon- strated by Pringsheim. The " horns " of Vaucheria which Yaucher half a century ago observed and conjectured to be male organs, Pringsheim proved to be so, having seen them open at the summit and emit a great number of free-moving corpuscles (spermatozoids), many of which found their way into the now open orifice of the ^protuberance, which contains the forming spore, and were seen crowding against it, after which a membrane of cellulose appears over the surface of the mass of protoplasm and completes the spore. Whether one or more of the spermatozoids actually penetrates the protoplasm and so is included within the cell-membrane is uncertain; but Pringsheim thought it was the case, from having detected a colorless corpuscle like one of the sperma- tozoids inside of the membrane. Next Pringsheim demon- strated a similar fecundation in CEdogonium. His results, briefly published in the proceedings of the Berlin Academy, and thence translated into French and English, are now given in detail in the first part of his " Jahrbiicher," noticed above. CEdogonium consists of a row of cylindrical cells. 1 Der Befruchtungsprocess im PJlanzenreiche. L. Radlkofer. Leipsic, 1857. (English translation by Arthur Henfrey in Annates and Magazine of Natural History, October and November, 1857.) (American Journal of Science and Arts, 2 ser., xxv. 112.) 92 REVIEWS. Some of these cells, usually shorter than the rest, become tumid, and, without conjugation, have their whole green con- tents transformed into a large spore. Pringsheim has ascer- tained that other cells of the same individual plant have their green contents transformed into a multitude of active corpuscles or zoospores, which, from their subsequent evolu- tion and office, he names androspores ; these escape by the opening of the mother cell moving about freely by the vibra- tion of a crown of cilia attached near the smaller end. One or more of these androspores fix themselves by the smaller end upon the surface of the cell in which a large ordinary spore is forming, or in the vicinity, and germinate there, growing longer and narrower at the point of attachment, while near the free end a cross partition forms, and some- times another, making one or two small cells ; this is the true antheridium : for in it a crowd of spermatozoids are formed, also endowed with motivity by means of vibratile cilia. Now the top of the antheridium falls off as a lid, the spermatozoids escape ; the spore-cells at this time open at the top ; one of the spermatozoids enters the opening, its pointed end fore- most; this becomes stationary upon or slightly penetrates the surface of the young spore, into which its contents are doubtless transferred, and a coat of cellulose is then, and not until then, deposited upon it, completing its organization as a spore, which in due time germinates, and grows directly into a plant like the parent. But in Bulbochaete, and especially in Sphseroplea, so beauti- fully investigated by Cohn ("Annales des Sciences Natu- relles," 4 ser.,v.), the spore does not directly develop into the normal or fruit-bearing plant. Instead of this, by an alterna- tion of generations (to adopt that well understood phrase), the spore proceeds to convert its contents by successive division into a large number of zoospores, different from the andro- spores, namely, small oval or oblong bodies, furnished with two long cilia on a short beak at one end, and for a time mov- ing actively about by their vibration. Coming to rest, these zoospores germinate, by elongation and the formation of trans- verse partitions, into adult thread-like plants, consisting of a FECUNDATION IN THE VEGETABLE KINGDOM. 93 row of cells. In Spbseroplea the whole contents of the cells of some adult individuals condense into large green spores, as yet without a coat ; while those of different individuals give rise to myriads of slender spermatozoids, moving* by means of a pair of cilia fixed at the narrow end. The latter escape from the parent cell through a small perforation which now appears, enter the spore-bearing cells of the fertile plant through a similar perforation in them, play around the spores, and at length one or more of them drives its pointed extremity into their naked surface ; after which, fertilization being accomplished, a thick coat of cellulose is deposited to complete the spore. " Cohn does not consider that observa- tions justify his assuming a direct penetration of the sperma- tozoids into the primordial spore-cell. It rather seemed to him as if they attached themselves on the outside of the spore, and were finally converted into mucilaginous glo- bules." Reproduction by conjugation of course had long been familiarly known in the lower Algce. But it was questioned whether this was really analogous to sexual reproduction, since what appeared to be similar spores are often formed of the contents of a single cell without conjugation. Areschoug shows that these are abortive spores, incapable of germina- tion ; while those which result from actual conjugation will grow into new plants, without further metamorphosis ; Vau- cher's old observations to this effect having been confirmed by Braun and Pringsheim. That in the Fucacece or olive-green AlgcB the large spores are fecundated by spermatozoids, produced in antheridia, was demonstrated by Thuret in the year 1850. And in more recent memoirs he has shown that the fertilization takes place through direct contact of the spermatozoids with the naked surface of the unimpregnated spore, then having only a protoplasmic coating ; and that these spores will not de- velop nor hardly acquire a cell-wall unless so fertilized. His experiments upon dioecious species are j^erfectly decisive upon these points. He observed the lively spermatozoids playing over the surface of the still naked spore, fix them- 94 REVIEWS. selves to it by the ciliated end, apparently by one of the cilia, and at length come to rest in contact with it ; but he could not detect any material penetration of them into the body of the spore. Pringsheim, confirming all Thuret's ob- servations, thinks also that the spermatozoids actually pene- trate the sperm-mass ; but there is no direct proof of it. Indeed Thuret, in a very recent article (" Annales des Sci- ences Naturelles," 4 ser., vii., 1857), indicates the grounds of Pringsheim's probable mistake. The most interesting point in this last article by Thuret relates to the suddenness with which the cell-membrane is formed on the spore of Fucus after the access of the spermatozoids and the accomplishment of the act of fecundation. In six or eight minutes traces of the forma- tion of the membrane are recognizable upon a considerable number of the spores. In ten minutes the presence of a mem- brane may be clearly made manifest by the application of chloride of zinc. In an hour the membrane has acquired con- siderable firmness and thickness, and the presence of cellulose is revealed by the action of sulphuric acid and iodine; an hour later and the blue coloration under the test is decided. In the higher Cryptogamia and in the Phanerogamia^ Radlkofer's treatise, though interesting for the history, offers nothing new to our readers. In fact, its date precluded it from containing much of what is referred to in the preced- ing paragraphs. But the subject is still to be continued. DR. HOOKER ON THE BALANOPHORE^. Although read before the Linnsean Society nearly three years ago, this fine memoir ^ was published only last summer. The delay has probably been owing, in a great part, to the time requisite for the engraving of the very beautiful and 1 On the Structure and Affinities of BalanopJioreoe, by J. D. Hooker, (sep- arately issued from the Transactions of the Linncean Society of London, xxii.) London, 1857. (American Journal of Science and Arts, 2 ser., XXV. 116.) DR. HOOKER ON THE BALANOPHORE^. 95 elaborate plates whicli illustrate the memoir. It is a clear, patient, and philosophical elucidation of an extremely anoma- lous group of plants, and a succinct exposition of the principal lessons to be learned from their study, both organographically and systematically ; and it bears the impress throughout of the spirit, freshness, and independence which so distinguish this author, and make all his writings so attractive and in- structive. While the whole subject is developed in proper order, the divisions are not quite clearly marked out in the essay. The first sectional heading is: "1. Parasitism and structure of the Rhizome." But there is no section 2 answer- ing to the first, which moreover continues, without a break, to treat of the general anatomy, organography, and morphol- ogy of these plants, the structure of the flowers, ovules, and seeds, and of the diverse doctrines which have been propounded respecting them. The affinities of Balcmoj^horece are then considered under a special heading ; their Classification is then the subject of a few general remarks ; also their Geo- graphical Distribution and Variation. Then a Synoptical Table of the genera is given ; and the fourteen genera with their known species (28 in all) are finally described and illustrated. As to the structure and affinities of BaIanoj)liorece, and the curious questions that have arisen about their place in the natural system, Dr. Hooker, in the first place, affirms them to be truly ph^enogamous. It now seems strange that this should ever have been doubted. The arguments to the contrary, says our author, " all appear to have originated, on the one hand, in mistaking feeble analogies between the forms of organs that are not homologous, for affinities ; and, on the other, in overlooking a multitude of positive characters. These argu- ments may be summed up as : — 1. An erroneous view of the nature of the seeds, by Endlicher, Martins, Blume, and others, who describe them as a sporulif erous mass, — a term which, even if it were applicable, has no meaning. 2. An erroneous view of their origin being in a diseased state of the plants they grow upon, adopted by Junghuhn and Trattinick. 3. A supposed similarity in appearance to Fungi., and an erroneous 96 REVIEWS. idea that their appearance is meteoric and their growth rapid ; — a theory advanced by Endlicher, who says of the horizontal rhizome of Ilelosis and Langsdorffia, 'mycelio Fungorum quam maxima analogum.' 4. The resemblance between the articulated filaments on the capital a of the Ilelosidece and the paraphyses of Muscl ; and between the pistils of Balano- 2)horece and the pistillidia of Mosses ; strongly advocated by Griffith and Lindley. 5. The resemblance of the cellular and vascular tissues in some of their characters to some of those of Ferns, as indicated by Unger and Gceppert. 6. A very peculiar view of the nature and relations of the parts of the female flower entertained by Weddell ; who hence considers Balanophorece (together with EafflesiaceGe) to approach nearer to Gymnosperms than to any other group of plants." Instead of discussing at length opinions which "had the authors who advocate them been sufficiently furnished with specimens and facts they would never have entertained," Dr. Hooker merely recalls attention to the essential facts that these plants exhibit true flowers with stamens and pistils, gen- uine ovules, and even embryo, and so accord in no one par- ticular with Cryptogams. He shows moreover that the embryo is dicotyledonous in the few cases where it is sufficiently de- veloped to manifest the character, and that the stem is con- structed upon the exogenous plan. Even with these facts before him, Lindley has retained his Rhizogens, as " logically a class " ; as an intermediate form of organization between Endogens and Thallogens, and characterized by vegetation rather than fructification. But there is little or nothing really peculiar in their vegetation ; and, as Lindley himself reduces the differences to questions of degree, it suffices to say that the classes are not founded upon degradation of type, but upon change of type. Viewing Balanophorem., then, as degraded members of the Dicotyledonous class, Dr. Hooker follows Brown and Griffith in regarding Raffle sicicem as near to Aristolochiacece, and in denying all affinity between these and Balanophorece. In searching for the affinities of the latter. Dr. Hooker is guided by the sound rule of disregarding " the negative characters, DR. HOOKER ON THE BALANOPHOREuE. 97 as those may be termed wliicli are founded on the imperfec- tions of organs ; " and he takes the most perfectly developed species as the best exponents of the typical structure of any group, — a principle laid down, we believe, by Mr. Brown. This gives a substantial scientific basis for the estimation of affinity. Agreement in plan of structure is just what consti- tutes affinity ; agreement in grade of evolution may indicate only distant analogy, can indicate only collateral relationship, — not to be neglected, indeed, but in itself of no account in assigning a family to its true position in a system. The principle as aj^plied in the present case leads Dr. Hooker to the conclusion that the nearest relatives of Balanophorem are the Heloragece., a group itself, " consisting for the most i)art of reduced forms of Onagrariem^'' or, more strictly speaking, that the link which connects these plants with the higher forms of vegetation is furnished by Gunnera. The qualifying phrase above is appropriate ; for it is hard to conceive of Gunnera with its minute embryo as a reduced Onagraceay while it is impossible to sever the chain of evidence which binds the genus to Loudonia and Haloragis. Be this as it may, Dr. Hooker has surely made a happy hit in seizing upon Gunnera as the key to the true affinities of Balanophorece, Of all the objections that may be urged against this approxi- mation not the strongest, but rather the least valid, in our opinion (so long as the question is one of alliance and not of co-ordination), is that to be derived from the habit and the imperfection of the foliar organs. Any type is liable to have its parasitic phase, and this is generally a degraded one in these respects ; the Gesneriaceous has it in Orhancheoe^ which it might with the greatest propriety include ; the Scrophula- riaceous graduates insensibly into similar parasitic forms ; the Ericaceous has them in Monotropem ; and the Cornaceous or Olacaceous degrades through Santalaccce into Loran- thacece. It is quite probable that our author would deny the degra- dation in the latter case, judging from some points whicli he makes when considering whether the group of Balanophorcce^ '' putting aside any consideration of its relationship with other 98 REVIEWS, orders, and regarding it ^9er se, . . . should abstractedly be considered as ranking high, or the contrary." This is an ab- straction of which we are hardly capable, — that of determin- ing the rank of an order ^^e?' se. Still our author's ideas are clear and clearly expressed ; the comparison is really between these plants and the ideal plant-type. And what is wanting to make the comparison practical is a settled idea as to what constitutes the highest stjde of plant, and what is the relative importance of deviations from it ; questions too large to be en- tered upon here, if indeed the science is yet ready for their dis- cussion, but which underlie the most important inquiries which good systematic botanists are everywhere tentatively prosecut- ing. Assuming that the conventional definition of perfection in use among zoologists is applicable to the vegetable king- dom, and which argues that a high degree of specification of organs and morphological differentiation of them for the per- formance of the highest functions indicate a high rank, Dr. Hooker ingeniously argues that " Balanoijhorece may in some respects be considered to hold a very high one ; " and the points are presented under seven heads. Now we will not deny that the principles are logically applied in the present case, nor that the considerations of the kind are perhaps as applicable to the vegetable as to the animal kingdom. But we should a priori expect that principles of fundamental importance in the latter could have no sound application to the former ; that even such as relate to functions common to the two, or to structures analogous, would require to be based each upon its own ground. As to morphology, and as to what constitutes perfection of type, we should look to the fundamental differ- ences rather than to the resemblances of the two for our start- ing-point. Plants for obvious reasons are constructed on the principle of extension of surface. Concentration or consolidation, wher- ever it occurs in the vegetable kingdom, is a sj^eclal jjrovlsion against some peculiar danger. Animals, on the contrary, are formed on the principle of restriction of surface. As if to withdraw them as much as practicable from the direct action of the external world, their shape is compact, their extent as DR. HOOKER ON THE BALANOPHORE^. 99 individuals strictly limited, the external organs by which they take their sustenance comparatively few and small, while the most essential organs are safely sheltered within. Consolida- tion of organs and even their restriction in number, accord- ingly, are not likely to be indications of high rank in the vegetable kingdom. Not the latter, because the object of the plant in vegetation is attained by the indefinite repetition of the same organs ; nor the former, for the type of the plant is real- ized only in the distinct elimination of leaves from the axis. A Melon- Cactus and a Cuscuta are low forms of plants as to vegetation. As it is a fundamental character of plants that their organs of reproduction are only specialized organs of vegetation ; as the higher great divisions of plants are those in which the leaf-type is most apparent throughout ; as the perfect accomplishment of the end in view — the produc- tion, protection, and nourishment of the embryo even of the highest or most developed kind — does not require the con- fluence of homogeneous parts, why should such confluence be regarded as indicating higher rank, merely because the type is more disguised in such cases ? We see no sufficient ground for ranking a monopetalous plant higher than a polypetalous one on that account ; and still less for regarding a Loranthus or a Viscum as the highest style of plant. On the contrary, we incline to look upon the consolidation of heterogeneous parts in the blossom not as high specialization at all, but as want of development, i. e. imperfect elimination ; and in this light those who maintain an inferior ovary to be one immersed in a receptacle, must needs regard it. Again, suppression or abortion of organs that belong to the tyjDC of the blossom cannot be considered as other than an imperfection, although the loss of the corolla is no great mat- ter, and the abortion of one of the sexes little more. Still hermaphroditism is plainly in the type of the highest style of plant ; while the opposite is the case in the animal kingdom. But we cannot here enter further into the discussion of this class of questions. No one feels more deepl}^ than our author the want of fixed and philosophical principles for the subor- dination of characters and the study of aflinities in plants ; 100 REVIEWS. and no botanist of his age is more competent, or so well placed and furnished for the investigation of this problem, to which we invite him as to a task worthy of his powers. As to the rank of Balanophoreoe^ if our author has demon- strated anything, it is that they belong to the highest class of l^lants, but that they are probably the most degraded members of it. BOUSSINGAULT ON THE INFLUENCE OF NITRATES ON THE PRODUCTION OF VEGETABLE MATTER. Several years ago Boussingault demonstrated, in the clearest way, that plants are incapable of assimilating the free nitrogen of the atmosphere. Two years ago, in a paper communicated to the French Academy of Sciences, he showed that nitrates eminently favor vegetation. He now shows,^ by decisive experiments, — (1) That the amount even of ternary vegetable matter produced by a plant depends absolutely upon the supply of assimilable nitrogen (ammonia and nitrates). A plant, such as a sunflower, with a rather large seed, may grow in a soil of recently calcined brick, watered with pure water, so far as even to complete itself with a blossom ; but it will only have trebled or quadrupled the amount of vegetable matter it had to begin with in the seed. In the experiments, the seeds weighing 0.107 grams, in three months of vegetation formed plants which when dried weighed only 0.392 grams, — a little more than trebling their weight. The carbon they had acquired from the decomposition of carbonic acid of the air was only 0.114 grams ; the nitrogen they had assimilated from the air in three months was only 0.0025 grams. (2) Phosphate of lime, alkaline salts, and earthy matters 1 Researches upon the Influence which assimilable Nitrogen in manures exerts upon the production of Vegetable Matter ; and (2) Upon the Quantity of Nitrates contained in the Soil and in Water of various kinds. J. B. J. D. Boussingault. Annales des Sciences Naturelles, 4 ser., vii., No. 1, 1857. (American Journal of Science and Arts, 2 ser., xxv. 120.) INFLUENCE OF NITRATES. 101 indispensable to the constitution of j)lants exert no appre- ciable action upon vegetation, except when accompanied by matters capable of furnishing assimilable nitrogen. Two plants of the same kind, grown under the same conditions as above, but with the perfectly sterile soil adequately supplied with phosphate of lime, alkali in the form of bicarbonate of potash, and silex from the ashes of grasses, resulted in only 0.498 grams of dried vegetable matter, from seeds weighing 0.107 grams : and had acquired only 0.0027 grams of nitrogen beyond what was in the seeds. (3) But nitrate of potash furnishing assimilable nitrogen, associated with phosphate of lime and silicate of potash, forms a complete manure, and suffices for the full develop- ment of vegetation. Parallel experiments with nitrate in place of bicarbonate of potash resulted in the vigorous growth of the Sunflower plants, and the formation of 21.248 grams of organic matter, from seeds weighing as before only 0.107. This 21.111 grams of new vegetable matter, jiroduced in three months of vegetation, contained 8.444 of carbon derived from the carbonic acid of the air and 0.1666 grams of nitrogen. The 1.4 grams of nitrate of potash suj^plied to the soil con- tained 0.1969 grams of nitrogen, leaving a balance of 0.0303, nearly all of which was found unappropriated in the soil. Finally Boussingault made a neat series of comparative experiments, introducing into calcined sand the same amount of phosphate of lime and carbonate of potash, but different proportions of nitrate of soda, or in other words of assimi- lable nitrogen, and watering with w^ater free from ammonia but containing a quarter of its volume of carbonic acid. The soil was divided among four pots, each having two seeds of Sunflower (^H. argophylliis was the species used in all the ex- periments) ; the pot No. 1 received of nitrate of soda 0.00 grams. No. 2 " « " 0.02 " No. 3 " " " 0.04 " No. 4 " " " O.IG " The results of fifty days' vegetation are given in the rate 102 REVIEWS. of growth, size and number of leaves, weight of the pro- duct, etc. : — No. 1 made of new vegetable matter . . . 0.397 grams. No. 2 " " "... 0.720 " No. 3 " " "... 1.130 '' No. 4 " " "... 3.280 " In No. 2 so little as three milligrams of assimilable nitrogen introduced into the soil enabled the plant to double the amount of organic matter. The proportion of the weight of the seeds to that of the plant formed was in No. 1, as 1 : 4.6 grams. No. 2, as 1 : 7.6 No. 3, as 1 : 11.3 No. 4, as 1 : 30.8 In no case did the nitrogen acquired by the plant exceed that of the nitrate added to the soil. In the experiments where no nitrate was added to the soil the two or three milligrams of nitrogen acquired by the plants during three months of vegetation came in all probability from ammoniacal vapors and nitrates existing or formed in the at- mosphere. To establish their presence, Boussingault arranged an apparatus which detected the production of some nitrates. And, in exposing to the air 500 grams of calcined sand, which had 10 grams of oxalic acid mixed with it, in a glass vessel with an open surface equal to that of one of the flower-pots used in the above experiments, the sand took 0.0013 grams of nitrogen from the air, of which a part was certainly ammonia. The object of the researches of which a summary is given in the second paper was, to determine the quantity of nitrates contained, at a given moment, in one hectare of cultivated ground, one of meadow, one of the forest soil, and in one metre of river or spring water. The quantity in the soil was of course found to vary extremely with the extremes of wet or dry weather. Garden soil, highly manured every autumn, con- tained on the 9th of August, 1856, after fourteen dry and warm days, 316.5 grams of nitre in a cubic litre of soil. On the 29th of the month, after twenty rainy days, the same quan- INFLUENCE OF NITRATES. 103 tity of the same soil contained only 13 grams of nitre. The greater part had been dissolved out of the superficial soil. Some specimens of forest-soil, in a state of nature, fur- nished no indication of nitrates ; others gave 0.7 and 3.27 grams of nitre to the cubic metre. The soil of meadows and pastures afforded from 1 to 11 grams of nitre to the cubic metre. Nineteen sjDccimens of good cultivated land gave, four of them none ; others from 0.8 to 1.33 ; the richer ones from 10.4 to 14.4, and one fallow, of exceptional richness, as much as 108 grams of nitre to the cubic metre. To the latter much calcareous matter has been added. The soil of the conservatory, from which the nitrates would not be washed away by rains, contained 89, or 161, and some rather deep soil 185 grams of nitre to the cubic metre. The sources of the nitre are not difficult to understand when we reflect that a manured soil, especially a calcareous one, is just in the condition of an artificial nitre-bed. The ultimate result of the decomposition of ordinary manure is a residuum of alkaline and earthy salts, phosphates, and ni- trates, the latter, with the ammonia furnishing the assimi- lable nitrogen, all-essential to productive vegetation. In in- corporating with the soil undecomposed manure, instead of the ultimate results of decomposition, less loss is suffered from prolonged rains wasting out the formed nitrates. The soluble matters washed out of the soil are to be sought in the water. River and spring waters therefore act as manure by the silex and alkali, the organic matter, and the nitrates which they hold. The spring waters, poorest in nitre of those examined, contained from 0.03 to 0.14 milligrams of nitre to the litre ; the richer ones from 11 to 14 grams in the cubic metre. As to the river-water, the Yesle in Champagne held 12 grams, the Seine at Paris 9 grams the cubic metre. These were the richest. The Seine at Paris carries to the sea, in times of low water, 58,000 kilograms, in times of high water, 194,000 kilograms, of nitre every twenty-four hours. What enormous amounts of nitre must be carried into the sea by the 104 REVIEWS. Mississippi, the Amazon, and by every great continental river ; and bow active, beyond all ordinary conception, must the process of nitrification be all over the land ; and how vast the su^Dply of assimilable nitrogen for the use of vegetation ! BENTHAM'S HAND-BOOK OF THE BRITISH FLORA. One of the best systematic botanists — of the soundest judgment and the largest experience, both in European and exotic botany — has deemed it no unfit employment of a por- tion of his valuable time to prepare a volume ^ by which beginners, having no previous acquaintance with the science, may learn to know most advantageously and readily, the wild flowers and plants of his native land. The result is a genuine popular Flora, and a clear proof that the plants of a limited country may be described, by one who understands them thor- oughly, in comparatively simple language, without any sacri- fice of scientific accuracy, or of scientific interest. No really good work of this kind was ever made by a compiler ; and no one who has not essayed the task, can comprehend how thor- oughly faithful writing for beginners brings one's knowledge to the proof. The characteristic features of the work before us are : 1. The full use of analytical keys, after the mode of Da Can- doUe's " Flora Fran^aise," leading easily not only to the order and the genus, but also to the species of the plant in hand. These keys, or analyses, are here made to supersede specific characters as such, neat and free in descriptions, longer or shorter according to circumstances, occupying their place. But generic characters are given with considerable fullness. 2. The exclusion of all technical terms which were not re- quired for the purpose in view, and " the omission, in nu- merous instances, of microscopical, anatomical, or theoretical 1 Hand-hook of the British Flora ; a Description of the Flowering Plants and Ferns indigenous to, or naturalized in, the British Isles : for the use of Beginners and A mateurs. By George Bentham. London, 1858. (Amer- ican Journal of Science and Arts, 2 ser., xxvi. 413.) HAND-BOOK OF THE BRITISH FLORA. 105 characters, often of the greatest importance in scientific botany, but useless to the mere amateur." 3. The descrii> tions are original, and have been drawn up from British specimens in the first instance, and afterwards compared with the characters given in the standard Floras, and verified upon continental specimens from various parts of the geographical range of the species. As a describer of species (which is something very different from a describer of specimens), i\Ir. Bentham has no superior. 4. The geographical range of each species, at least its European range, is carefully specified; then the British stations are given in general terms, the object being to state where the plant is likely to be found, rather than to indicate the precise spot where it has been gathered. 5. The judicious limitation of species, and the reduction of a crowd of nominal or " critical " species to their supposed types, with a thoroughness which only a botanist of Mr. Bentham's great experience and authority could well venture upon. The following extract from the preface will explain his views : — " Taking into account the omission of all plants erroneously indicated as British, it will still, no doubt, be a matter of astonishment that, whilst the last edition of Hooker and Arnott's Flora contains 1571 species, and that of Babington's Manual as many as 1708 (exclusive of Chara), the number, in the present work, is reduced to 1285. This is not owing to any real difference of opinion as to the richness and di- versity of our vegetable productions, but is occasioned by a different appreciation of the value of the species themselves. The author has long been persuaded that the views originally entertained by Linnaeus, of what really constitutes a species, were far more correct than the limited sense to which many modern botanists seem inclined to restrict the term ; and that in most cases where that great master had good means of observation, he succeeded admirably in the practical applica- tion of his principles. At any rate, if those minute distinctions by which the innumerable varieties of Brambles, of Roses, of Hawk-weeds, or of Willows, have of late years been charac- terized, are really more constant and more important than the author's experience has led him to conclude, they cannot 106 REVIEWS. be understood witliout a more complete acquaintance with ti'iflino\ vasfue, and sometimes theoretical characters, than he has himself been able to attain, or than can ever be expected from the mere amateur. . . . The species are limited accord- ing to what are conceived to have been the original principles of Linnaeus : and the author, in submitting his views to the judgment of the scientific world, trusts that they will not be attributed to hasty generalizations, or conjectural theories, but that they will be generally recognized as founded on personal observation of living plants, made during many years' resi- dence on the continent, as v/ell as in this countrj^, and on repeated comparison of specimens collected from the most varied and distant points of the geographical areas of the several species." 6. Popular names are employed and reduced to a system in accordance with the principles of botanical nomenclature. " An attempt has, on the present occasion, been made to give prominence to a series of English names to the British plants, rendering them as far as possible consistent with the recog- nized principles of systematic nomenclature, so essential for the study of plants. It was at first intended merely to have adopted those which are appended to all the genera and species in Hooker and Arnott's Flora ; but the first attempts to apply them practically, gave evidence that they had never been framed with a view to being used by botanists, or ama- teurs, in the place of the Latin ones. It will be observed that there is among them a continual confusion between popu- lar, trivial, and generic names ; between epithets and specific names ; betv/een substantives and adjectives ; that on frequent occasions one name is applied to several genera, or several names to one genus ; that the number of words forming the name of a plant varies from one to &Ye, instead of being con- stantly two ; and that some of the names put forward as Eng- lish, are very local, almost unknown or obsolete, and no easier to learn than the more useful Latin ones they represent. It became necessary, therefore, thoroughly to revise the whole system, and to recast it upon the Linnsean principles univer- sally adopted for the Latin botanical names. . . . The full HAND-BOOK OF THE BRITISH FLORA. 107 statement of the principles whicli have induced the rejection of certain names, and the substitution of others, and the details of their application to individual cases, . . . are given at length in a paper prepared by the author, to be laid before the Linnaean Society on the publication of this Flora." Criticism may well be deferred until this paper comes to hand. Of the propriety of an English nomenclature of some kind in a Flora where a great part of the plants have well- known vernacular names, there is no room for doubt ; and if used at all, it is desirable that these names should be reduced to a systematic form. This is readily done for perhaps half of the common plants of the British Flora ; but for the rest, the difficulties are various and much greater than one would imagine before making the attempt. The present undertak- ing must be deemed a decided success. What imperfections it has, are on the safer side. We should have inclined to a larger use of the vernacular for the generic names ; and where they were inapplicable to the whole genera, to apply them to subgenera, e. (/., Apple and Pear, Gooseberry and Currant. As these are real and universal English generic names, they ought, if possible, to be given as such. Still we appreciate the reasons which appear to have compelled the adoption of Pyrus and Ribes as English names, though Eng- lish they never can become. Our author is strongly dis2)osed throughout to make the Latin name do duty as an English one, doubtless supposing that they may become popular appel- lations in time, as Geranium and Aster have done. Some- times he adopts the Latin word entire ; sometimes he trun- cates or anglicizes the termination. Happy instances of the latter sort are : — Trigonel, from Trigonella. Limosel, from Limo sella. Corydal, from Corydalis. Corrigiole, from Corrigiola. Chrysosplene, from Chrysosplenium ; but why not Golden- spleen ? Samole, from Samolus ; but why not Brookweed ? Limnanth, from Limnanthemum. 108 REVIEWS. Scleranth, from Scleranthus ; but why not Knawel ? Osmund, from Osmunda. Myriophyll, from Myriopliyllum ; but why not Milfoil ? Matricary, for Matricaria. Eupatory, for Eupatorium, etc. Those names which are not at all to our tastes are : — Cerast, for Cerastium ; but if such a word must be coined, why not Holost, for Holosteum on the preceding page ? Doronic, for Doronicum. Onopord, for Onopordon ; why not Cotton-Thistle ? Polycarp (newly martyred), for Polycarpon. Myosote, for Myosotis ; in place of Forget-me-not. Capsell, for Capsella ; in the place of the vernacular Shep- herd's-purse. The best coinage of an English name is Rockcist for Heli- anthemum. An Introduction of thirty-six pages teaches the elements of botany to beginners, and explains the technical terms used in the flora, and many besides. The definitions of perigynous and epigynous, however conformable to etymology, are not the quite usual ones, and are not adhered to in the work itself. We were not aware that " in general the word ovary is used to designate all the ovaries of a flower," unless when united into one body, and are glad to observe that the author does n©t use the word in this way in the body of the work, one or two instances excepted. We always supposed the word to be an exact synonym of the Linnsean germen. And if we may not use it, as botanists always have done, for the ovule-bearing portion of the pistil, whether simple or com- pound (reserving carpel for the simple of elementary pistil, whether separate or combined), then a new word must needs be coined for this very purpose. To mistake the radicle of the embryo for the root, is common to all English botanists. The short sections upon classification and the examination and determination of plants, are full of practical wisdom. IMPROVEMENT OF CULTIVATED PLANTS. 109 VILMORIN'S IMPROVEMENT OF CULTIVATED PLANTS. This very interesting pamphlet ^ is a collection and reprint of several of Louis Vilmorin's important communications to the Central Agricultural Society of France and to the Acad- emy of Sciences : to which is prefixed a French translation of a memoir upon the Amelioration of the Wild Carrot, con- tributed by his venerable father to the Transactions of the London Horticultural Society (but not before published in the vernacular of the author), which memoir, as the younger Vilmorin informs us, was the point of departure for his own investigations in this field, and even contains the germ of most of the ideas which he has since developed upon the the- ory of the amelioration of the plants from the seed. These papers claim the attention of the philosophical naturalist, no less than of the practical horticulturalist. Most of our esculents are deviations from the natural state of the species, which have arisen under the care and labor of man in very early times. New varieties of these cultivated races are originated almost every year, indeed ; but between these particular varieties, the differences, however well marked, are not to be compared for importance with those changes which the wild plant has generally undergone in assuming the esculent state. In this amelioration dt altera- tion, as in other cases, c'est la premiere pas qui coute. For the altered race, once originated, has more stability than the wild stock ; it accordingly tends not only to degenerate (as the cultivator would term it) towards its original and less useful state, but also to sport into new deviations, in various directions, with a freedom and facility not manifested by its wild ancestors. This explains the readiness with which we continually obtain new varieties of those esculent plants which have been a long time in cultivation, while a newly-introduced 1 Notice sur V Amelioration des Plantes par le Semis et Consideratiom sur VHeredite dans les Vegetaux. Par M. Louis Vilmorin. Paris, 1859. (American Journal of Science and Arts, 2 ser., xxvii. 440.) no REVIEWS. plant exhibits little flexibility. To detect tbe earliest indica- tions of sporting, and to select for the parents of the new race those individuals which begin to vary in the requisite direc- tion, is the part of the scientific cultivator. In this way, the elder Vilniorin succeeded in producing the esculent carrot from the wild stock in the course of three generations, — no addition to our resources, indeed, but significant of what may be done by art directed by science. By adopting and skill- fully applying these principles, the younger Vilmorin has con- ferred a benefit upon France which (if she wall continue to make sugar from the beet) may almost be compared with that of causing two blades of grass to grow where only one grew before, having, so to say, created a race of beets containing twice as much sugar as their ancestors, and indicated the practicability of its perpetuation. The mode of procedure, and the ingenious methods he contrived for rapidly selecting the most saccharine out of a whole crop of beets, as seed-bear- ers for the next season, are detailed in these papers. Once originated, and established by selection and segrega- tion for a few generations, the race becomes fixed and per- petuable in cultivation, with proper care against intermixture, in virtue of the most fundamental of organic laws, namely, that the offspring shall inherit the characteristics of the parent, — of which law that of the general permanence of species is one of the consequences. The desideratum in the production of a race is, how to initiate the deviation. The divellent force, or idiosyncrasy, the source of that " infinite variety in unity which characterizes the works of the Creator," though ever active in all organisms, is commonly limited in its practical results to the production of those slighter differences which ensure that no two descendants of the same parent shall be just alike, being overborne by that opposite or centri- petal force, whatever it be, which ensures the particular re- semblance of offspring to parents. Now the latter force, as Mr. Louis Vilmorin has well remarked, is really an aggrega- tion of forces, composed of the individual attraction of a series of ancestors, which we may regard as the attraction of the type of the species, and which we perceive is generally all- IMPROVEMENT OF CULTIVATED PLANTS. Ill powerful. There is also the attraction or influence of the im- mediate parent, less powerful than the aggregate of the ances- try, but more close, which ever tends to impress upon the off- spring all the parental peculiarities. So, when the parent has no salient individual characteristics, both the longer and the shorter lines of force are parallel, and combine to produce the same result. But whenever the immediate parent deviates from the type, its influence upon its offspring is no longer parallel with that of the ancestry ; so the tendency of the off- spring to vary no longer radiates around the type of the species as a centre, but around some point upon the line which represents the amount of its deviation from the type. Left to themselves, as Mr. Vilmorin proceeds to remark, such varie- ties mostly perish in the vast number of individuals which annually disappear, — or else, we may add, are obliterated in the next generation through cross-fertilization by pollen of the surrounding individuals of the typical sort, — whence results the general fixity of species in Nature. But under man's protecting care they are preserved and multiplied, perhaps still further modified, and the better sorts fixed by selection and segregation. Keeping these principles in view, Mr. Vilmorin concluded that, in order to obtain varieties of any particular sort, his first endeavor should be to elicit variation in any direction whatever ; that is, he selected his seed simply from those in- dividuals which differed most from the type of the species, however unlike the state it was desired to originate. Repeat- ing this in the second, third, and the succeeding generations, the resulting plants were found to have a tendency to vary widely, as was anticipated ; being loosed, as it were, from the ancestral influence, which no longer acted upon a straight and continuous line, but upon one broken and interrupted by the opposing action of the immediate parents and grand- parents. Thus confused by the contrariety of its inherited tendencies, it is the more free to sport in various ways ; and we have only to select those variations which manifest the qualities desired, as the progenitors of the new race, and to develop and fix the product by selection upon the same prin- ciple continued for several generations. 112 EE VIEWS. It is In tills way that Mr. Vilmorin supposes cross-fertiliza- tion to operate in the production of new varieties ; and even in the crossing of two distinct species, the result, he thinks, is rarely, if ever, the production of a fertile hybrid, but of an offspring which, thus powerfully impressed by the strange fertilization and rendered productive by the pollen of its own female j^arent, is then most likely to give origin to a new race. We cannot follow out this interesting but rather recondite subject in a brief article like this. But we are naturally led to inquire whether the history of those plants with which man has had most to do, and the study of the laws which regulate the production and perpetuation of domesticated races, may not throw some light upon the production of varieties in Na- ture ; and whether races may not have naturally originated, occasionally, under circumstances equivalent to artificial selec- tion and segregation. Some recent attempts which have been made in this direction we may hope to notice upon another occasion. THE BUFFALO-GRASS. The Buffalo-Grass,^ so abundant and so widely diffused over the broad, arid region which separates our Pacific from our Atlantic possessions, is one of the humblest plants of its order, rising only a few inches above the surface of the soil ; but at the same time it is one of the most important and use- ful, since it forms the principal subsistence of the buffalo for a part of the year, and no less so of the cattle of the emigrant. The botanical history of this little grass, now happily com- pleted by Dr. Engelmann, is remarkable. Nuttall first named and described it nearly thirty years ago ; and, wdiile he re- ferred it to Sesleria, susj^ected it to be sui generis^ and threw ^ Two neio Genera of Dioscious Graffses of the United States. By George Engelmann, M. D. Extr. from the Transactions of the Academy of Nat- ural Sciences of St. Louis, i. p. 431 ; with three plates. 1859. (American Journal of Science and Arts, 2 ser., xxviii. 439.) THE BUFFALO-GRASS. 113 out a happy conjecture as to its natural relationship. Torrey figured it twelve years ago, and also announced its affinity to the Chloridece ; he at the same time discovered its dioecious character, and showed that only the male plant was known. At length Dr. Engelmann has detected the female plant in a rather rare grass, the Anthephora axilUflom of Steudel, which is so unlike the common Buffalo-Grass that it naturally had been referred to a widely different tribe. Struck by the similarity of their foliage and stoloniferous growth, as they occurred together in a collection made by his brother, Dr. Engelmann shrewdly suspected the relationship, and finally set the question at rest by finding a male Buffalo-Grass which happened to bear a stalk of female flowers from the same root- stock ; and these flowers were those of the so-called Anthe- phora. So different are the two that nothing short of this ocular proof would have been convincing. It hardly need be said that the male plant is not a Sesleria, nor the female an Anthephora ; although they severally resemble these genera, or at least the female spikelets have a very great external re- semblance to the Paniceous genus Anthephora. So that Dr. Engelmann, having to characterize this new generic type, very naturally named it Biichloe (shorter and more euphonious than Bubalochloe), i. e. Buffalo-Grass ; and he retained the speciflc appellation of Dactyloides, although the male plant is not much like a Dactylis, and the female wholly unlike. Very glad we are to see the genus established under so ap- propriate a name, — the more so as it has narrowly escaped a different fate. That is to say, two inchoate attempts seem to have been made to found a genus upon the male sex. First, in Sir William Hooker's enumeration of the plants of Geyer's western collection we find " Calanthera dactyloides. Kth. — Nutt. Sesleria, Nutt. Gen. i. p. Qb'' But neither Kunth nor any other author has described a genus Calanthera. AVe have a suspicion that the " Kth." is a slip of the pen, and that the name is really Nuttall's, given by him to a specimen in the Hookerian herbarium. But if this be so, the manuscri})t name (which, moreover, is destitute of any particular signifi- cance) can by no means now supersede Engelmann's published 114 REVIEWS. one ; though we might have been constrained by courtesy to adopt it, if this suspicion had occurred to him, and he had been able to confirm it. Again, in the corrections at the close of the " Plantae Hartwegianse," Mr. Bentham applies the name of ^^ Lasiostega humlUs, Rupprecht (ined.) " to No. 250, which he had before called a Triodla. The plant is un- doubtedly a male Buffalo-Grass. But no genus Lasiostega is found to be published, nor has this name any appropriateness as applied to the plant in question. It is curious to remark that the male plant, being more proliferous by stolons than the female, has nearly displaced the latter, or has (so far as known) attained a wider geo- graphical range as well as a far greater abundance. Prob- ably, in accordance with a general law, the tendency to bar- renness from seed which accompanies copious multiplication by offshoots, has also assisted in the production of this re- sult, — a state of things quite contrary to the genius of that polygamous community which has effected a lodgment in the region of Buffalo-Grass. Dr. Engelmann's second genus, Monanthochloe, is founded upon a singular, exceedingly stoloniferous, littoral grass with leaves scarcely half an inch long, with solitary sessile spilie- lets, which has long been known to occur on the coast of Texas and Florida (collected by Berlandier, Drummond, and Blodgett), but has never been studied until now. In fact, it has been thought to be something abnormal, on account of its showing as its most interesting feature, a regular transition from the foliage to the paleJB of the flowers. Dr. Engelmann notes that the glumes are wanting (perhaps represented by ordinary leaves of the axis of which the spikelet is a direct continuation), the uppermost leaf representing the lowest palea of the spikelet. The latter consists of from three to five flowers, of which the lowest flower and sometimes the next are neutral or rudimentary, from one to three succeed- ing ones are staminiferous or pistllliferous, according to the sex, and the uppermost is also reduced to a rudiment. In the hands of agrostologists such a grass as this will be likely further to elucidate the floral structure of the order, the THE TREES OF NORTH CAROLINA. 115 theory of which is by no means settled yet. Dr. Engel- mann's three excellent plates, displaying all the details of the flowers, will facilitate this investigation. The youthful Academy of Natural Sciences of St. Louis is well inaugurating its public career by publications of such character as this paper, and the more elaborate "Mono- graph of Cuscuta " by the same author, which is now in press. THE TREES OF NORTH CAROLINA. We have turned over the pages of this popular exposition ^ with much interest, and gleaned some valuable information. " Botanists will of course find fault with it," says the author, who we well know could write scientifically and profoundly enough, if he so pleased, but who has here come down to the level of his most unlearned readers, discoursed separately of trees, shrubs, and vines, and classified these in a fashion which might well shock the susceptibilities of a stickler for tech- nical nomenclature and natural system in botany. Now, we are not shocked at all ; indeed we quite enjoy a glimpse of Flora en deshabille and slip-shod, and are well aware how much easier it is, and how much better in such cases, to fit your book to its proper readers than to fit the readers to it. The fault we should find is not with the plan of this Report but with the quantity. We could wish for more of it, for a volume as large at least as Mr. Emerson's Report on the Trees and Shrubs of Massachusetts. We quite like to see the popular names put foremost, but would suggest that the botanist who does this should lead as well as follow the in- digenous nomenclature, so far as to correct absurd or in- congruous local names and introduce right or fitting ones as far as practicable. For instance, " Virgin's Bower " is not a proper name for Wistaria frutescens, and is rightly applied to 1 Geological and Natural History Survey of North Carolina. Part III. Botany. The Woody Plants of North Carolina. By M. A. Curtis. Raleigh, 18G0. (American Journal of Science and Arts, 2 ser., xxx. 275.) 116 HE VIEWS. Clematis Virginiana over the leaf. (We venture to add, in passing, C Viorna to the list, having gathered it in Ashe County.) And although the people alongshore call Baccha- ris by the name of the English annual weed, " Groundsel," it were better to write it '' Groundsel-tree." " Yellow-wood " is the name of Cladrastis, rather than of Symplocos, which the Carolinians call " Horse-Sugar." Dr. Curtis can coin a name upon occasion ; for surely nobody in Carolina knows 3Ien- ziesia glohularis as False Heath, nor has it any scientific claim to this appellation. While in critical mood we may express a strong dissent from the proposition that Rhodo- dendron punctatum is too inferior to the other two species " to attract or deserve much attention." With us, it is sur- passingly beautiful in cultivation, none the less so because its habit is so different, having light and pendent branches, when well grown forming broad and thick masses, and loaded with its handsome rose-colored blossoms. While Leucothoe Catesbcel is called " a very pretty shrub," the far hand- somer A7idromeda Jlorihunda,so much prized by our nursery- men, gets no commendation. JIagnolia Fraseri may not only be " cultivated in the open air near Philadelphia," but is perfectly hardy near Boston, and the earliest to blossom ; but we never noticed the fragrance of the flowers. On the other hand, as it is a native so far south as Florida, it might thrive in plantations anywhere in North Carolina. The flowers of M, cordata are described as if larger than those of 31. Fra- seri., instead of the contrary ; we could hardly say much for their beauty, except in comparison to those of the common Cucumber-tree. Frunus Virginiana is omitted; yet surely it is not wanting in North Carolina. And it is almost an excess of conscientiousness to leave out Cladrastis, the hand- somest tree of the country, all things considered, when it is known to grow only a few rods over the Tennessee line. On the other hand, we are disposed to doubt if the gen- uine White Spruce (^Ahies alha) occurs in North Carolina. At length we know this tree, but only in Canada and parts adjacent. It is more, instead of less, northern in its range than A. nigra. But since President Wheeler has pretty BENTHAM'S FLORA OF HONGKONG. 117 nearly determined the existence of A. Fraseri on the Green Mountains in Vermont, we could not deny that A. alba grows with the latter on the high mountains of North Carolina. We make our little criticism freely, — as we know the ex- cellent author would wish, — for we think it likely that this part of the Keport will pass to a second edition, — when we hope it will be largely augmented. BENTHAM'S FLORA OF HONGKONG. The present work^ is the third of the series of British Colonial Floras, upon a new and simple plan, compact in form, written in English throughout, authorized and supported by the British Government. The Colonial department pays a very moderate recompense to the authors, and turns the work over to a publisher upon such terms as to render the volume generally accessible to working botanists and colonists. This is a much wiser as well as vastly more economical plan of government patronage to scientific publication than that adopted in this country, one which secures that the publications are just what is wanted and that they reach the hands which are to use them, and not others, — one which, when our pres- ent task is done and we again cultivate the arts of peace, we might profitably adopt. The present work is by a master- hand ; for Mr. Bentham is one of the most experienced, in- dustrious, and judicious of systematic botanists. The island of Hongkong has an area of scarcely thirty square miles, its general aspect is bleak and barren ; yet it has already yielded about a thousand phaenogamous species. " At a first glance," as the author observes, " one is struck with the very large total amount of species crowded upon so small an island, which all navigators depict as apparently so bleak and bare ; 1 Flora HongJcongensis ; a Description of the Flowering Plants and Ferns of the Island of Hongkong, by George Bentham. With a :Map of the Island. Published under the Authority of Her Majesty's Secre- tary of State for the Colonies. London, 1861. (American Journal of Science and Arts, 2 ser., xxxii. 124.) 118 REVIEWS. — with the tropical character of the great majority of species, when botanists agree in representing the general aspect (de- rived from the majority of individuals) to present the fea- tures of a much more northern latitude ; — with the large pro- portion of arborescent and shrubby species, on a rocky mass where the woods are limited to a few ravines, or short narrow valleys half monopolized by cultivation ; — and with the very great diversity in the species themselves, the proportion of orders and genera to species, and the comparative number of monotypic genera, being far greater in the Hongkong Flora than in any other Flora of similar extent known to me. The very large number of endemic species — of species known to us only from the island — is probably occasioned by cur igno- rance, already alluded to, of the vegetation of continental south China." A fitting acknowledgment is given for the important con- tribution to this Flora furnished by the botanical collection (of above 500 species) made by Charles Wright, as botanist of the U. S. North Pacific Exploring Expedition under Cap- tains Ringgold and Rodgers, duplicates of which were oblig- ingly and most properly furnished by direction of the Com- mander and the enlightened Secretary of the Smithsonian Institution. In aid of the colonial botanists or amateurs who may use this Flora, the author has prefixed (with some minor altera- tions) the admirable brief outlines of Botany and Glossary prepared for his popular British Flora. In these Outlines the subject is regarded, not from the mor- phological or the physiological, but from the descriptive point of view. It opens with a statement of the nature and design of a Flora, and of what a botanical description ought to be. " These descriptions should be clear, concise, accurate, and characteristic, so that each one should be readily adapted to the plant it relates to, and to no other ; they should be as nearly as possible arranged under natural divisions, so as to facilitate the comparison of each plant with those nearest allied to it ; and they should be accompanied by an artificial key or index, by means of which the student may be guided BENTHAM'S FLORA OF HONGKONG. 119 step by step in the observation of such peculiarities, or char- acters, in his x^hant as may lead him, with the least delay, to the individual description belonging to it. " For descriptions to be clear and readily intelligible, they should be expressed as much as possible in ordinary well- established language. But, for the purpose of accuracy, it is necessary not only to give a more precise technical meaning to many terms used more or less vaguely in common conver- sation, but also to introduce purely technical names for such parts of plants or forms as are of little importance except to the botanist. In the present chapter it is proposed to define such technical or technically limited terms as are made use of in these Floras. " At the same time mathematical accuracy must not be ex- pected. The forms and appearances assumed by plants and their parts are infinite. Names cannot be invented for all ; those even that have been proposed are too numerous for ordinary memories. Many are derived from supposed resem- blances to well-known forms and objects. These resemblances are differently appreciated by different persons ; and the same term is not only differently applied by two different botanists, but it frequently happens that the same writer is led on differ- ent occasions to give somewhat different meanings to the same word. The botanist's endeavors should always be, on the one hand to make as near an approach to precision as circum- stances will allow, and on the other hand to avoid that prolix- ity of detail and overloading with technical terms which tends rather to confusion than to clearness. In this he will be more or less successful. The aptness of a botanical description, like the beauty of a work of imagination, will always vary with the style and genius of the author." These Outlines are throughout so well sketched, and so worthy to be regarded as of standard authority, that we must still venture a criticism or two, looking to their possible im- provement. In the first place, referring to paragraphs 8 and 88, we must dissent from the proposition that the subject of homol- ogy does not belong to " morphology in the proper sense of 120 REVIEWS. the term ; " — unless, indeed, morphology relates simply to form in the lowest sense, to mere shape, arbitrarily viewed, — which would belittle the subject down to mere terminology, and empty that of all scientific interest. If the comparison even of a perfoliate or clasping with a cordate leaf, or of membranaceous or coriaceous wdth thickened leaves, such as those of a Houseleek, a Mesembryanthemum, and an Aloe, falls within the province of morphology, surely so also must the comparison of an ordinary leaf with a cotyledon, with a bulb-scale, a bud-scale, and no less with a sepal, a petal, a carpel, etc. In the latter we merely trace morphological relations of the very same kind somewhat further and higher. The relation of a leaf as foliage to the scale of a bud, or to the thorn of a Barberry, is clearly of the same category as its relation to a sepal or a petal, — the latter, as we regard it, bringing in no new idea, and requiring no new point of view. Next, Qulncimcial imbrication is defined by Mr. Bentham to be that arrangement in which " one petal is outside, an ad- joining one wholly inside, the three others intermediate and overlapping on one side." But why give this name to a mixed form, to that which is merely convolute aestivation deranged by one of the five petals getting both edges under ? And why change the uniform usage from De Candolle's " Theorie Ele- mentaire," if not earlier, down to the present time, which de- fines the quincuncial mode as having two members exterior, two interior, and one with one edge overlapping its neighbor and the other overlapped ; an arrangement which especially merits a distinguishing name, since it is the normal imbrication in a pentamerous perianth, answering as it does to two fifths phyllotaxis. So that current usage and reason tell against the innovation. In the third place, we are equally inclined to demur to the proposed modifications of the sense of the terms 2)C7ngynous and epifjynous (paragraph 140), Mr. Bentham restricting the former to those cases in which the petals, etc., are adnate to a perfectly free calyx, as in the Cherry, and applying the latter in cases where the calyx, equally bearing the petals, etc., is adnate even merely to the base of the ovary, if only the BENTHAM'S FLORA OF HONGKONG. 121 adhesion readies above the level of the insertion of the lowest ovule ; which would make most Saxifrages epigynous. Besides the etymological objections, and the inconvenience of a change, the new definitions seem to us to be at least as ambiguous as the old in practice ; and it is not surprising that they are not uniformly adopted in the Hongkong Flora itself. Finally, as to paragraph 1G6, we are not much better satisfied with the definition that the radicle is the " base of the future root," than with the original statement that it is " the future root." To us nothing in botany is clearer, or more patent to observation during germination, than that while the radicle is, if you please, "the base of the future root" inasmuch as it is that from which the root proceeds, it is itself the first inter- node of the stem. This view, to which morphological considera- tions and observation of the development long since brought us, appears to be generally adopted by the French and Ger- man botanists, but not by the English. If the radicle univer- sally failed to elongate, as in Monocotyledons, and in the Pea, Oak and others with hypogseous germination, this organ might be deemed to be merely the base of the future root ; but its more usual elongation, in the manner of any other internode, plainly reveals the cauline nature which analogy would also assign to it. The chapter on Vegetable Anatomy and Physiology is new, is very condensed, and considering that it deals with matters to which Mr. Bentham has never specially attended, is remark- ably good and accurate. We merely observe in passing, of paragraphs 195, 197, that the distinction between exogenous and endogenous stems is as obvious during the first season, and even at its beginning, as ever afterward, and it is then that the purely systematic botanist will more commonly have occasion to examine the structure in this regard; of § 198", that " the liber or inner bark " is by no means always " formed of bast cells ; " of § 200, that we cannot accept the statement that " in the leaf the structure of the petioles and priuclpal ribs or veins is the same as that of the young branches of which they are ramifications," at least in any sense in which 122 REVIEWS. the sentence would be understood by the learner. Paragrai3h 207, that roots grow in length at the extremities, " in propor- tion as they find the requisite nutriment," might imply the popular fallacy that they grow directly by means of what they take in from the soil, which surely they do not, unless they live in the manner of Fungi. To say that the starch, etc. in a tuber or in a seed " appears to be a store of nourishment " for the early growth of the buds or the embryo, is a remark- ably over-cautious statement (how could these grow without some store of elaborated matter to feed upon ?) ; nor does the consideration that similar accumulations in the pericarps of many fruits " perish long before germination," and so do not nourish the embryo, afford to us any presumption to the con- trary, even if we could not conceive — as we readily can — of other final causes, some of them important to the continuance of the species thereby subserved. The fourth chapter, on the Collection, Preservation, and Determination of Plants, and upon Aberrations from the ordinary type or appearance, is most excellent. DR. HOOKER'S DISTRIBUTION OF ARCTIC PLANTS. The immediate subjects of the treatise ^ are the Arctic plants, of every phaenogamous species known to occur spon- taneously anywhere within the Arctic circle ; the geographical distribution of which, so far as known, is carefully indicated : 1. Within the Arctic region, under the several divisions — Europe, Asia, western America (Behring's Straits to the Mackenzie River), eastern America (Mackenzie River to Baf- fin's Bay), and arctic Greenland. 2. Without this circle, and under the general divisions of north and central European and north Asiatic Distribution, with three longitudinal subdivi- sions ; American Distribution, with appropriate subdivisions ; 1 Outlines of the Distribution of Arctic Plants. By Joseph D. Hooker. Extr. Transactions Linncean Society, of London. Vol. xxiii. pp. 251-348. 1861. (American Journal of Science and Arts, 2 ser., xxxiv. 144.) DISTRIBUTION OF ARCTIC PLANTS. 123 south European and African Distribution ; central and south Asiatic Distribution. The theory upon which the facts are collocated and discussed, and which they are thought strongly to confirm, is that of Edward Forbes, which was completed, if not indeed originated, by Darwin : — " first, that the exist- ing Scandinavian flora is of great antiquity, and that pre- vious to the glacial epoch it was more uniformly distributed over the Polar Zone than it is now ; secondly, that during the advent of the glacial period this Scandinavian vegetation was driven southward in every longitude, and even across the tropics into the south temperate zone ; and that, on the suc- ceeding warmth of the present epoch, those species that sur- vived both ascended the mountains of the warmer zones, and also returned northward, accompanied by aborigines of the countries they had invaded during their southern migration. Mr. Darwin shows how aptly such an explanation meets the difficulty of accounting for the restriction of so many Ameri- can and Asiatic arctic types to their own peculiar longi- tudinal zones, and for what is a far greater difficulty, the representation of the same arctic genera by closely allied species in different longitudes. . . . Mr. Darwin's hypothesis accounts for many varieties of one plant being found in va- rious alpine and arctic regions of the globe, by the competi- tion into which their common ancestor was brought with the aborigines of the countries it invaded. Different races sur- vived the struggle for life in different longitudes ; and these races again, afterwards converging on the zone from which their ancestor started, present there a plexus of closely allied but more or less distinct varieties, or even species, whose geographical limits overlap, and whose members, very prob- ably, occasionally breed together." A further advantage claimed for this hypothesis is, that it explains a fact brought out by Dr. Hooker in a former publication, namely : " that the Scandinavian flora is present in every latitude of the globe, and is the only one that is so." Moreover, Dr. Hooker discovers in the flora of Greenland a state of things explicable upon this hypothesis, but hardly by any other, namely : its almost complete identity with that of 124 REVIEWS. Lapland ; its general paucity, as well as its poverty in pecul- iar species ; the rarity of American species there ; the few- ness of temperate plants in temperate Greenland ; and the presence of a few of the rarest Greenland and Scandinavian species in enormously remote alpine localities of west America and the United States. Our author reasons thus : " If it be granted that the polar area was once occupied by the Scandi- navian flora, and that the cold of the glacial epoch did drive this vegetation southwards, it is evident that the Greenland individuals, from being confined to a peninsula, would have been exposed to very different conditions from those of the great continents. In Greenland many species would, as it were, be driven into the sea, that is, exterminated ; and the survivors would be confined to the southern portion of the peninsula, and, not being there brought into competition with other types, there could be no struggle for life amongst their progeny, and, consequently, no selection of better adapted varieties. On the return of heat, survivors would simply travel northwards, unaccompanied by the plants of any other country." The rustic denizens of Greenland, huddled upon the point of the peninsula during the long glacial cold, have never en- joyed the advantages of foreign travel ; those of the adjacent continents on either side have " seen the world," and gained much improvement and diversity thereby. Considering the present frigid climate of Greenland, the isotherm of 32° just impinging upon its southern point, its moderate summer and low autumnal temperature, we should rather have supposed the complete extermination of the Greenland ante-glacial flora; and have referred the Scandinavian character of the existing flora (all but eleven of the 207 arctic sj)ecies, and almost all those of temperate Greenland, being European plants) directly to subsequent immigration from the eastern continent. Several geographical considerations, and the course of the currents, which Dr. Hooker brings to view on p. 270, would go far towards explaining why Greenland should have been re-peopled from the Old rather than from the New World ; while the list (on pp. 272, 273) of upwards of 230 DISTRIBUTION OF ARCTIC PLANTS. 125 Arctic-European species, which are all likewise American plants, but are remarkable for their absence from Greenland, would indicate no small difficulty in the westward migration, and render it most probable that the diffusion of species from the Old World to the New was eastward through Asia, for the arctic no less than (as has elsewhere been shown) for the temperate plants. Was it that Greenland and the adjacent part of the American continent remained glacial longer than the rest of the zone ? And if our northern regions were thus colonized by an ancient Scandinavian flora, this seems to have been in return for a still earlier donation of American plants to Europe, to which a very few existing but numerous fossil remains bear testimony. Speculative inquiries of this sort are enticing, and the time is approaching in which they may be fruitful. Indeed, the characteristic features and the immediate in- terest and importance of the present memoir, as of others of the same general scope and interest, are found in this : 1. That the actual geographical distribution of species is something to be accounted for ; 2. That our existing species, or their originals, are far more ancient than was formerly thought, mainly if not wholly antedating the glacial period ; and, 3. That they have therefore been subject to grave climatic vicissitudes and changes. There may be many naturalists who still hesitate to accept these propositions, as there are one or two who deny them ; but these or similar conclusions have evidently been reached by those botanists, paleontolo- gists, and geologists in general who have most turned their thoughts to such inquiries, and who march foremost in the advancing movement of these sciences. In this position, the author of the present memoir, — prepossessed with Darwin's theory of the diversification of species through natural selec- tion, — having occasion to revise systematically the materials of the arctic flora, is naturally led to compare the new theory with the facts of the case in this regard ; to see how far the vicissitudes to which it is all but demonstrated that the plants of the northern hemisphere have long been subjected, and the modifications and extinctions which he thinks must have 126 REVIEWS. ensued under such grave changes and perils, during such lapse of time, may serve to explain the actual distribution of arctic species and the remarkable dispersion of many of them. That the enquiry is a legitimate and a hopeful one we must all agree, whether we favor Darwinian hypotheses or not. How well it works in the present trial we could not venture to pronounce without a far more critical examination than could now be undertaken. But there are good reasons for the opinion that this is just the ground upon which the elements of the new hypothesis figure to the best advantage. The mass of facts, so patiently and skillfully collected and digested in this essay, have a high and positive value, irre- spective of all theoretical views. We cannot undertake to offer an abstract, but may note here and there a point of in- terest. The flowering plants which have been collected within the arctic circle number 762, namely, 214 Monocoty- ledons and 548 Dicotyledons. They occupy a circumpolar belt of 10° to 14° of latitude. The only abrupt change in the vegetation anywhere along this belt is at Baffin's Bay, the opposite shores of which present, as has been already in- timated, an almost purely European flora on the east coast, but a large admixture of purely American species on the west. " Regarded as a whole, the arctic flora is decidedly Scan- dinavian ; for Arctic Scandinavia, or Lapland, though a very small tract of land, contains by far the richest arctic flora, amounting to three fourths of the whole." This would not be very surprising, since this is much the least frigid por- tion of the zone, and has the highest summer temperature ; but " upwards of three-fifths of the species, and almost all the genera of Arctic Asia and America, are likewise Lap- ponian ; " so that the Scandinavian character pervades the whole. In the section on the local distribution of plants within the arctic circle. Dr. Hooker shows that there is no close relation discoverable between the isothermal lines (whether annual or monthly) and the amount of vegetation, beyond the general fact that the scantiness of the Siberian flora is associated with a DISTRIBUTION OF ARCTIC PLANTS. 127 great soutliern bend in Asia, and its richness in Lapland, with an equally great northern bend there, of the annual isotherm of 32°. Yet " the same isotherm bends northwards in passing from eastern America to Greenland, the vegetation of which is the scantier of the two ; and it passed to the northward of Iceland, which is much poorer in species than those parts of Lapland to the southward of which it passes." A glance at the supposed former state of things would suggest the ex- planation of all that is anomalous here. "The June isothermals, as indicating the most effective temperatm^es in the arctic regions (when all vegetation is torpid for nine months, and excessively stimulated during the three others), might have been expected to indicate better the positions of the most luxuriant vegetation. But neither is this the case ; for the June isothermal of 41°, which lies within the arctic zone in Asia, wdiere the vegetation is scanty in the extreme, descends to lat. 54° in the meridian of Behr- ing's Straits, where the flora is comparatively luxuriant." The aridity of the former and the humidity of the latter district here offers an obvious explanation ; also the great severity of the winter in the former, and its mildness in the latter. And Great Britain, in which a far greater diversity of species are capable of surviving without protection than in the eastern United States under the same annual isotherms, indicates the advantage of a mean over an extreme climate in this respect, if only there be a certain amount of summer heat. For lack of that, doubtless, very many of the introduced denizens of Britain would soon disappear, if deprived of human care. " The^ northern limit to which vegetation extends varies in every longitude ; the extreme is still unknown ; it maj^ in- deed, reach to the pole itself. Phsenogamic plants, however, are probably nowhere found far north of lat. 81°. Seventy flowering plants are found in Spitzbergen ; and Sabine and Ross collected nine on AValden Island, towards its northern extreme, but none on Ross's Islet, fifteen miles further to the north. " Saxifraga oppositifolia is probably the most ubiquitous, and may be considered the commonest and most arctic flower- 128 REVIEWS. ing plant." There are only eight or nine phaenogamous species peculiar to the arctic zone, and only one peculiar genus, namely, the grass Pleuropogon.^ Of the 7G2 found south of the circle, all but 150 have advanced beyond lat. 40^ N. in some part of the world ; about 50 of them are identified as natives of the mountainous regions of the tropics, and 105 as inhabiting the south temperate zone. " The proportion of species which have migrated southward in the Old and New World also bear a fair relation to the facilities for migration presented by the different continents." The tables given to illustrate this " present in a very striking point of view the fact of the Scandinavian flora being the most widely distributed over the world. The Mediterranean, south African, Malayan, Australian, and all the floras of the New World, have narrow ranges compared with the Scandi- navian, and none of them form a prominent feature in any other continent than their own. But the Scandinavian not only girdles the globe in the arctic circle, and dominates over all others in the north temperate zone of the Old World, but intrudes conspicuously into every other temperate flora, whether in the northern or southern hemis23here, or on the Alps of tropical countries. ... In one respect this migration is most direct in the American meridian, where more arctic species reach the highest southern latitudes. This I have accounted for (' Flora Antarctica,' p. 230) by the continuous chain of the Andes having favored their southern dispersion." In presenting the actual number of arctic si^ecies, and in delineating their geographical ranges, the question, what are to be regarded as species, becomes all-important. As to this, it does not so much matter what scale is adoj^ted, as to know clearly wdiat the adopted scale is. Here we are not left in 1 Douglasia is mentioned in another place (p. 269) as an absolutely pe- culiar arctic or arctic-alpine genus of eastern America. But we have con- sidered this genus as identical with Gregoria, of Duby. It would appear as if these two genera were established in the same year, since Lindley himself, in the "Botanical Register," refers to Brando's Journal for Jan- uary, 1828, for his original article. But this article will be found in the volume of that Journal for 1827 ; so that the name Douglasia is to be adopted, if the genus is sufficiently distinct from Androsace. DISTRIBUTION OF ARCTIC PLANTS. 129 doubt. Taking European botanists by number, we are confi- dent that nine out of ten would have enlarged the list of TG2 phaenogamous arctic species to 800 or more, and would not have recognized a goodly number of the synonyms adduced, thereby considerably affecting the assigned ranges, especially into temperate and austral latitudes. In this regard we should side with Dr. Hooker on the whole, but with differ- ences and with questionings — with halting steps following his bold and free movement, but probably arriving at the same goal at length. Indeed, we freely receive the view which Dr. Hooker presents as aj)propriate to his particular purpose, and as the most useful expression of our knowledge of the rela- tionships of the plants in question, when collocated in refer- ence to the ideas upon wdiich this memoir is based. That is : " if, with many botanists, we consider these closely allied va- rieties and species as derived by variation and natural selec- tion from one parent form at a comparatively modern epoch, we may with advantage, for certain purposes, regard the aggregate distribution of such very closely allied species as that of one plant." " An empirical grouping of allied plants, for the purposes of distribution, may thus lead to a practical solution of difficulties in the classification and synonymy of species. My thus grouping names must not be regarded as a committal of myself to the opinion that the plants thus grouped are not to be held as distinct species. . . . My main object is to show the affinities of the polar plants, and I can best do this bykeej^ing the specific idea comprehensive." And further : " I wish it then to be clearly understood, that the catalogue here appended is intended to include every species hitherto found within the arctic circle, together with those most closely allied forms which I believe to have branched off from one common parent within a comparativel}- recent geo- logical epoch, and that immediately previous to the glacial period or since then" (p. 279). All we coidd ask more would be some distinction (typographical or other), to mark, 1, undoubted and complete synonyms ; 2, mere variations or states, local or otherwise, or undoubted varieties ; 3, such as, theory apart, would claim to be regarded as distinct but 130 REVIEWS. closely related species. For example ; to take one order, while lilunanthus ??zi?ior may well be considered as "not a sufficiently constant form to rank as a race even," while Ximo- sella tenuifolia could rank for no more than a race, and while Castilleia se2'>tentrionalis and C. iiallida^ we are now con- vinced, however distinct in this single character, differ only (and inconstantly) in the relative development of the galea, we think it likely that Pedicularis lanata^ Willd., does not rightfully merge in P. Mrsiita this side of the glacial period, although it perhaps may into P. Langsdorffii^ and that into P, Sudetica. But this is no place for criticisms upon the limitation of species, upon which the opinions of botanists will so greatly depend vipon the amount of their materials, and upon which the best considered opinions must be subject to frequent revisal. Nor does the value of the present memoir at all depend upon the settlement of such points. To the philosophical naturalist, as to the archaeologist, just now the most interesting and pregnant epoch of the world's natural history is that immediately antecedent to the present, that near past from which the present has proceeded, and upon which so much light, from very diverse sources, is now being concentrated: towards its elucidation the memoir we have been considering is a very valuable contribution. ALPHONSE DE CANDOLLE ON THE VARIATION AND DISTRIBUTION OF SPECIES. This is the title of a paper ^ by Monsieur De Candolle grow- ing out of his study of the Oaks. It was published in the November number of the " Bibliothoque Universelle," and separately issued as a pamphlet. A less inspiring task could hardly be assigned to a botanist than the systematic elabora- tion of the genus Quercus and its allies. The vast materials assembled under De Candolle's hands, while disheartening ^ Etude sur VEspece, a V occasion d'une Revision de la Famille des Cupu- lifers. Par M. Alplionse De Candolle. (American Journal of Science and Arts, 2 ser., xxxv. 431.) VARIATION AND DISTRIBUTION OF SPECIES. 131 for their bulk, offered small lioj^e o£ novelty. The subject was both extremely trite and extremely difficult. Hapjnly it occurred to De Candolle that an interest might be imparted to an onerous undertaking, and a work of necessity be turned to good account for science, by studying the Oaks in view of the question of Species. ^ What this term Species means, or should mean, in natural history, what the limits of species, mter se or chronologically, or in geographical distribution, their modifications, actual or probable, their origin, and their destiny, — these are ques- tions which surge up from time to time ; and now and then in the progress of science they come to assume a new and hope- ful interest. Botany and Zoology, Geology, and what our author, feeling the want of a new term, proposes to name Epio7itology,^ all lead up to and converge into this class of questions, while recent theories shape and point the discus- sion. So we look with easier interest to see what liirht the study of Oaks, by a very careful, experienced, and conserva- tive botanist, particularly conversant with the geographical relation of plants, may throw upon the subject. The course of investigation in this instance does not differ from that ordinarily pursued by working botanists ; nor, in- deed, are the theoretical conclusions other than those to which a similar study of other orders might not have equally led. The Oaks afford a very good occasion for the discussion of questions which press upon our attention, and perhaps they offer peculiarly good materials on account of the number of fossil species. Preconceived notions about species being laid aside, the 1 A name which, at the close of his article, De CantloUe proposes for the study of the succession of organized beings, to comprehend, therefore, palaeontology and everything included under what is called geograpliical botany and geographical zoology, — the whole forming a science parallel to geology, — the latter devoted to the history of unorganized hodies, the former, to that of organized beings, as respects origin, distribution, and succession. We are not satisfied with the word, notwithstanding the prece- dent of jDateon^o^o^rz/y since ontology, the science of being, has an estab- lished meaning as referring to mental existence, — i. c., is a synonym or a department of metaphysics. 132 REVIEWS. specimens in hand were distributed, according to their ob- vious resemblances, into groups of apparently identical or nearly identical forms, which were severally examined and compared. Where specimens were few, as from countries little explored, the work was easy, but the conclusions, as will be seen, of small value. The fewer the materials, the smaller the likelihood of forms intermediate between any two, and — what does not appear being treated upon the old law-maxim as non-existent — species are readily enough defined. Where, however, specimens abound, as in the case of the Oaks of Eu- rope, of the Orient, and of the United States, of which the specimens amounted to hundreds, collected at different ages, in varied localities, by botanists of all sorts of views and pre- dilections, — here alone were data fit to draw useful conclu- sions from. Here, as De Candolle remarks, he had every advantage, being furnished with materials more complete than any one person could have procured from his own herboriza- tions, more varied than if he had observed a hundred times over the same forms in the same district, and more impartial than if they had all been amassed by one person with his own ideas or predispositions. So that vast herbaria, into which contributions from every source have flowed for years, furnish the best possible data — at least are far better than any prac- ticable amount of personal herborization — for the compara- tive study of related forms occurring over wide tracts of terri- tory. But as the materials increase, so do the difficulties. Forms, which appeared totally distinct, approach or blend through intermediate gradations ; characters, stable in a lim- ited number of instances or in a limited district, prove unsta- ble occasionally, or when observed over a wider area ; and the practical question is forced upon the investigator, — what here is probably fixed and specific, and what is variant, pertaining to individual, variety, or race ? In the examination of these rich materials, certain char- acters were found to vary upon the same branch, or upon the same tree, sometimes according to age or develojDment, some- times irrespective of such relations or of any assignable rea- sons. Such characters, of course, are not specific, although VARIATION AND DISTRIBUTION OF SPECIES. 133 many of them are such as would have been expected to be constant in the same species, and are such as generally enter into specific definitions. Variations of this sort, De Candolle, with his usual painstaking, classifies and tabulates, and even expresses numerically their frequency in certain species. The results are brought well to view in a systematic enumera- tion : — (1) Of characters which frequently vary upon the same branch : over a dozen such are mentioned. (2) Of those which sometimes vary upon the same branch : a smaller number of these are mentioned. (3) Those so rare that they might be called monstrosi- ties. Then he enumerates characters, ten in number, which he has never found to vary on the same branch, and which therefore may better claim to be employed as specific. But, as among them he includes the duration of the leaves, tlie size of the cupule, and the form and size of its scales, which are by no means wholly uniform in different trees of the same species, even these characters must be taken with al- lowance. In fact, having first brought together, as groups of the lowest order, those forms which varied upon the same stock, he next had to combine similarly various forms which, though not found associated upon the same branch, were thoroughly blended by intermediate degrees. " The lower groups (varieties or races) being thus constituted, I have given the rank of species to the groups next above these, which differ in other respects, i, e., either in characters which were not found united upon certain individuals, or in those which do not show transitions from one individual to another. For the Oaks of regions sufficiently known, the species thus formed rest upon satisfactory bases, of which the proof can be furnished. It is quite otherwise with those which are represented in our herbaria by single or few specimens. These are provisional species, — species which may hereafter fall to the rank of simple varieties. I have not been in- clined to prejudge such questions ; indeed, in this regard, I am not disposed to follow those authors whose tendency is, as they say, to reunite species. I never reunite them without proof in eacli partic- 134 REVIEWS. ular case ; while the botanists to whom I refer do so on the ground of analogous variations or transitions occurring in the same genus or in the same family. For example, resting on the fact that Quer- cus Hex, Q. coccifera, Q. acutifolla, etc., have the leaves sometimes entire and sometimes toothed upon the same branch, or present transitions from one tree to another, I might readily have united my Q. TlcqnixaJmiensis to Q. Sartorii of Liebmann, since these two differ only in their entire or their toothed leaves. From the fact that the length of the peduncle varies in Q. Eohur and many other Oaks, I might have combined Q. Seemanni'i, Liebm., with Q. sallcifolia, Nee. I have not admitted these inductions, but have demanded visible proof in each particular case. Many species are thus left as provisional ; but in jDroceeding thus, the progress of the science will be more regular, and the synonymy less dependent upon the caprice or the theoretical opinions of each author." This is safe and to a certain degree judicious, no doubt, as respects published species. Once admitted, they may stand until they are put down by evidence, direct or circumstantial. Surely a species may rightfully be condemned on good cir- cumstantial evidence. But what course does De CandoUe pursue in the case — of every-day occurrences to most w^ork- ing botanists having to elaborate collections from countries not so well explored as Europe — when the forms in question, or one of the two, are as yet unnamed ? Does he introduce as a new species every form which lie cannot connect by ocular proof with a near relative, from which it differs only in particulars which he sees are inconstant in better known species of the same group ? We suppose not. But if so, little improvement for the future upon the state of things revealed in the following paragraph can be expected. " In the actual state of our knowledge, after having seen nearly all the original specimens, and in some species as many as 200 rep- resentatives from different localities, I estimate that, out of the 300 species of OupulifercB which will be enumerated in the ' Prodromus,' two thirds at least are provisional species. In general, when we consider what a multitude of species were described from a single specimen, or from the forms of a single locality, of a single country, or are badly described, it is difficult to believe that above one third of the actual species in botanical works will remain unchanged." VAlilA TION AND DISTRIBUTION OF SPECIES. 135 Such being the results of the want of adequate knowledge, how is it likely to be when our knowledge is largely in- creased ? The judgment of so practised a botanist as De Can- dolle is important in this regard, and it accords with that of other botanists of equal experience. " They are mistaken," he pointedly asserts, " who repeat that the greater part of our species are clearly limited, and that the doubtful species are in a feeble minority. This seemed to be true, so long as a genus was imperfectly known, and its species were founded upon few specimens, that is to say, were provisional. Just as we come to know them better, intermediate forms flow in, and doubts as to specific limits augment." De Candolle insists, indeed, in this connection, that the higher the rank of the groups, the more definite their limita- tion, or, in other terms, the fewer the ambiguous or doubtful forms ; that genera are more strictly limited than sjiecies, tribes than genera, orders than tribes, etc. We are not con- vinced of this. Often where it has appeared to be so, advanc- ing discovery has brought intermediate forms to light, per- plexing to the systematist. "They are mistaken," we think more than one systematic botanist will say, " who repeat that the greater part of our natural orders and tribes are absolutely limited," however we may agree that we will limit them. Provisional genera we suppose are proportionally hardly less common than provisional species ; and hundreds of genera are kept up on considerations of general propriety or general convenience, although well known to shade off into adjacent ones by complete gradations. Somewhat of this greater fixity of higher groups, therefore, is rather apparent than real. On the other hand, that varieties should be less definite than species, follows from the very terms employed. They are ranked as varieties, rather than species, just because of their less definiteness. Singular as it may appear, we have heard it denied that spontaneous varieties occur. De Candolle makes the impor- tant announcement that, in the Oak genus, the best known species are just those which present the greatest number of 136 REVIEWS. spontaneous varieties and sub-varieties. The maximum is found in Q. Rohiu\ with twenty-eight varieties, all spon- taneous. Of Q. Lusltanica eleven varieties are enumerated, of Q. CaUij)rinos ten, of Q. coccifera eight, etc. And he significantly adds that " these very species which offer such numerous modifications are themselves ordinarily surrounded by other forms, provisionally called species, because of the absence of known transitions or variations, but to which some of these will probably have to be joined hereafter." The in- ference is natural, if not inevitable, that the difference be- tween such species and such varieties is only one of degree, either as to amount of divergence, or of hereditary fixity, or as to the frequency or rarity, at the present time, of interme- diate forms. This brings us to the second section of De Candolle's article, in which he passes on, from the observation of the present forms and affinities of Cupuliferous plants, to the considera- tion of their probable history and origin. Suffice it to say, that he frankly accepts the inferences derived from the whole course of observation, and even contemplates with satisfac- tion a probable historical connection between congeneric spe- cies. He accepts and, by various considerations drawn from the geographical distribution of European Cupiiliferce^ for- tifies the conclusion — long ago arrived at by Edward Forbes — that the present species, and even some of their varieties, date back to about the close of the Tertiary epoch, since which time they have been subject to frequent and great changes of habitation or limitation, but without appreciable change of specific form or character ; that is, without pro- founder changes than those within which a species at the present time is known to vary. Moreover, he is careful to state that he is far from concluding that the time of the appearance of a species in Europe at all indicates the time of its origin. Looking back still further into the Tertiary epoch, of which the vegetable remains indicate many analo- gous, but few, if any, identical forms, he concludes, with Heer and others, that specific changes of form, as well as changes of station, are to be presumed. And finally, that " the theory VARIATION AND DISTRIBUTION OF SPECIES. 1S7 of a succession of forms through the deviation of anterior forms is the most natural hypothesis, and the most accordant with the known facts in palaeontology, geographical botany and zoology, of anatomical structure and classification : but direct proof of it is wanting, and moreover, if true, it must have taken place very slowly ; so slowly indeed, that its efPects are discernible only after a lapse of time far longer than our historic epoch." In contemplating the present state of the species of Cupu- liferce in Europe, De Candolle comes to the conclusion that, while the Beech is increasing, and extending its limits south- ward and westward (at the expense of Conifer ca and Birches), the Common Oak, to some extent, and the Turkey Oak de- cidedly, are diminishing and retreating, and this wliolly ir- respective of man's agency. This is inferred of the Turkey Oak from the great gaps found in its present geographical area, which are otherwise inexplicable, and which he regards as plain indications of a partial extinction. Community of descent of all the individuals of species is of course implied in these and all similar reasoninsfs. An obvious result of such partial extinction is clearly enough brought to view. The European Oaks (like the American species) greatly tend to vary, — that is, they mani- fest an active disj^osition to produce new forms. Every form tends to become hereditary, and so to pass from the state of mere variation to that of race ; and of these competing in- cipient races some only will survive. Qitercus Eohur offers a familiar illustration of the manner in which one form may in the course of time become separated into two or more distinct ones. To Linnseus this Common Oak of Europe was all of one species. But of late j^ears the greater number of European botanists have regarded it as including three species, (^.^;r(/^/;^- culata, Q. sessiliflora^ and Q. j^uhescens. De Candolle looks with satisfaction to the independent conclusion which lie reached from a long and patient study of the forms (and which Webb, Gay, Bentham and others had equally reached), that the view of Linnaeus was correct, inasmuch as it goes to 138 REVIEWS. show that the idea and the practical application of the term species have remained unchanged during the century which has elapsed since the publication of the "Species Plantarum." But the idea remaining unchanged, the facts might appear under a different aspect, and the conclusion be different, un- der a slight and very supposable change of circumstances. Of the twenty-eight spontaneous varieties of Q. Bohui\ whicli De Candolle recognizes, all but six, he remarks, fall naturally under the three sub-species, jiedunculata., sessiliflora^ and piihesceiis, and are therefore forms grouped around these as centres ; and, moreover, the few connecting forms are by no means the most common. Were these to die out, it is clear that the three forms which have already been so fre- quently taken for species, would be what the group of four or five provisionally admitted species which closely surround Q, Robur (see p. 435) now are. The best example of such a case, as having in all probability occurred, through geo- graphical segregation and partial extinction, is that of the Cedar, thus separated into the Deodar, the Lebanon, and the Atlantic Cedars, — a case admirably w^orked out by Dr. Hooker two or three years ago.^ A special advantage of the CupuUferm for determining the probable antiquity of existing species in Europe, De Can- dolle finds in the size and character of their fruits. However it may be with other plants (and he comes to the conclusion generally that marine currents and all other means of distant transport have played only a very small part in the actual dispersion of species), the transport of acorns and chestnuts by natural causes across an arm of the sea in a condition to germinate, and much more the spontaneous establishment of a forest of Oaks or Chestnuts in this way, De Candolle con- ceives to be fairly impossible in itself, and contrary to ail experience. From such considerations, /. e., from the actual dispersion of the existing species, with occasional aid from Post-tertiary deposits, it is thought to be shown that the prin- cipal Cupuliferce of the Old World attained their actual ex- tension before the present separation of Sicily, Sardinia and Corsica, or of Britain, from the European continent. ^ Natural History Review, Jan., 1862. VARIATION AND DISTRIBUTION OF SPECIES. 139 This view once adopted, and this course once entered upon, has to be pursued farther. Quercus Rohiir of Europe with its bevy of admitted derivatives, and its attending species only provisionally admitted to that rank, is very closely related to certain species of eastern Asia, and of Oregon and California, — so closely that " a view of the specimens by no means forbids the idea that they have all originated from Q. Rohiir^ or have originated, with the latter, from one or more preceding forms so like the present ones that a natural- ist could hardly know whether to call them species or varieties." Moreover, there are fossil leaves from diluvian deposits in Italy, figured by Gaudin, which are hardly distinguishable from those of Q. Rohm' on the one hand, and from those of Q, DougJasis etc., of California on the other. No such leaves are found in any Tertiary deposit in Europe ; but such are found of that age, it appears, in northwest America, where their remote descendants still flourish. So that the probable genealogy of Q, Rohur., traceable in Europe up to the commencement of the present epoch, looks eastward and far into the past on far distant shores. Q. llex^ the Evergreen Oak of southern Europe and north- ern Africa, reveals a similar archseology ; but its presence in Algeria leads De Candolle to re^fard it as a much more ancient denizen of Europe than Q. Rohur ; and a Tertiary Oak, Q, ilicoides, from a very old Miocene bed in Switzer- land, is tliought to be one of its ancestral forms. This high antiquity once established, it follows almost of course that the very nearly related species in central Asia, in Japan, in Cal- ifornia, and even our own Live Oak with its jMexican rela- tives, may probably enough be regarded as early offshoots from the same stock with Q. Hex. In brief, — not to continue these abstracts and remarks, and without reference to Darwin's particular theory (which De Candolle at the close very fairly considers), — if existing species, or many of them, are as ancient as they are now generally thought to be, and were subject to the physical and geographical changes (among them the coming and the going of the Glacial epoch) which this antiquity implies; if in 140 BE VIEWS. former times they were as liable to variation as they now are ; and if the individuals of the same species may claim a common local origin, — then we cannot wonder that " the theory of a succession of forms by deviations of anterior forms " should be regarded as " the most natural hypothesis," nor at the general advance made towards its acceptance in some form or other. The question being, not, how plants and animals originated, but, how came the existing animals and plants to be just where they are and what they are ; it is plain that naturalists interested in such inquiries are mostly looking for the answer in one direction. The general drift of opinion, or at least of expectation, is exemplified by this essay of De Candolle ; and the set and force of the current are seen by noticing how it carries along naturalists of widely different views and prepos- sessions, — some faster and farther than others, — but all in one way. The tendency is, we may say, to extend the law of continuity, or something analogous to it, from inorganic to organic nature, and in the latter to connect the present with the past in some sort of material connection. The generali- zation may indeed be expressed so as not to assert that the connection is genetic, as in Mr. Wallace's formula : " Every species has come into existence coincident both in time and space with preexisting closely allied species." Edward Forbes, who may be called the originator of this whole line of inquiry, long ago expressed a similar view. But the only material sequence we know, or can clearly conceive, in plants and animals, is that from parent to progeny ; and, as Dc Candolle implies, the origin of species and that of races can hardly be much unlike, nor governed by other than the same laws, whatever these may be. The progress of opinion uj^on this subject in one generation is not badly represented by that of De Candolle himself, who is by no means prone to adopt new views without much con- sideration. In an elementary treatise published in the year 1835, he adopted, and, if we rightly remember, vigorously maintained, Schouw's idea of the double or multiple origin of species, at least of some species, — a view which has been VARIATION AND DISTRIBUTION OF SPECIES. 141 carried out to its ultimate development only perhaps by Agassiz, in the denial of any necessary genetic connection among tlie individuals of the same species, or of any original localization more restricted than the area now occupied by the species. But in 1855, in his " Geographic Botanique," the multiple hypothesis, although in principle not abandoned, is seen to lose its point, in view of the probable high antlcpiity of existing species. The actual vegetation of the world being now regarded as a continuation, through numerous geological, geographical, and more recently historical, changes, of ante- rior vegetations, the actual distribution of plants is seen to be a consequence of preceding conditions and geological consid- erations, and these alone may be expected to explain all the facts, many of them so curious and extraordmary, of the actual geographical distribution of the species. In the present essay, not only the distribution but the origin of congeneric species is regarded as something derivative ; whether derived by slow and very gradual changes in the course of ages, according to Darwin, or by a sudden, inexplicable change of their Tertiary ancestors, as conceived by Heer, De Candolle hazards no opinion. It may, however, be inferred that he looks upon "natural selection" (which he rather underrates) as a real, but insufficient, cause ; while some curious remarks (pp 57, 58), upon the number of monstrosities annually pro- duced, and the possibility of their enduring, may be regarded as favorable to Heer's view. As an index to the progress of opinion in the direction re- ferred to, it will be interesting to compare Sir Charles Lyell s well-known chapters of twenty or thirty years ago, m which the permanence of species was ably maintained, with his treat- ment of the same subject in a work just issued m England, which, however, has not yet reached us. ^ ^ A belief in the derivation of species may be maintained alon- with a conviction of great persistence of specific char- acters. This is the idea of the excellent Swiss vegetable pale- ontologist Heer, who imagines a sudden change of specihc type at certain periods, and perhaps is that of Pictet. lal- coner adheres to somewhat similar views in his elaborate 142 REVIEWS. paper on Elephants, living and fossil, in the " Natural History Review " for January last. Noting that " there is clear evi- dence of the true Mammoth having existed in America long after the period of the northern drift, when the surface of the country had settled down into its present form," and also in Europe so late as to have been a contemporary of the Irish Elk, and on the other hand that it existed in England so far back as before the deposition of the boulder Clay ; also that four well-defined species of fossil Elephant are known to have existed in Europe ; that " a vast number of the remains of three of these species have been exhumed over a large area in Europe ; and, even in the geological sense, an enormous in- terval of time has elapsed between the formation of the most ancient and the most recent of these deposits, quite sufficient to test the persistence of specific characters in an Elephant," he presents the question : " Do then the successive Elephants occurring in these strata show any signs of a passage from the older form into the newer ? " To which the reply is : " If there is one fact which is im- pressed on the conviction of the observer with more force than any other, it is the persistence and uniformity of the charac- ters of the molar teeth in the earliest known Mammoth and his most modern successor. . . . Assuming the observation to be correct, what strong proof does it not afford of the per- sistence and constancy, throughout vast intervals of time, of the distinctive characters of those organs which are most con- cerned in the existence and habits of the species ? If we cast a glance back on the long vista of physical changes w^hich our planet has undergone since the Neozoic Epoch, we can nowhere detect signs of a revolution more sudden and pronounced, or more important in its results, than the intercalation and sud- den disappearance of the glacial period. Yet the ' dicyclo- therian ' Mammoth lived before it, and passed through the ordeal of all the hard extremities it involved, bearing his organs of locomotion and digestion all but unchanged. Tak- ing the group of four European fossil species above enumer- ated, do they show any signs in the successive deposits of a transition from the one form into the other ? Here again VARIATION AND DISTRIBUTION OF SPECIES. l43 the result of my observation, in so far as it has extended over the European area, is, that the specific characters of the molars are constant in each, within a moderate range of varia- tion, and that we nowhere meet with intermediate forms." . . . Dr. Falconer continues (p. 80) : — " The inferences which I draw from these facts are not opposed to one of the leading propositions of Darwin's theory. AVith liim, I have no faith in the opinion that the Mammoth and other extinct Elephants made their appearance suddenly, after the type in which their fossil remains are presented to us. The most rational view seems to be, that they are in some shape the modified descendants of earlier progenitors. But if the asserted facts be correct, they seem clearly to indicate that the ohler elephants of Europe, such as E. meridionalis and E, antiquns, were not the stocks from which the later species, E. prwiir/enius and E. Afrlcanus, sprung, and that we must look elsewhere for their origin. The nearest affinity, and that a very close one, of the European E. meridionalis is witli the Miocene E. planifrons of India ; and of E. priviir/enius, with the existing India species. " Another reflexion is equally strong in my mind, — that the means which have been adduced to exjilain the origin of the species by ' Natural Selection,' or a process of variation from external in- fluences, are inadequate to account for the phenomena. The law of phyllotaxis, which governs the evolution of leaves aromid the axis of a plant, is as nearly constant in its manifestation as any of the physical laws connected with the material world. Each instance, however different from another, can be shown to be a term of some series of continued fractions. When this is coupled with the geo- metrical law governing the evolution of form, so manifest in some departments of the animal kingdom, e. g., the spiral shells of the Mollusca, it is difficult to believe that there is not, in nature, a deeper-seated and innate principle, to the operation of which Natural Selection is merely an adjunct. The whole range of the Manunalia, fossil and recent, cannot furnish a species wliich has had a wider geographical distribution, and passed through a longer term of time, and through more extreme changes of climatal conditions, than tlie Mammoth. If species are so unstable, and so susceptible of nmta- tion through such influences, why does that extinct form stand out so signally a monument of stability ? By his admirable researches and earnest writings, Darwin has, beyond all his cotcmporaries, given 144 REVIEWS. an impulse to the philosophical investigation of the most backward and obscure branch of the biological sciences of his day ; he has laid the foundations of a great edifice ; but he need not be surprised, if, in the progress of erection, the superstructure is altered by his suc- cessors, like the Duomo of Milan from the Roman to a different style of architecture." Entertaining ourselves the opinion that something more than natural selection is requisite to account for the orderly production and succession of species, we offer two incidental remarks upon the above extract. First, we find in it — in the phrase " Natural Selection, or a process of variation from external influences " — an example of the very common confusion of two distinct things, namely, variation and natural selection. The former has never yet been shown to have its cause in " external influences," nor to occur at random. As we have elsewhere insisted, if not inex- plicable, it has never been explained ; all we can yet say is, that plants and animals are prone to vary, and that some con- ditions favor variation. Perhaps in this Dr. Falconer may yet find what he seeks : for '' it is difficult to believe that there is not in [its] nature, a deeper-seated and innate prin- ciple, to the operation of which Natural Selection is merely an adjunct." The latter, which is the e?ise7nhle of the external influences, including the competition of the individuals them- selves, picks out certain variations as they arise, but in no proper sense can be said to originate them. Secondly, although we are not quite sure how Dr. Falconer intends to apply the law of phyllotaxis to illustrate his idea, we fancy that a pertinent illustration may be drawn from it, in this way. There are two species of phjdlotaxis, perfectly distinct, and, we suppose, not mathematically reducible the one to the other, — namely, 1, that of alternate leaves, with its varieties ; and 2, that of verticillate leaves, of which opposite leaves present the simplest case. That, although generally constant, a change from one variety of alternate phyllotaxis to another should occur on the same axis, or on successive axes, is not surprising, the different sorts being terms of a regular series, — although indeed we have not the VARIATION AND DISTRIBUTION OF SPECIES. 1^7 least idea as to how the change from the one to the takcT- comes to pass. But it is interesting, and in this cor-i'e now perhaps instructive, to remark that, while some dIcot;-;onsistent plants hold to the verticillate, i. tures and consumes animals, but perhaps avails itself of the nitrogenous matter only when passing into decay. Drosophyllum, a rare representative of the order, confined to Portugal and Morocco, grows on the sides of dry hills near Oporto ; so that, as to station, it is the very counterpart of Aldrovanda. Its leaves are long and slender, in the manner of our Dvosera filiformis., and are covered with much larger glands. To these, flies adhere in vast numbers. " The latter fact is well known to the villagers, who call the plant the ' fly-catcher,' and hang it up in their cottages for this purpose." Mr. Darwin found the glands incapable of movement, and their behavior in some other respects differs from that of Drosera ; but they equally secrete a digestive juice. Insects usually drag off this secretion instead of being fixed on the glands by it ; but their fate is no better ; for as the poor ani- mal crawls on and these viscid drops bedaub it on all sides, it sinks down at length exhausted or dead, and rests on a still more numerous set of small sessile glands which thickly cover the whole surface of the leaf. These were till then dry and inert, but as soon as animal matter thus comes in contact with them, they also secrete a digestive juice, which, as Mr. Darwin demonstrated, has the power of dissolving bits of coagulated albumen, cartilage, or meat, with even greater readiness than that of Drosera. DARWIN'S INSECTIVOROUS PLANTS. 209 Mr. Darwin next records various observations and experi- ments upon more ordinaxy glandular hairs of several plants. To certain Saxifrages his attention was naturally called, on account of the presumed relationship of Di^oseracece to this genus. He declares that " their glands absorb matter from an infusion of raw meat, from solutions of nitrate and carbon- ate of ammonia, and apparently from decayed insects." To such plants the vast number of little insects caught may not be useless, as they may be to many other plants (Tobacco, for instance) with sticky glands, in which Mr. Darwin could detect no power of absorption. The prevalent idea, that glandular hairs in general serve merely as secreting or excret- ing organs, and are of small or no account to the plant, must now be reconsidered. Those of the common Chinese Primrose (^Primula Sinensis)^ although indifferent to animal infusions, were found to absorb quickly both the solution and vapor of carbonate of ammonia. Now, as rain-water contains a small percentage of ammonia, and the atmosphere a minute quan- tity of the carbonate or nitrate, and as a moderate-sized plant of this Primrose was ascertained (by estimate from a count on small measured surfaces by Mr. Francis Darwin) to bear between 2| to 3 millions of these glands, it begins to dawn upon us that these multitudinous organs are neither mere excrescences nor outlets, nor in any just sense insignificant. Mr. Darwin next investigates the densely crowded short glandular hairs, with their secretions, which form the buttery surface of the face of the leaves of Pinguicula, the Butter- wort. He finds that the leaves of the common Butterwort have great numbers of small insects adhering to them, as also grains of pollen, small seeds, etc. ; that most substances so lodged or placed, if yielding soluble matter to the glands, excite them to increased secretion ; but that if non-nitrogenous the viscid fluid poured out is not at all acid, while if nitro- genous it invariably has an acid reaction and is more copious ; that in this state it will quickly dissolve the muscles of insects, meat, cartilage, fibrin, curds of milk, etc. ; that when the surface of a plane leaf is fed, by placing upon it a row of flies along one margin, this margin, but not the other, folds 210 REVIEWS. over within twenty hours to envelop them ; and when placed on a medial line, a little below the apex, both margins incurve. He concludes " that Pinguicula vulgaris^ with its small roots, is not only supported to a large extent by the extraordinary number of insects which it habitually captures, but likewise draws some nourishment from the pollen, leaves, and seeds of other plants, which often adhere to its leaves. It is therefore partly a vegetable as well as an animal feeder." The leaves in one or two other species were found capable of greater and more enduring inflection, and the glands excitable to increased secretion even by bodies not yielding soluble nitrogenous matter. The aquatic type of this family is Utricularia ; and the bladder-bearing species of this genus are to Pinguicula nearly what Aldrovanda is to Dionaea and Drosera — the bladders imprisoning minute aquatic animals by a mechanism almost as ino:enious as that of Dionsea itself. Observations of the same kind were made in this country by Mrs. Treat, of Vine- land, New Jersey, before Mr. Darwin's investigations were made known. These submerged aquatic stomachs, ever del- uged with water, apparently do not really digest their cap- tures, but merely absorb the products of their decay. The same must in all probability be said of such Pitcher- plants as Sarracenia and Darlingtonia, which Mr. Darwin merely alludes to at the close of his volume but does not treat of. Nepenthes, how^ever, according to Dr. Hooker's investi- gations, has attained a higher dignity, and converted its pitcher into a stomach. This parallelism, and this higher and lower mode of appropriating organic products by each of the three well-marked carnivorous families of plants, are highly sug- gestive. In concluding this notice of a book for which we have no room to do justice, — but wdiich is sure to be in the hands of many interested readers, — there is something to be said in regard to the discovery of the lure in some of our Sarracenias. We have by degrees to discover our discoverers. In this Journal, only so far back as the number for August, 1873, is a notice of the discovery of a sweet secretion at the orifice of DARWIN'S INSECTIVOROUS PLANTS. 211 tlie pitcher o£ Sarracenia flava^ by Mr. B. F. Grady, of Clin- ton, North Carolina (in the article by an oversight called " Mr. Hill "), which effectively lures flies to their destruction. This statement, made in a letter, had been for several months in our hands, awaiting the opportunity of confirmation, when an allusion to the same thing appeared in the English edition of LeMaout and Decaisne's System of Botany, without refer- ence to any source, and on inquiry we learned that the author- ity for the statement was forgotten. But early in the follow- ing year, when the monograph of the order appeared in the last volume of De Candolle's " Prodromus," a reference was found to a paper by Dr. Macbride in the Transactions of the Linnsean Society. His observations (made upon S, vario- lar is^, it appears, were communicated to Sir J. E. Smith, read before the Linnsean Society in 1815, and published soon after. They are referred to by his surviving friend and associate, Eliott, in his well-known work, and therefore need not have gone to oblivion, or needed rediscovery here in our days by Mr. Grady and Dr. Mellichamp, the latter greatly extending our knowledge of the subject. Probably the main facts were all along popularly known in the regions these species affect, and where their use as fly-traps is almost im- memorial. But the gist of these remarks is, that a colleague has just called our attention to an earlier publication than that of Dr. Macbride, namely, an article on " Certain Vege- table Muscicapse," by Benjamin Smith Barton (one of our botanical fathers), published in " Tilloch's Philosophical Magazine " for June, 1812. Among other matters not bear- ing directly upon this point, he says of Sarracenia, without reference to any particular species : " A honeyed fluid is secreted or deposited on the inner surface of the hollow leaves, near their J'aux or opening ; and this fluid allures great num- bers of the insects which they are found to contain into the ascidia." Here is earlier publication by three years. Yet we suspect that Dr. Barton knew little about it at first hand, and we find clear evidence that he had not anticipated Dr. Macbride. All his references have an indefiniteness quite in contrast with 212 BE VIEWS. Dr. Macbricle's narrative ; he says that " some if not all the species of the genus appear to possess a kind of glandular function," without mentioning those that have it, or the absence of it in the only species growing around him at the north ; and he adds that he " was entirely unacquainted with this curious economy . . . when I published the first edition of my ' Elements of Botany,' and even when I printed the appendix (in vol. i.) to the second edition of this work." Now his paper is dated September 11, 1811 ; and the volume referred to, as just printed, is dated 1812. But Macbride states that his observations were chiefly made 1810 and 1811 ; he corresponded intimately with Eliott, through whom, if not directly, his observations would probably find their way at once to the Philadelphia naturalists. NAUDIN ON THE NATURE OF HEREDITY AND VARIABILTIY IN PLANTS.^ Why is it the nature and essence of species to breed true, and why do species sometimes vary ? In other words, why is offspring like parent, and when unlike in certain particulars, what is the cause and origin of the difference? We com- monly and properly enough take these two associated yet opposed facts as first principles. But it is equally proper and legitimate to enquire after the cause of them. M. Naudin, a good many years ago, took up the study of hybrid plants, and followed up for a series of generations, the course of life of certain self -fertile ones, notably of Datura. We gave at the time an abstract of his observations of the manner in which the characters of two closely related common species, D. Stramonium and D. Tatula^ were mixed, and in which the characters of the two began to separate in the close- bred progeny of the next generation, ending in a complete division of the amalgamated forms into those of the two con- stituent species after a few generations. ^ American Journal of Science and Arts, 3 ser., xi. 153. HEREDITY AND VARIABILITY IN PLANTS. 213 The " Comptes Renclus " of September 27 and October 4, 1875, contain an abstract of a paper communicated by M. Naudin to the French Academic des Sciences, of which the text was suggested by a hybrid between the wild Lactuca virosa and a variety of X. sativa, the common Lettuce. The hybrid was an accidental one : its seeds were fully fertile ; a great number of young plants were raised from them, of which twenty were preserved for full development and study. Like other hybrids the original showed no character which was not evidently derived from the two parents ; and, fer- tilized by its own pollen, the offspring all agreed in this respect, although they varied exceedingly among themselves in the division of the j^arental heritage, no two being quite alike. This exceeding vacillation between the two parental forms, but not overpassing the limits on either hand, — which Naudin finds to be the common characteristic of fertile hybrids, close-bred, — -.he names disordered variation (variation dSsordonnee). His explanation is that the hybrid is a j^iece of living mosaic, that two specific natures are at strife in it ; in the progeny each endeavors to reclaim its own, like seeks like ; whence in the course of a very few generations (as he first showed in Datura) a segregation takes j^jlace, part of the progeny reverting completely to one ancestral type, part to the other. What Naudin now insists upon is that out of all this disturbance comes nothing new ; that there is here no varia- tion beyond the line of inheritance ; and therefore from cross- ing no possible development of species. To this proposition we accede, so far as respects the direct consequence of crossing. To fill up the interval more or less between two forms or species with intermediate patterns may tend to the fusion or confusion of the two, but not to the orig- ination of new forms or species. Although Naudin's own experiments lead him to deny all tendency to variation over- passing these limits, we do not forget that his countryman, the late M. Vilmorin, — working in a different way and with another object, — arrived at a different conclusion. He suc- ceeded, as we understand, in originating floricultural novelties from species which refused to vary ^^er se, by making a cross, 214 REVIEWS. — not to infuse the character of the male parent, for he fer- tilized the progeny with the pollen of the female parent, and thus early bred out the other blood, but to induce variation, which, once initiated in the internal disorder consequent upon the crossing, was apt to proceed, or might be led on by selec- tion, to great lengths, according to Vilmorin. The variations in question, being mainly such as are sought in floriculture, may not have passed the line laid down by Naudin, or actually have introduced new features. But such plants would surely have no exemption from the ordinary liability to variation. If other plants vary, in the sense of producing something new, so may these. This brings us to another inference which Naudin draws. Having observed that his hybrids in their manifold variation exhibited nothing which was not derivable from their im- mediate ancestry, he directly (and in our opinion too confi- dently) concludes that all variation is atavism, — that when real variations are set up in ordinary species, this is not an origination but a reversion, a breaking out of some old ances- tral character, a particular and long deferred instance of this variation desordonnee, which w^ould thus appear to be the only kind of variation. This view has been presented before, but not, perhaps, so broadly. Adducing some theoretical considerations in its favor — to whicli we may revert — and some sound reasons against the view that variation is caused by external influences, he declares it " infinitely more prob- able that variation of species properly so called is due to ancestral influences rather than to accidental actions." We might think so if these two categories were exhaustive, and external conditions must be supposed to act immediately, as the cause rather than the occasion of variation. But the sup- position that " accidental actions," whatever they may be, and external influences of every sort do not produce but educe and conduct variation — which is our idea of what natural selec- tion means — avoids the force of Naudin's arguments. Moreover, Naudin's view, regarded as an hypothesis for explaining variation, leaves the problem just where it finds it. To explain the occurrence of present and actual variations, HEREDITY AND VARIABILITY IN PLANTS. 215 hypothetical ones like those of a former time are assumed ; the present diversity implies not only equal but the very same anterior diversity, and so on backwards. Or rather it demands a much greater diversity at the outset than now ; for these aberrant forms are the rare exception, and if due to atavism they imply the loss of the many and the incidental reappear- ance of the few. Else they would be the rule instead of the exception, and atavism would be simply heredity. This comes to the view which Mr. Agassiz strongly maintained, that really there are no varieties, — meaning, we understand, that all the forms are aboriginal, except the transient ones evidently due to circumstances. That some variation is atavism is clear enough. This is the natural explanation of the appearance of characters want- ing in the immediate parents but known in their ancestors or presumed ancestors. But the assumption of hypothetical ancestors to account for variation generally is quite another thing. Besides its inutility as an explanation, to which we have adverted, its improbability as an hypothesis is set in a strong light by Naudin's own forcible conception of the nature of heredity. What is heredity ? he asks. In other words, what keeps species so true, offspring like parent, through the long line of generations ? He illustrates hereditary force by comparing its action with that of physical force, in which the movement from one state of equilibrium to another is always that in which there is least resistance. From which it follows that when it has once begun to proceed in a certain course, its tendency to continue in that direction increases, because it facilitates its way as it overcomes obstacles. In other words, this line becomes fixed by habit ; vires acquirit eundo ; the stream deepens its bed by flowing ; and the more remote the commencement of a certain course, the more fixed its direc- tion, and the greater its power of overcoming opposition. The species is kept true in its course by the sum of the hered- ities which press each individual forward in its actual direc- tion. So that, as Naudin remarks, if we could calculate the energy with which millions of ancestors tend to impel the liv- ing representative of the line onward in the same direction, we 216 REVIEWS. should better appreliend the persistence of species, and feel the great improbability that the stream will ever escape from its ancient and well-worn bed, and strike into new courses. Now, in the first place, the more lively the conception we thus form of the invariability of species, through a happy metaphorical illustration of it, the more unlikely does it apj)ear that early characters, long lost in the flow, should reappear through atavism as varieties. To continue the simile, the more impetuous the stream, the less the possibility of its turning back upon itself, and resuming old characteristics. The eddies of atavism (the resumj^tion of dropped characters) are not likely to extend back very far ; and it seems gratuitous to have recourse to them in explanation of new forms. More- over, although the stream has made its bed and lies in it, not escaping from its own valley, it is flexible enough to obstacles, is ever changing its particular course as it flows, and may by its own action send off here and there a bayou (variety) or branch into a delta of channels (derivative species). Like Agassiz, Naudin conceives of species as originating with a large number of individuals of the same structure, and of which numerous reciprocal crosses have determined the direction of the line in which their posterity have evolved. But he maintains that these individuals, and all existing spe- cies, had a common origin in a " proto-organisra " ; and that the various lines of descent acquired fixity into species only as they acquired sexuality. If w^e rightl}- apprehend it, Nau- din's idea of the purport of sexual reproduction (as contrasted with that by buds) is to give fixity to species. Our idea is a different one, both as to the essential meaning of sexuality and as to its operation in respect to fixity. His conception may be tested by inquiring which are the more variable, or sportive, seedlings or p)lants propagated from buds. This we suppose can be answ^ered only one way. M. Naudin is a veteran and excellent investigator, and nothing which he w^rites is to be slighted. We have frankly set down our impressions upon a first perusal of his important communication ; but are ready to revise them, if need be, upon more deliberate consideration. FERTILIZATION IN THE VEGETABLE KINGDOM. 217 CROSS AND SELF-FERTILIZATION IN THE VEGE- TABLE KINGDOM. Mr. Darwin, in the title of Lis new work,^ refers only incidentally to adaptations for cross-fertilization, — a subject which has given origin to a copious literature since he opened it anew in his book on the Fertilization of Orchids, in 1862. A new edition of this latter book is on the eve of publication in England, and we believe that this author's scattered papers on cross-fertilization, as secured by various contrivances, are about to be collected, revised, and published in book form. In the volume now before us, Mr. Darwin deals with the effects of cross and self-fertilization, recounts at length the experiments he has devised and carried on, collects and criti- cises the results, glances at the means of fertilization, and the habits of insects in relation to it, and ends with some theoreti- cal considerations of inferences suggested by or deduced from the facts which have been brought to light. If writing for the popular press, we should be bound to say that the book is not light reading. Three fourths of its pages and of the chapters are devoted to the details of the experi- ments and the sifting and the various presentation of the results ; and the remainder, although abounding in curious facts and acute suggestions, is yet of a solid character. The bearings of various points upon what is called " Darwinism " are merely touched or suggested, here and there, in a manner more likely to engage the attention of the thoughtful scientific than of the general reader. That cross-fertilization is largely but not exclusively aimed at in the vegetable kingdom, is abundantly evident. As Mr. Darwin declares, "it is unmistakably plain that innumerable flowers are adapted for cross-fertilization, as that the teeth and talons of a carnivorous animal are adapted for catching prey, or that the plumes, hooks, and wings of a seed are adapted 1 The Effects of Cross aiid S el f -Fertilization in the Vegetable Kingdom. By Charles Darwin. London and New York, 187G. (American Journal of Science and Arts, 3 ser., xiii. 125.) 218 REVIEWS. for its dissemination." That the crossing is beneficial, and consequently the want of it injurious, is a teleological infer- ence from the prevalence of the arrangements which promote or secure it, — an inference the value of which increases with the number, the variety, and the effectiveness of the arrange- ments for which no other explanation is forthcoming. That the good consisted in a re-invigoration of progeny, or the evil of close-breeding in a deterioration of vigor, was the sugges- tion first made (so far as we know), or first made prominent, by Knight, from whom Darwin adopted it. However it be as to animals, there was until now no clear and direct evidence that cross-fertilization in the vegetable kingdom did re-invigo- rate. Indeed, the contrary might be inferred from the long and seemingly indefinite perpetuation of bud-propagating varieties, which have no fertilization at all. But the inference from this is not as cogent as would at first appear. For, although bud-propagation is, we think, to be considered as the extreme of close-breeding, yet in it the amount of material contributed by parent to offspring is usually vastly more than in sexual reproduction ; and, accordingly, the diminution to an injurious degree of any inherited quality or essence might be correspondingly remote. Yet, as sexual reproduction may be and often must be much closer in plants than it can be in most animals, the ill effects of self-fertilization, or the good of cross-fertilization, might the sooner be noticeable. Mr. Darwin arranged a course of experiments to test this question, prosecuted it as to some species for eleven years ; and the main object of this volume is to set forth the results. Ipomcea j)urinirea., the common Morning Glory of our gardens, was the leading subject. The flowers of this species self-fertilize, but must also be habitually cross-fertilized, as they are visited freely by bumble-bees and other insects. Ten flowers of a plant in a green-house were fertilized with their own pollen ; ten others were crossed with pollen from a dif- ferent plant. The seeds from both were gathered, allowed to germinate on damp sand, and as often as pairs germinated at the same time the two were planted on opposite sides of the same pot, the soil in which was well mixed, so as to be uniform FERTILIZATION 12^ THE VEGETABLE KINGDOM. 219 in composition. *' The plants on the two sides were always watered at the same time and as equally as possible, and even if this had not been done the water would have spread almost equally to both sides, as the pots were not large. The crossed and self -fertilized plants were separated by a superficial par- tition, which was always kept directed towards the chief source of light, so that the plants on both sides were equally illuminated." Five pairs were thus planted in two pots, and all the remaining seeds, whether or not in a state of germina- tion, were planted on the ojjposite sides of a third pot, so that the plants were crowded and exposed to a very severe compe- tition. Rods of equal diameter were given to all the plants to twine up, and as soon as one of each pair had reached the summit, both were measured. But a single rod was furnished to each side of the crowded pot, and only the tallest plant on each side was measured. This was followed up for ten gen- erations ; the close-fertilization being always self-fertilization, i. e., by pollen to stigma of the same flower; the crossing, between individuals in successive generations of this same stock, except in special instances, when aa extraneous stock was used as one parent, — to eminent advantage, as will be seen. The difference in vigor between the cross-bred and the close-bred progeny, as measured by early growth, was well marked throughout. In the mean of the ten generations it was as 100 to 77. In the tenth generation it was 100 to 54, that is, five cross-bred plants grew to the average height of 93.7 inches while the close-bred were reaching: the averajre of 50.4 inches. This was a notably greater difference than in any previous generation. But this was probably accidental or anomalous, for it was not led up to by successive steps. Indeed, the difference in the first generation was a trifle greater than the average of all ten, being as 100 to 76. The second generation was as 100 to 79 ; the third as 100 to 68 ; the fourth as 100 to 86 ; the fifth as 100 to 75 ; the sixth as 100 to 72 ; the seventh as 100 to 81 ; the eighth as 100 to 85 ; the ninth as 100 to 79 ; the tenth as already stated, 100 to 54. The general result is made striking in the following illustration. 220 REVIEWS. " If all the men in the country were on an average six feet high, and there were some families which had been long and closely interbred, these would be almost dwarfs, their average height during ten generations being only four feet eight and one-quarter inches." (p. 53.) It is remarkable that the difference between the close-bred and the cross-bred individuals should have been as great as it was in the first generation ; and, this being the case, it might have been expected that the difference would have gone on increasing in the succeeding generations. If self- fertilization is injurious, the ill effects would be expected to be cumulative. " But," instead of this, " the difference be- tween the two sets of plants in the seventh, eighth, and ninth o-enerations taken too^ether is less than in the first and second generations together." Upon this Mr. Darwin remarks : " When, however, we remember that the self-fertilized and crossed j^lants are all descended from the same mother plant, that many of the crossed plants in each generation were re- lated, often closely related, and that all were exposed to the same conditions, which, as we shall hereafter find, is a very important circumstance, it is not at all surprising that the difference between them should have somewhat decreased in the later generations." (p. 56.^ Further light was thrown upon these points by two kinds of subsidiary experiments. In one case, the cross was made between two flowers of the same plant of Ipomsea, while other flov/ers were self-fertilized as before. On raising seed- lings from the two lots, it was found that such crossing gave no superiority ; indeed, the offspring of the self-fertilized flowers appeared to be rather more vigorous than the close- crossed. And other experiments led to the same conclusion, namely, that there was no particular benefit from cross-fer- tilization on the same plant. In the other case, the cross was made not only between the flowers of distinct plants, but between those from different sources, and which had pre- sumably grown under somewhat different conditions. For instance, several flowers of the ninth generation of crossed plants of Ipomaea were crossed with pollen taken from the FERTILIZATION IN THE VEGETABLE KINGDOM. 221 same variety but from a distant garden. The resulting seed- lings showed the benefit of the fresh stock remarkably, being as much superior in vigor to those of the tenth intercrossed generation as the latter were to the self-fertilized plants of a corresponding generation. In height they were as 100 to 78, over the ordinary intercrossed; and in fertility, as 100 to 51. Indeed, Mr. Darwin's main conclusion from all his observations is, '' that the mere act of crossing by itself does no good. The good depends on the individuals which are crossed differing slightly in constitution, owing to their pro- genitors having been subjected during several generations to slightly different conditions, or to what we call in our igno- rance spontaneous variation." The greater constitutional vigor of the crossed plants of Ipomaea was manifested in other ways than their rate or amount of growth : they better endured exposure to a low temperature or sudden changes of temperature ; they blos- somed earlier ; and they were more fertile. The difference in fertility varied greatly in degree (the extremes in dif- ferent experiments and in different generations being 100 to 99 and 100 to 26), but was always sustained. Also, "the impaired fertility of the self -fertilized plants was shown in another way, namely, by their anthers being smaller than those in the flowers on the crossed plants. This was first observed in the seventh generation, but may have occurred earlier. . . . The quantity of pollen contained in one of the self-fertilized was, as far as could be judged by the eye, about half of that contained in one from a crossed plant. The im- paired fertility of the self-fertilized plants of the eighth gen- eration was also shown in another manner, which may often be observed in hybrids — namely, by the first-formed flowers being sterile." Similar experiments were made, but not carried to the same extent, upon fifty-seven other species of plants, belong- ing to fifty-two genera, and to thirty great natural families, the species being natives of all parts of the world. The re- sults — the details and discussion of which occupy the bulk of this volume — vary greatly, some plants making a better and 222 REVIEWS. others a less good showing for the advantage of cross-fertili- zation, and this advantage manifesting itself in different ways, some in vigor or amount of growth, some in hardiness, most in fertility ; but with twelve cases in which the crossed plants showed no marked advantage over the self-fertilized. There were, however, fifty-seven cases in which the crossed exceeded the self-fertilized by at least five per cent., gener- ally by much more. Increase of vigor, as evinced in growth, appears generally to be accompanied by increased fertility ; but sometimes the good of crossing was manifested only in productiveness, i. e., in a larger amount of seed. This proved to be the case in Eschseholtzia, in which — strange to say — self -fertilized plants of several generations were superior in size and weight to intercrossed plants, even when the crossing was between flowers derived on one side from American, on the other from English seed, from which, upon Mr. Darwin's view, the maximum benefit should be gained. This instance, however, stands alone. Yet it is approached by several others, in a manner which might have negatived the general conclusions of the research, if they had been hastily gathered from a small number of trials. For example, in the sixth self-fertilized generation of Ipo- mcea purjnirea^ one of these plants took the lead of its com- petitor, kept it almost to the end, and was ultimately over- topped only by half an inch on a total height of several feet. To ascertain whether this exceptionally vigorous plant would transmit its power to its seedlings, several of its flowers were fertilized with their own pollen, and the seedlings thus raised were put into competition with ordinary self-fertilized and with intercrossed j)lants of the corresponding generation. The six children of Hero (the name by which this individual was designated) beat the ordinary self-fertilized competitors at the rate of 100 to 84, and the intercrossed competitors at the rate of 100 to 95 ; and in the next generation the self- fertilized grandchildren beat those from a cross between two of the children at the rate of 100 to 94. In the next genera- tions the seedlings were raised in winter in a hot-house, became FERTILIZATION IN THE VEGETABLE KINGDOM. 223 unhealthy, and the experiment terminated without marked result. Moreover, the remarkable vigor of growth in Hero and its progeny was attended by somewhat increased fertil- ity. Here, then, an idiosyncrasy arose from some utterly un- known cause, — a spontaneous variation of constitution, which was transmitted to posterity, and which gave all the benefit of cross-fertilization, and somewhat more, both as to vigor and fertility. A similar idiosyncrasy made its appearance in the third generation of seedlings of 3Iimulus luteus. Discordant or anomalous facts like these seem confusing, even though too few to affect seriously the grand result of the numerous experiments ; but upon Darwinian principles, in which adaptations are ultimate results, they are to be ex- pected, as a consequence of the general and apparently vague proclivity to vary. In Foxglove, — the flowers of which are naturally self -sterile or nearly so, and in which crossing gave a marked advantage over self-fertilizing, both as to growth and productiveness, — a decided, thougli small advantage appeared to come from the crossing of flowers on the same plant. In Origanum vulgare., crosses w^re made between different plants of a large clump, long cultivated in a kitchen-garden, which had evidently spread from a single root by stolons, and which had become in a good degree sterile, as is usual under such conditions. The crossing caused rather more seed to form ; but the seedlings from the crossed did not surpass in grow^th those of the self-fertilized ; " a cross of this kind did no more good than crossing two flowers on the same plant of Ipomaea or Mimulus. Turned into the open ground, and both self and cross-fertilized the following summer, and equal pairs of the resulting seeds planted on opposite sides of two very large pots, the crossed plants from seed showed a clear supe- riority over their self-fertilized brethren, at the rate of 100 to 86. But this excess of height by no means gives a fair idea of the vast superiority in vigor of the crossed over the self-fertilized plants. The crossed flowered first and j^roduced thirty flower-stems, while the self-fertilized produced only fifteen, or half the number. The pots were then bedded out, 224 REVIEWS. and the roots probably came out of the holes at the bottom, and thus aided their growth. Early in the following summer, the superiority of the crossed plants, owing to their increase by stolons, over the self -fertilized plants, was truly won- derful. . . . Both the crossed and the self-fertilized plants being left freely exposed to the visits of bees, manifestly produced much more seed than their grandparents, — the plants of original clumps still growing close by in the same garden, and equally left to the action of bees." These few cases must here suffice, and they give a fair gen- eral idea of the main results reached, — somew^hat qualified, however, by certain instances in which little or no benefit was observed. Let it be remarked that while most of the cases show decided and unequivocal good from the crossing, none of them unequivocally tell to the contrary, as the ad- vantage appears sometimes in one direction, sometimes in another. "Thus, the crossed and self - fertilized plants of Ipomsea, Papaver, Reseda odorata, and Limnanthes were almost equally fertile, yet the former exceeded considerably in height the self-fertilized plants. On the other hand, the crossed and seK-fertilized plants of Mimulus and Primula differed in an extreme degree in fertility, but by no means to a corresponding degree in height or vigor." We must wholly omit — among many other things — the interesting account of self-sterile plants, meaning here not those in which the pollen does not reach the stigma unaided, but those in which it is impotent, or nearly so, when applied, although efficient upon the stigma of another individual. Verbascum, Passiflora, Corydalis, and many Orchids afford instances of this sort. In these, the advantage of cross-fer- tilization arises to a necessity. A noteworthy fact respect- ing them (of which Mr. Darwin makes much) is, that such self-sterility, or the reverse, is influenced by slight changes in the conditions, such as difference in temperature, grafting on another stock, and the like. In South Brazil, Fritz Miiller found that for six generations all his plants of Eschsclioltzia Californica were completely sterile, unless supplied with pollen from a distinct plant, when they were completely fer- FERTILIZATION IN THE VEGETABLE KINGDOM. 225 tile. This was not the case in English plants, which, when covered by a net, set a considerable number of capsules, the seeds o£ which, by weight, were as 71 to 100 of those on plants intercrossed by bees. These Brazilian seeds, sent to England, yielded plants with moderately self-fertile flowers, and this limited self-fertility was increased in two generations of English growth. Conversely, seeds from English plants grown in Brazil were more self-fertile than those reared in Brazil for several generations ; yet " one which did not flower the first year, and was thus exposed for two seasons to the climate of Brazil, proved quite self-sterile, like a Brazilian plant, showing how quickly the climate had acted on its sexual constitution." Having observed that certain individuals of Mignonette were self-sterile, Mr. Darwin secured several such plants under separate nets, and by intercrossing these for a few generations, obtained plants which inherited this peculi- arity, so that " without doubt a self -sterile race of Mignonette could easily have been established." Nine of the twelve cha23ters are devoted strictly to the effects of cross and self-fertilization. The tenth considers the " means of fertilization." Cross-fertilization is favored or ensured by : 1, the separation of the sexes ; 2, the maturity of the male and female sexual elements at different 23eriods; 3, dimor- phism, or even trimorphism ; 4, various mechanical contriv- ances ; 5, the more or less complete inefficiency of the flower's own pollen on its stigma, and the jDrepotency of pollen from any other individual over that from the same plant. We understand that Mr. Darwin is just now occupied in revising and extending his various papers upon these topics, with a view to their publication in a volume. Here he gives a list of plants which, when insects are excluded, are either quite ster- ile or produce less than half the number of seeds yielded by unprotected plants. This is followed by a list of plants which, when protected from insects, are either quite fertile or yield more than half the number of seeds produced by unprotected plants. " Each of these lists contains by a mere accident the same number of genera, namely, forty-nine. The genera in the first 226 REVIEWS. list include sixty-five species, and those in the second sixty species ; the Orchldece in both being excluded. If the genera in this latter order, as well as in the Asclej^iadece and A2^ocy- nacece^ had been included, the number of species which are sterile if insects are excluded would have been greatly in- creased ; but the lists are confined to species which were actu- ally experimented upon. The results can be considered as only approximately accurate, for fertility is so variable a character, that each species ought to have been tried many times. The above number of species, namely, 125, is as nothing to the hosts of living plants ; but the mere fact of more than half of them being sterile within the specified degree, when insects are excluded, is a striking one ; for whenever pollen has to be carried from the anthers to the stigma in order to insure full fertility, there is at least a good chance of cross-fertilization. I do not, how^ever, believe that if all known plants were tried in the same manner, half would be found to be sterile within the specified limits ; for many flowers were selected for experiment which presented some remarkable structure ; and such flowers often require insect- aid."— (p. 370.) It is worth noticing that Tnfolium repens and T. 2'>ro,tense (the common White and Eed Clovers) have a place in the first list ; T. arvense and T. iwoQUiiibens in the second. Darwin refers to Mr. Miner's statement that "in the United States hive-bees never suck the Red Clover," and says it is the same in England, except from the outside through holes bitten by humble-bees ; yet that H. Miiller has seen them visiting this plant in Germany for the sake both of pollen and nectar, which latter they obtained by breaking apart the petals. Darwin has not qualified his statement, long ago made, of the complete sterility of Red Clover protected from insects ; but Mr. Meehan asserts that protected plants are fertile in this country, without, however, giving details or the rate of fertil- ity. In T. arvense^ " the excessively small flowers are inces- santly visited by hive and humble-bees ; when insects were excluded the flower-heads seem to produce as many and as fine seeds as the exposed heads." FERTILIZATION IN THE VEGETABLE KINGDOM. 227 As to cross-fertilization, "the most important of all the means by which the pollen is carried from the anthers to the stigma of the same flower, or from flower to flower, are in- sects, belonging to the orders of Plymenoptera, Lepidoptera, and Diptera ; and in some parts of the world, birds." In a note the author cites all the cases known to him of birds fertil- izing flowers. These are chiefly humming-birds. " In North America they are said to frequent the flowers of Impatlens " (for which Gould, " Trochilidae," is referred to as authority, and a reference is given to the " Gardener's Chronicle," which we find relates to something else in South America) ; and this is all concerning the United States. Can it be that there are no references in print to the most familiar fact that our hum- ming-bird is very fond of sucking the blossoms of Trumpet Creeper QTecoma radicans) and of Honeysuckles? Both these are, in size and arrangement of parts, well adapted to be thus cross-fertilized. Flowers are rendered conspicuous to birds and still more to insects, by bright colors. And as " almost every fruit which is devoured by birds presents a strong contrast in color with the green foliage, in order that it may be seen and its seeds disseminated," so the proportionally large size and the bright colors of the corolla, or, in some cases, the equally bright hues of the adjoining parts of the flower, or of the inflorescence, are correlated to visiting insects, — have come to pass, as Dar- win would say, in consequence of the visits of insects, through the advantages in vigor and productiveness gained by cross- fertilization. He is ready to adopt even the idea of Conrad Sprengel, which seemed to be so fanciful, that marks and streaks on the corolla serve as guides to the nectary ; for, although insects are well able to discover the nectar without the aid of guiding marks, yet they are of service by facilitating the search and enabling insects to suck a greater number of blossoms within a given time, which is tantamount to greater opportunity for cross-fertilization. That odors attract insects is certain, and many flowers are both conspicuous and odoriferous, while others make up in fragrance what they lack in show. " Nageli affixed artificial 228 REVIEWS. flowers to branches, scenting some with essential oils, and leavins: others unscented ; and the insects were attracted to the former in an unmistakable manner." " Of all colors white is the prevailing one ; and of white flowers a considerably larger proportion smell sweetly than of any other color, namely, 14.6 per cent. ; of red, only 8.2 per cent, are odoriferous. The fact of a larger proportion of white flowers smelling sweetly may dej^end in part on those which are fertilized by moths requiring the double aid of conspicuous- ness in the dusk and of odor. So great is the economy of nature, that most flowers which are fertilized by crepuscular or nocturnal insects emit their odor chiefly or exclusively in the evening. Some flowers, however, which are highly odor- iferous depend solely on this quality for their fertilization, such as the night flowering-stock (Hesperis) and some species of Daphne ; and these present the rare case of flowers which are fertilized by insects being obscurely colored." " The shape of the nectary and of the adjoining parts are likewise related to the particular kinds of insects which habit- ually visit the flowxrs : this has been well shown by H. Miiller, by his comparison of lowland species, which are chiefly visited by bees, with alpine species belonging to the same genera, v/hich are visited by butterflies." "Pollen contains much nitrogen and phosphorus, — the two most precious of all the elements for the growth of plants, — but in the case of most open flowers, a large quantity of pollen is consumed by pollen-devouring insects, and a large quantity is destroyed during long-continued rain. With many plants this latter evil is guarded against, as far as possible, by the anthers opening only during dry weather, by the position and form of some or all of the petals, by the presence of hairs, etc. ; also, as Kerner has shown in his interesting essay, by the movements of the petals or of the whole flower during cold and wet w^eather. In order to compensate the loss of pollen in so many ways, the anthers produce a far larger amount than is necessary for the fertilization of the same flower. I know this from my own experiments on Ipomsea, given in the Introduction ; and it is still more plainly shown FERTILIZATION IN THE VEGETABLE KINGDOM. 229 by the astonishingly small quantity produced by cleistogene flowers, which lose none of their j^ollen, in comj)arison with that produced by the open flowers borne by the same plants ; and yet this small quantity suffices for the fertilization of all their numerous seeds. Mr. Hussall took pains in estimating the number of pollen-grains produced by a flower of the Dan- delion, and found the number to be 243,600, and in a Peony 3,654,000 grains. The editor of the ' Botanical Register ' counted the ovules in the flowers of Wistaria sinensis^ and carefully estimated the number of pollen-grains, and he found that for each ovule there were 7,000 grains." — (pp. 376, 377.) These are probably fair averages of the numerical ratio of pollen to ovules in flowers which are adapted to be fertil- ized by insect agency. Their meaning in the " economy of nature " is seen by a comparison on the one hand with ane- mo'pliiloiis^ i. e. wind-fertilized, flowers, in most of which there is a vastly greater disproportion between the numbers, — com- pensating for inevitable waste, — and on the other hand with cleistogenous flowers, namely, those small and less developed blossoms which some plants produce in addition to the ordi- nary sort, and which fertilize as it were in the bud, necessarily by their own pollen. Here is no waste, and accordingly the anthers are very small, and the pollen-grains are not many times more than the ovules : also such flowers are never brightly colored, never odoriferous, and they never secrete nectar. The only advantages of this close-fertilization which we can think of are sureness and strict likeness ; both of which are quite as well secured by budding-reproduction. Now, as cleistogene flowers are borne, we believe, chiefly and perhaps only, by species whose normal blossoms are adapted for insect- fertilization, they must be regarded as a subsidiary arrange- ment, a safeguard against failure of proper insect-visitation. As the volume before us amply shows, this failure is in gen- eral provided for by a more or less wide margin of self-fertil- ization in the very flowers which are adapted for crossing. In Impatiens, Yiola, and the like, it is provided for by separate flowers, the sj^ecial adaptations of which are unmistakable. H. Miiller appears to have shown " that large and conspic- 230 REVIEWS, uous flowers are visited much more frequently and by many more kinds of insects than are small inconspicuous flowers. He further remarks that the flowers which are rarely visited must be capable of self-fertilization, otherwise they would quicldy become extinct." Mr. Darwin's list seems to show that, as a rule, they are so ; yet many very small flowers, like those of Trifolium arvense, and small and dingy ones, like those of Asparagus, are freely visited by bees ; and, con- versely, many large and conspicuous flowers which are fre- quented by insects are none the less self-f ertilizable. Through- out we find that such things do not conform to arbitrary or fixed rules ; and this favors the idea that the differences have been acquired. Mr. Darwin conjectures that the self -fertiliz- ing capabilities of many small and inconspicuous flowers may be comparatively recent acquisitions, on the ground that, if they were not occasionally intercrossed, and did not profit by the process, all their flowers would have become cleisto- genous, " as they would thus have been largely benefited by having to produce only a small quantity of safely protected poUen." Mr. Darwin's experiments tending to prove that cross-fer- tilization between flowers on the same plant is of little or no use, he is naturally led to consider the means which favor or ensure their fertilization with pollen from a distinct plant. This must needs take place with dioecious plants, and is likely to occur with the monoecious, and is in some cases secured (as in Walnut and Hazelnut) by some trees being proterandrous and others proterogynous, so that they will reciprocally fer- tilize each other. In ordinary hermaphrodite species the ex- pansion of only a few blossoms at a time greatly favors the intercrossing of distinct individuals, although in the case of small flowers it is attended with the disadvantage of render- ing the plants less conspicuous to insects. Our common Sundews furnish a good illustration of this. They abound wherever they occur, and are for a long while in blossom, but each plant or spike opens but one flower at a time. The fact of bees visiting the flowers of the same species as long as they can, instead of promiscuously feeding from the various FERTILIZATION IN THE VEGETABLE KINGDOM. 231 blossoms nearest within reach, greatly favors such intercross- ing. So does the remarkable number of flowers which bees are able to visit in a short time (of which mention wdll be made), and the fact that they are unable to perceive with- out entering a flower whether other bees have exhausted the nectar. Then dichogamy (the maturation of one sex in a I hermaphrodite flower earlier than the other) is so prevalent that it may almost be regarded as the rule ; and this ensures such crossing between few-flowered plants, and greatly favors it in the case of spikes, racemes, and the like. For, proteran- dry being the commonest arrangement, so that the younger flowers act as male, and the older as female, the bees habitu- ally alighting at the bottom and proceeding upward, they carry the pollen from the upper and younger flowers to stig- mas of the lower and older flowers of the next spike, and so on. Heterogonism, which is less common, operates precisely like complete dioecious separation of the sexes in this respect, and with the advantage that all the individuals are seed-bear- ing. Most of the special arrangements peculiar to certain families, such as Orchids, — or to plants, such as Posoqueria, with its wondrous mechanism for quickly stopping out access to the stigma when the pollen is violently discharged upon some insect, but opening the orifice the next day, — are of a kind to favor the crossing of distinct plants. Prepotency of other pollen, which may accompany the other arrangements or exist independently, acts largely and powerfully toward the same end. Our author investigates this at some length : we cite for illustration a single but strong case. The stig- mas of a long-styled Cowslip were supplied with pollen from the same plant, and again after twenty-four hours, with pol- len of a short-styled, dark-red Polyanthus, a variety of the same species : from the resulting seeds tw^enty seedlings were raised, and all of them bore reddish flowers ; so that the effect of the plant's own pollen, though placed on the stigmas twenty-four hours previously, was destroyed by that of the red variety. The same thing is shown by the impossibility in many cases of raising two varieties of the same species pure if they grow near each other. " No one who has had any 232 REVIEWS. experience would expect to obtain pure cabbage-seed, for in- stance, if a plant of another variety grew within 200 or 300 yards." And a veteran cultivator once had his whole stock of seeds seriously bastardized by some plants of purple Kale which flowered in a cottager's garden half a mile away. Mr. Gordon records a case of the crossing between Primroses and Cowslips through pollen carried by bees over more than two kilometers, or an English mile and a quarter. We must copy the close of this section — long though it be — because of its capital illustration of the topic in hand, and for the teleological lesson which it teaches. " The case of a great tree covered with innumerable her- maphrodite flowers, seems at first sight strongly opposed to the belief in the frequency of intercrosses between distinct individuals. The flowers which grow on the opposite sides of such a tree will have been exposed to somewhat different conditions, and a cross between them may perhaps be in some degree beneficial; but it is not probable that it would be nearly so beneficial as a cross between flowers on distinct trees, as we may infer from the inefficiency of the pollen taken from plants which have been propagated from the same stock thouG^h crrowins' on different roots. The number of bees which frequent certain kinds of trees when in full flower is very great, and they may be seen flying from tree to tree more frequently than might have been expected. Nevertheless, if we consider how numerous are the flowers, for instance, on a Horse-Chestnut or Lime-tree, an incomparably larger num- ber of flowers must be fertilized by pollen brought from other flowers on the same tree, than from flowers on a distinct tree. But we should bear in mind that with the Horse-Chest- nut, for instance, only one or two of the several flowers on the same peduncle produce a seed ; and that this seed is the pro- duct of only one out of the several ovules within the same ovarium. Now we know from the experiments of Herbert and others that if one flower is fertilized with pollen which is more efficient than that applied to the other flowers on the same peduncle, the latter often drop off ; and it is probable that this would occur with many of the self-fertilized flowers FERTILIZATION IN THE VEGETABLE KINGDOM. 233 on a large tree, if other and adjoining flowers were cross- fertilized. Of the flowers annually produced by a great tree, it is almost certain that a large number would be self- fertilized ; and if we assume that the tree produced only 500 flowers, and that this number of seeds were requisite to keep up the stock, so that at least one seedling should here- after struggle to maturity, then a large proportion of the seedlings would necessarily be derived from self-fertilized seeds. But if the tree annually produced 50,000 flowers, of which the self-fertilized dropped off without yielding seeds, then the cross-fertilized flowers might yield seeds in sufficient number to keep up the stock, and most of the seedlings would be vigorous from being the product of a cross between dis- tinct individuals. In this manner the production of a vast number of flowers, besides serving to entice numerous insects and to compensate for the accidental destruction of many flowers by spring-frosts or otherwise, v/ould be a very great advantage to the species ; and when we behold our orchard- trees covered with a white sheet of bloom in the spring, we should not falsely accuse Nature of wasteful expendi- ture, though comparatively little fruit is produced in the autumn." The Horse-Chestnut is not altogether a well-chosen ex- ample, for in it, as in our Buckeyes, a very large projwrtion of the flowers in the thyrsus are usually male, with barely a vestige of pistil. These serve, however, to increase the show, in the manner here illustrated, as well as to furnish abun- dance of pollen. The section on anemophilous (wind-fertilized) plants, — their interest as survivals of the earlier phtenogamic vegetatiou, the speculation as to how, when flying insects came to prevail, an anemophilous plant may have been rendered entomophi- lous, — how pollen, being a most nutritious substance, would soon have been discovered and devoured by insects, and by adhering to their bodies be carried from anthers to stigma and from one flower to another, — how a waste secretion, such as honey-dew or glandular exudations, may have been devel- oped into nectar and utilized as a lure, — the interesting illus- 234 REVIEWS. trations of the vast amount of pollen produced by anemophi- lous plants, and the great distances to which their light pollen is often carried by the wind, — all these inviting topics we must now pass by. In passing we note the remark that " the excretion of a sweet liquid by glands seated outside of a flower is rarely utilized as a means of cross-fertilization by the aid of in- sects ; " and the sole exception alluded to is that of the bracts of Marcgraviacece, But a parallel case is afforded by many species of Euphorbia, and notably in a striking species culti- vated in conservatories, under the name of Poinsettia. Here the attraction to the eye is supplied by the intense red colora- tion of ordinary leaves placed next to the inflorescence, and that to the palate or tongue (if either term be allowed), by a large cup-shaped gland on the side of the involucre, wdiich contains or surrounds the naked and greatly simplified flowers of both sexes. That anemophilous plants are prevailingly declinous (either monoecious or dioecious) is speculatively connected with their antiquity ; that they are very largely trees or shrubs is because " the long life of a tree or bush permits of the separa- tion of the sexes with much less risk of evil from impregna- tion occasionally failing, and seeds not being produced, than in the case of short-lived plants. Hence it is probably, as Lecoq has remarked, that annual plants are rarely dioecious." The number of anemophilous species is comparatively small, but that of individuals of the species strikingly large, so that they form of themselves, in cold and temperate regions, where plant-fertilizing insects are fewer, either vast forests, as of Coniferce.^ Birches, Beeches, etc., or meadows and glades, as of Grasses, Sedges, and Kushes. Being thus either neces- sarily or prevailingly cross-fertilizable and gregarious, it is not wonderful that they should hold their own unchanged in various parts of the world. Still their advantage is gained at the expense of the production of an enormous superfluity of pollen, a costly product ; and, when dioecious, half the in- dividuals produce no seed. Hermaphroditism with dicho- gamy, or some equivalent, and transportation by an appeal FERTILIZATION IN THE VEGETABLE KINGDOM. 235 to the senses and appetites of insects, secures all the advan- tages with least expenditure. The earliest fertilization in plants took place by the locomotion of the fertilizing or even of the fertilized material, in manner of most of the Algae. : mainly losing this as vegetation became terrestrial, the trans- portation was committed to the winds, and finally in the higher plants more economically consigned to insects. The eleventh chapter, on the habits of insects in relation to the fertilization of flowers, is one of the interesting and readable although one of the shortest. It appears that the prince of naturalists, Aristotle, had observed more than two thousand years ago that the hive-bees visited the flowers of the same species as long as possible before going to a dif- ferent species. This holds true of all kinds of bees and cer- tain other insects, generally, but not absolutely ; although, as Lubbock has recently proved, bees are much guided by color, yet they hold to the practice just mentioned in spite of difference in this respect, being botanists enough to know that color is not a good specific character. Mr. Darwin has repeatedly seen humble-bees flying straight from a red Frax- inella to a white variety, from one Larkspur to a different- colored variety ; and the same as to Primroses and Pansies. But two species of Poppy were by some bees treated as one ; and H. Miiller traced hive-bees from blue Hyacinths to blue Violets. On the other hand, Darwin's bees fly straight from clump to clump of yellow Oenothera without turning an inch in their course to Eschscholtzias with yellow flowers which abound on either side. This constancy to species, however, is manifested only when their flowers abound ; a fact which may have led Mr. Darwin to his explanation of the reason of it. " The cause probably lies in insects being thus enabled to work quicker ; they have just learned how to stand in the best position on the flower, and how far and in what direc- tion to insert their proboscides. They act on the same prin- ciple as does an artificer who has to make a half a dozen en- gines, and who saves time by making consecutively each wheel and part for all of them. Insects, or at least bees, seem much influenced by habit in all their manifold operations ; 236 REVIEWS. and we sliall presently see that this holds good in their felo- nious practice of biting holes through the corolla." — (p. 420.) As to this latter practice — " The motive which impels bees to gnaw holes through the corolla seems to be the saving of time, for they lose much time in climbing into and out of large flowers, and in forcing their heads into closed ones. They were able to visit nearly twice as many flowers, as far as I could judge, of a Stachys and Pentstemon by alighting on the upper surface of the corolla and sucking through the cut holes, than by entering in the proper way. Nevertheless each bee before it has had much practice, must lose some time in making each new perforation, especially when the perforation must be made through both calyx and corolla. This action therefore implies foresight, of which faculty we have abundant evidence in their building operations ; and may we not further believe that some trace of their social instinct, that is, of working for the good of other members of the community, may here likewise j^lay a part ? Many years ago I was struck with the fact that humble-bees as a general rule perforate flowers only when these grow in large numbers near together,'* etc., etc. (p. 433.) It appears that the cutting of these holes is done only by humble-bees, never by hive-bees. Yet the latter are quick to take advantage of them. " In the early part of the summer of 1857 I was led to observe during some weeks several rows of the scarlet Kidney- bean (^Phaseolus multiflorus)^ whilst attending the fertiliza- tion of this plant, and daily saw humble and hive-bees suck- ing at the mouths of the flowers. But one day I found sev- eral humble-bees employed in cutting holes in flower after flower ; and on the next day every single hive-bee, without exception, instead of alighting on the left wdng petal and suck- ing the flower in the proper manner, flew straight without the least hesitation to the calyx, and sucked through the holes which had been made only the day before by the humble-bees, and they continued this habit for many following days. Mr. Belt has communicated to me (July 28, 1874) a similar case, FERTILIZATION IN THE VEGETABLE KINGDOM. 237 with the sole difference that less than half of the flowers had been perforated by the humble-bees ; nevertheless all the hive-bees gave up sucking at the mouths of the flowers and visited exclusively the bitten ones. Now how did the hive- bees find out so quicldy that holes had been made ? Instinct seems to be out of the question as the plant is an exotic. The holes cannot be seen by bees whilst standing on the wing-petals, where they had always previously alighted. From the ease with which bees were deceived when the petals of Lobelia Erinus were cut off, it was clear that in this case they were not guided to the nectar by its smell ; and it may be doubted whether they were attracted to the holes in the flowers of the Phaseolus by the odor emitted from them. Did they perceive the holes by the sense of touch in their proboscides, whilst sucking the flowers in the proper manner, and then reason that it would save them time to alight on the outside of the flower and use the holes ? This seems almost too abstruse an act of reason for bees ; and it is more prob- able that they saw the humble-bees at work, and understand- ing what they were about, imitated them and took advantage of the shorter path to the nectar. Even with animals liigh in the scale, such as monkeys, we should be surprised at hear- ing that all the individuals of one species within the space of twenty-four hours understood an act performed by a dis- tinct species and profited by it." — (pp. 430, 431.) But we must cut short our citations and remarks ; passing by one of the most important points, relative to the amount of fertilizing work done by insects, namely, the evidence of the extraordinary industry of bees and the number of flowers visited within a short time ; which, as well as the distance to which pollen is sometimes transported, is far greater than one would have supposed. But the volume is reprinting by the Appletons, and will soon be within the reach of all, — along with a new edition of the Orchid-fertilization book, the proper supplement to the present work, relating as it does to the class of plants in which the adaptation for fertilization by insects is carried to the highest degree of sjiecialization and perfection. 238 REVIEWS. THE HYBRIDIZATION OF LILIES. Mr. Parkman, under the above title, gives a summary ^ of liis exi3eriments, during ten or twelve years, in crossing Lilies. One of the earlier results, and that which the horticulturists count as the eminent one, was the production of that mag- nificent hybrid between X. auratum and L. speciosum^ with flower resembling the former in fragrance and form and the most brilliant varieties of the latter in color, which was brought out in England under the name of Lilium Parh- manni. The interesting physiological point which Mr. Park- man here records is, that this striking novelty was wholly unique ; that all the other seeds of the same parentage which germinated, over fifty in number, gave rise to plants which in the blossom showed no trace of the male parent, L. auratum^ but were exactly like the female parent, L. speciosum. That these plants were truly hybrids, notwithstanding, is well made out : 1, by the precautions taken against any possible access of own pollen ; 2, by the scantiness of seed, most of which was abortive ; 3. " such good seed as there was differed in appear- ance from the seed of the same Lily fertilized by the pollen of its own species," which is smooth, while this was rough and wrinkled ; and 4, the stems were mottled after the manner of the male parent. It would naturally be thought that this slight but evident impression of the character of the male j^arent might be deepened by iteration. That was tried next year, when the flowers of several of these plants were fertilized with the pollen of L. auratum precisely as their female parent had been fer- tilized. The result was an extremely scanty crop of seed, " but there w^as enough to produce 8 or 10 young bulbs. Of these, when they bloomed, one bore a flower combining the features of both parents, but though large, it was far inferior to L. Parhmanni in form and color ; the remaining flowers 1 The Hibridization of Lilies. By Francis Parkman, in Bull. Bussey In- stitution, ii., 1877. (American Journal of Science and Arts, 3 ser., xv. 144.) THE HYBRIDIZATION OF LILIES, 239 were not distinguishable from those of the pure L. speciosum.'* The article records the results of various similar attempts to hybridize other Lilies. For instance, our L. sup)erhum was poUenized with eight different old-world species. The result was, that capsules, apparently perfect, were abundantly pro- duced ; some of them contained nothing but chaff, others had a few imperfect seeds, still others gave a fair supply of good seed. From this seed several hundred young bulbs were pro- duced. " But when these came into bloom, not a single flower of them all was in the least distinguishable from the pure L. siqjerbum.'' Moreover, in this case (different from the other) " not one of the eight different male parents had im- posed his features on his hybrid offspring. Not only in their flowers, but in their leaves, stems, and bulbs, the young plants showed no variation from their maternal parent." The ex- periment proceeded one generation farther. " In the following year I set some of them apart from the rest, and applied to them, as to their mother before them, the pollen of several species of Lilies. This time the seeds were extremely scanty. A few, however, were produced ; but the plants and flowers that resulted from them were, to all appearance, X. siipcrhum pure and simple." In trials of other species results intermediate between these two cases were obtained. For instance, the pure white of the perianth of L. longiflorum came out unstained in the progeny raised by a crossing with L. S2:>eciosi(m, and the herbage was equally unaffected ; but in that or the next generation " dis- tinct evidence could be seen of the action of alien pollen " in the changed color of many of the anthers, and in the abortion of others. They also showed differences of habit among them- selves, some being very tall and vigorous, and others compact and bushy, with a tendency to bloom in clusters ; but these may have been mere seedling variations, with which the hy- bridization had nothing to do. Yet some of these marks cor- respond with known results of hybridization. That offspring should partake unequally of the characters of the two parents is a matter of common observation. That in the genus Lilium the hybrid offspring should in forty 240 REVIEWS. instances out of fifty take almost all its traits from the female parent, as Mr. Parkman has shown, is very remarkable. That, in not a few instances, it should take them all, so far as can be seen, — that the paternal influence should be repre- sented by zero, — is most extraordinary. If parthenogenesis in plants were more unequivocally demonstrated, so as to be placed in certain instances quite beyond doubt (which is hardly the case), then we should regard the supposition which Mr. Parkman mentions as having been suggested to him, namely, that in the case of X. superhinn the embryo was devel- oped without male influence, to be quite as likely as the alter- native of the progeny's inheriting everything from the female and nothing from the male parent ; in fact the two suppositions approximate to the same thing. We are supposing tlie total absence of male parent's characters, and also that the alterna- tive of fertilization by chance pollen of the species is absolutely excluded. Of this there is very high probability, yet not entire certainty. One of Mr. Parkman's "reasons for be- lieving that parthenogenesis had nothing to do with the cases in question," namely, that some of the Lilies were young plants that never had bloomed before, has no application, but comes from a slight confusion of the idea of parthenogenesis with the effect in some animals of a previous male influence upon next succeeding progeny, which is quite a different thing. The fact that more than one sort of hybrid may be gen- erated between the same two species, copulated in the same way, must do away with the old mode of naming hybrids by a combination of the name of the two parents, that of the male preceding. The plan had the double advantage of indicating the origin of the cross, and of distinguishing hybrids from species in nomenclature ; but in practice it proves insufficient. PHYTOGAMY, PHYTOGAMY. 241 If this name has not been coined ah^eady it ought to be. For " the loves of the plants," so mellifluously sung by Dr. Erasmus Darwin in the days of our grandfathers, have been in our time, through a felicitous atavism, more scientifically, if prosaically, expounded by his grandson, in a series of articles and volumes, of which the subjoined are the principal titles.^ If we have too long delayed our notice of these books, we make amends by calling attention to them at the season which invites and amply rewards the observations in field and garden which they suggest. Mainly in consequence of these writings, the subject which our new word connotes, namely, the connu- bial relations of plants, has become a popular and fruitful branch of biological science, which has its own laws and rules and technical terms, its distinction of legitimate and illegiti- mate unions, and tables of forbidden degrees. For example, it is not lawful, at least it is not en regie nor beneficial, for '' thrum-eyed " Primroses to interbreed, nor for " pin-eyed " Primroses to interbreed. Such are illegitimate unions, seldom blessed with progeny. To the uncurious observer in Words- worth's poem, — " A primrose by a river's brim A yellow primrose was to him, And it was nothing more." But as concerns the Primrose, where seed-bearing is in question, if it be one of the thrum-eyed stock, the pollen brouo-ht to it must come from the pin-eyed, and vice versa, m orde? to secure full fertility. Tiny blue-eyed Houstonias, en- amelling our meadows in early spring, and fragrant INIitchel- las, carpeting Pine-woods in midsummer, are in a similar case. It is this kind of arrangement for cross-breeding to which 1 The Different Forms of Flowers on Plants of the Same Species. — The Various Contrivances hy ivhich Orchids are Fertilized by Insects. Second Edition, revised. - The Effects of Cross and Self- Fertilization in the lege- table Kingdom. By Charles Darwin. Loudon and ISew York, 18.0-i/. (The Nation, No. 667, April 11, 1878.) 242 REVIEWS. the larger part of Darwin's latest volume on "The Different Forms of Flowers on Plants of the Same Species " is devoted. In such flowers — and they are rather numerous and of many families — the advantage of cross-breeding between different individuals of the same species is unquestionable, for it is essential to full fertility. The differences in structure, which consist of relative and reciprocal length of stamens and style in blossoms otherwise alike, have long been known ; the mean- ing of it was one of Darwin's happy thoughts, and the con- firmation is due to his labors. He demonstrated that the structure was correlated to the transport by insects of the pollen of the one sort to the stigma of the other, and that each pollen was inert, or nearly so, upon the stigma of the flow^er it belonged to, but potent upon the stigma of the other sort, uj^on which, in passing from blossom to blossom among the plants (of about equal number as to sort), the visiting insects are pretty sure to deposit it. It is noteworthy that this significant dimorj)hism belongs to certain species of a considerable number of natural families, while others, sometimes even of the same genus, and in most of their species, show no trace of it; as if certain favored species had acquired a peculiarity in which their brethren have not shared. We ourselves call to mind some species in which this acquisition is either incipient or the correlations imperfect. But in his earliest work of the present series, on " The Vari- ous Contrivances by which Orchids are Fertilized by Insects," — a fascinating volume, which has recently been brought out in a second edition, — the " contrivances," as they may well be termed, are the common property of the whole order, although each genus seems to have patented a modification of its own. Here there is no dimorphism, but (with rare exceptions) all the flowers are alike, and all agree in having the pollen placed tantalizingly near the stigma, but prevented from reaching it, as well as in having some arrangement for the pollen's being transported by insects from one flower to another, ulti- mately from one plant to another. Wonderful arrangements, indeed, they are, which it requires a volume to describe, and of which we can here offer no details. Suffice it to say that, PHYTOGAMY. 243 in this great order, cross-fertilization must be all but universal as between different flowers of the same plant, and commonly between different individual plants. In both these kinds of hermaphrodite flowers the practical separation of the sexes is hardly less than in -Oaks, Willows, and other trees and herbs, in which the stamens and j^istils occupy distinct plants or different blossoms. To these three classes, then, Mr. Charles Darwin's ajihorism, " Nature abhors perpetual self-fertilization," undoubtedly applies. But there remains an equal number of plants with hermaphrodite blos- soms, all alike, with no obvious obstacle to fertilization with their own pollen, while in many the adaptations are such as must apparently insure it, and indeed does very commonly insure it. Wherefore it is nowise sur23rising that self-fertili- zation was the orthodox doctrine — that there was thought to be a general adaptation for the falling of the pollen upon the stigmas of the same blossom. It is true that Christian Con- rad Sprengel taught the contrary, in his work entitled " The Secret of Nature Discovered," published eighty-five years ago, and that he — mainly upon good observations — in a measure anticijDated Mr. Darwin's aphorism ; but he was accounted whimsical and untrustworthy by his own generation, and was forgotten by the next. Not so the contemporary '' Loves of the Plants" — the hymnal of the old orthodox cult — which sings the — " Gay hopes and amorous sorrows of the mead," in verse which our fathers were fond of, but from w^hich we will not further quote. Had Dr. Erasmus Darwin known Sprengel's book, and brought to it the insight of the grand- son, how different and how much richer the poem might have been. What curious facts and teeming fancies have been left unsung ! To H. Miiller and to Hildebrand, two of Sprengel's coun- trymen, in our own day, may be credited the confirmation of the latter's thesis as respects the general run of hermaphrodite flowers ; and this by showing what a large proportion even of these are functionally unisexual, either by the shedding of 244 REVIEWS. their pollen before the stigma of that blossom is ready to re- ceive it, or by the development and subsequent shrivelling of the stigma before the pollen matures, or by various other arrangements of like effect. And here, too, comes in the sig- nificant fact for the evolutionist, that these arrangements be- long to widely different families, but only to certain of their species or groups of species, and not to their near relatives ; also that they are more pronounced in some species than in others. Yet, withal, there is much close-fertilization, and no one has demonstrated this better than Mr. Darwin, nor so well illustrated its meaning. The more particular and special the adaptations for cross-fertilization — depending, as they mainly do, upon insect-transportation, consequent upon visits for nec- tar or other floral products — the greater the chances of no fertilization through the failure of the proper insect visitation. So nature, not scorning a succedaneum, arranges for self-fer- tilization also as the next best thing, indicating her preference, however, by endowing the pollen with greater potency upon other stigma than its own ; the principle throughout being to place the pollen where it will do the most good, all things considered. But Mr. Darwin insists, apparently with reason, that cross-breeding is the general plan, and close-breeding the subsidiary proceeding, or at least that no species of flowering plants is deprived of its chance of wide-breeding, or fails to receive the benefit of it for any long number of generations. This assumes that wide-breeding is beneficial. The assump- tion is one which a teleologist like Darwin is bound to make, and which an investigator like Darwin is bound to verify, if possible. The assumption is that ends elaborately brought to pass in a large number of species, in a variety of ways, and by great nicety and exactness of adaptation, cannot be mean- ingless or useless — must somehow conduce to the well-being of the species. Happily, this inference holds equally good whether, with the old-fashioned teleologist, the word " end " denotes a result aimed at, or, as in Darwinian teleology, a result attained. The two senses are not contradictory, and, as concerns the validity of the inference, it matters not which PHYTOGAMY. 245 sense is adopted, or whether the two are combined. Darwin's investigation, undertaken to determine by experiments whether such crossing is beneficial, is published in the remaining volume of the series under consideration — that on " The Effects of Cross and Self-Fertilization in the Vegetable Kingdom." It does not fall within the scope and limits of this notice to set forth the nature and the extent of these experiments. Headers interested will go to the book, and probably have done so already. As to the results we may only say that, on the whole, they corroborate the inference — in some cases unequivocally and strongly, in others feebly, while in a very few the result was simply negative. AVhile the crossing in many cases showed astonishing reinvigoration, and self-fertilization evident injury, the maximum good was obtained at the first or second cross- ing ; and some close-fertilized plants soon became tolerant of that condition, and retained their fertility for several close- bred generations. If the Darwinian thesis was on the whole maintained, yet it was also shown that plants have many in- explicable idiosyncrasies, and that many unknown or obscure factors enter into the results of the experiment. On looking over the series we are reminded of the late Jeffries "Wyman's aphorism : " No single experiment in physiology is worth anything." It seems reasonably made out that the benefit of cross is, ccetei'is 2)ctribus, in direct relation to a certain difference in constitution between the two parents, or to some difference in their surroundings or antecedents, from which diversity of constitution may be inferred. The benefit is more decided when the parents come together from a distance than when grown side by side for several generations, and " a cross be- tween two flowers on the same plant does no good, or very little good." The qualification is a proper one. It would be hasty to infer that it does absolutely no good, even though the advantage be inappreciable in any single instance. Still, however just and fairly well sustained the princl2)le of Dar- win's aphorism may be, it is confronted by the immense and seemingly endless vitality of long-propagated varieties which do not seed at all. 246 REVIEWS. If we were writing a popular review of this volume on cross and self-fertilization, w^e should make much of the tenth and eleventh chapters, on the means of fertilization, and es- pecially of cross-fertilization ; on the plants which are sterile, or more than half -sterile, without insect aid ; and, above all, on the habits of insects in relation to the fertilization of flowers. A closing chapter in the volume, on the Forms of Flowers, should also receive attention — that in which cleis- togaraous blossoms are discussed, namely, small and incon- spicuous ones which never open, but are far more fertile than the showy ordinary blossoms of the same plants ; for capital converse testimony, to the effect that all ordinary flowers are in primary reference to cross-fertilization, may be derived from the structure and behavior of these blossoms, in which the contrary intent is unmistakable. When nature means close-fertilization she makes her purpose manifest. Also, we should note that this cleistogamy is sporadic, affects certain families only, and certain members only of families not other- wise particularly related ; so that this peculiarity also seems to be of special and apparently late acquisition. When we gather into one line the several threads of evidence of this sort, to which we have barely alluded, we find that they lead in the same direction with the clews furnished by the study of abortive organs : slender, indeed, each thread may be, but they are manifold, and together they bind us firmly to the doctrine of the derivation of species. BENTHAM'S FLORA OF AUSTRALIA. This volume ^ brings a great undertaking to a happy com- pletion. The first volume was issued in the year 1863, and the work has made steady progress to the end. It is the com- plete phaenogamous Flora of a continent, and the only one ; ^ Flora Australiensis : a Description of the Plants of the Australian Ter- ritory. By George Bentham, assisted by Baron Ferdinand von Mueller. Roxburghiacece to Filices. London, 1878. (American Journal of Science and Arts, 3 ser., xvi. 237.) BENTHAM'S FLORA OF AUSTRALIA. 247 is worked up by one mind and hand, within a time and at an age which allows no sensible change of ideas or points of view, so that it is throughout comparable with itself. It is the work of the most experienced and wise systematic botanist of the day, and when we know that fully as much other work, of equal character, has been done within these fifteen years, it will not be denied that the author's industry and powers of accomplishment are unrivalled. No one else has done such good botanical work at such a rate. If, as some fear, the race of first-class systematic (pha^nogamous) botanists is des- tined to die out or dwindle, it will not be for the lack in our day of a worthy model. In the concluding Preface, Mr. Bentham turns over to his able and equally indefatigable coadjutor, Yon Mueller, the duty of incorporating addenda and corrections, and suggests the preparation of a methodical synopsis, for convenient use, especially in Australia, where such a handbook will be most helpful and needful. This trust, we doubt not. Von Mueller will duly undertake, and may be expected worthily to ac- complish. His fellow-workers over the world are not un- mindful of their great obligations to him in the development of Australian botany, and in rendering practicable the pro- duction of this " Flora Australiensis " which has been equally enriched by his vast collections and facilitated by his prelim- inary study of them. Mr. Bentham now declines to undertake " a detailed ex- amination of the relations, as well of the whole flora to that of other countries, as of its component parts to each other," referring instead to " the principles laid down by J. D. Hooker in the admirable essay prefixed to his ' Flora TasmaniiB,' '* but recapitulating shortly the general characteristics of the chief component parts of the present flora of Australia, the most peculiar one of any large part of the globe. Let us still hope that he may some day reconsider this determination, so far as to discuss in a general way the relations of Australian botany to the history of vegetation on the globe. Peculiar as the Australian vegetation is, its treatment not rarely touches points which concern the student of the Ameri- 248 REVIEWS. can flora. Especially interesting to us is the elaboration, in the present volume, of the Gramlnece^ in which General Mun- ro's matured views — as yet little known by publication — have passed under the independent consideration of a veteran general botanist, and in which the author's own conclu- sions regarding the morphology and terminology of the floral parts and their accessories are practically applied. We duly noticed ]\Ir. Bentham's essay on this subject, and had to acknowledge that its conclusions are apparently incontro- vertible. Next to this order in importance is the order CyperacecB, upon the arrangement of which sound judgment is brought to bear. The great order Liliacece is made to include the Smilacece^ and not the Roxhurghiacece. We should have excluded both, but Smilax in preference. Contrary to Mr. Bentham's opinion, we should insist that the anthers in Smilax are unilecular but bilocellate. The diagnosis of Box- hurghiacecB in the conspectus distinguishes the order from Australian Liliacese only, and by an oversight the second genus of the order is said to be restricted to Japan, whereas it was founded on a North American plant. DE CANDOLLE'S NEW MONOGRAPHS. In this form ^ and way we may hope to see the Monocoty- ledonous orders elaborated, and some of the earlier Dicoty- ledonous ones re-elaborated. The middle of this volume is filled by the monograph of Restiacece^ by Dr. Masters. This is an order allied on the one hand to Juncacece^ on the other to Cgperacece^ of twenty genera and two hundred and thirty- four species, wholly of the southern hemisiDhere, divided between South Africa (which has much the larger share), ^ MonographlrE Phanerogamarum Prodromi nunc continuatio, nunc revi- sio, anctoribus Alphonso et Casimir De Candolle, aliisque Botanicis ultra memoratis. Vol. I. Smilacece, Restiacece, Meliacece, cum tabulis ix. Paris, June, 1878. (American Journal of Science and Arts, 3 ser., xvi. 325 ; xxxiv. 490.) DE CANDOLLE'S NEW MONOGRAPHS. 249 and Australia with New Zealand, and a single species in Chili. It is not a prepossessing family, and presents peculiar diffi- culties to the systematist, on account of the dioecious character of most of them, and a striking difference between the plants of the two sexes, which in collections are hard to match. Much praise is due to Dr. Masters for his great labor, patience, and skill. The latter half of the volume is occupied by Casi- mir De Candolle with his neat revision of the Meliacem^ chiefly a tropical order. The stamineal tube in the monadel- phous Meliacem is concluded to be a staminiferous disk. The Smilacece by Alphonse De Candolle form the smaller but to us the most interesting part of the volume. This order is restricted to three genera : two of them dioe- cious, Heterosmilax with united sepals, no petals, and three monadelpLous stamens (east Asiatic), Smilax with separate sepals, petals, and (6-15) stamens ; the third, Rhipogonum (of New Zealand and Australia), with hermaphrodite flowers. Of Smilax one hundred and eighty-six species are character- ized, and a dozen or two more are obscure or doubtful. There are thirty-eight pages of prefatory generalia, in De CandoUe's best manner. We are pleased to find that he keeps up the sj^ecific 2)Jirase^ and with true Linnaean curtness, relegating all particulars, not truly diagnostic under the sections and other divisions, to the description. In discussing the nature and characters of the leaf (which in its general sense is called "recentement et assez inutilement phyllome") the morphology of the petiolar tendrils has to be considered ; the conclusion is that these answer rather to leaflets than to stipules, and the articulation, in some species well marked, between the blade and the petiole, or in the petiole, is noted as supplying good specific characters, which have been overlooked. The umbels are centrifugal or cymose. To distinguish, as is here done, the perianth into sepals and petals and to use these names when practicable, is most proper ; but it hardly follows that the term perianth or perigone will then have no raison d'etre. Whatever the number and position of the stamens, the carpels are superposed to the sepals, as indeed is the case in most Monocotyledons. It is pertinently noted that in Smilax, 250 REVIEWS. always dioecious, and with dull-colored perianth, the pollen is papillose as in most entomophilous flowers ; but that Rhipo- gouum, the only hermaphrodite genus, has a smoothish pollen, more like that transportable by the winds. Most have odor- ous blossoms, some pleasantly, some the reverse. De Can- dolle asks w^iether in our Coprosmanthus (the name of which indicates the ill odor) this is common to both sexes and the same in both. Can any of our readers speak to this ? An exposition of the geographical distribution of the order, and of what is known of it in a fossil state, is followed by a state- ment that all the four natural sections of Smilax and the two other genera — i. «_ _ a. d. 1883, «... 7,o85 Hooker, > If the last had been elaborated upon the scale of Endlicher, or with the idea of genera which is still common if not preva- lent, the number of genera would have amounted to at least ten thousand. An estimate of the number of known species of each genus and higher group has been made throughout the work — a rough approximation only, mentioning first the number in the books, and the number to which, in the opin- ion of the authors, these may probably be reduced by bota- nists who adhere to the Linnsean view of species ; from which it appears that upon the very strictest estimate their number, as now known to botanists, is at least 95,G20. In round numbers, it may fairly be said that about 100,000 species of phoenogamous plants are in the hands of botanists. The five largest orders, as well for genera as for species, are the follow- ing, and in this rank : Composite^, Leginninoscv, Orcliidem, Ru- hiacea?^ Gramlnece. The high standing of the Orchid family in the list will be a surprise to many. Linnaeus knew only a hundred species ; five thousand is now^ a moderate estimate — about half as many as there are of Comjjositcc^ which hold to their proportion of one tenth of the whole. In both fam- ilies every country and district is largely peculiar in its spe- cies and types. The far greater prominence of Compositce 358 REVIEWS. over Orchids is owing to tlie vast number of individuals in the former, their paucity in the latter. Those who desire to know the respective parts which the two authors have taken in the elaboration of the " Genera Plantarum " may be referred to a short article on the subject in a recent number of the " Journal of the Linnaean Society of London." Great thanks from all botanists are due to them both. BOTANICAL NOMENCLATURE.^ Sixteen years have passed since M. De Candolle laid before the International Botanical Congress held at Paris, August 16-26, 1867, a body of Laws of Botanical Nomen- clature, which he had drawn up for consideration by that assembly. The code was discussed by a special committee, afterward by the congress in full session, some modifications introduced, and it was then all but unanimously voted, " by about one hundred botanists of all countries: " "That these laws, as adopted by this assembly, shall be recommended as the best guide for nomenclature in the vegetable kingdom." The adopted code, with an extended comraentar}^, was pub- lished by De Candolle early in the autumn of the same year ; and an English translation, made by the lamented Dr. Wed- dell, appeared early in 1868. The " Laws," but mthout the more voluminous explanatory commentary, were reprinted from the English translation in this Journal in July of that year, occupying only twelve pages ; and some remarks and suggestions by the present writer were appended. As was then said, the code did not make, but rather declared, the com- mon law of botanists. It announced principles, systematically and perspicuously, and indicated their apj^lication in leading cases ; but many practical questions, as well as conflicts of ^ Nouvelle Remarques sur la Nomenclature Botanique. Par M. Alphonse De Candolle. Geneva, 1883. (American Journal of Science and Arts, 3 ser., xxvi. 417.) BOTANICAL NOMENCLATURE. 359 rules in particular instances, which would inevitably come up, were necessarily left to be settled when they arose. Ko small discussion upon certain detaHs has indeed ensued, in which our author, naturally appealed to, has taken an active part. In the first part of the present publication (of 79 pages 8vo), the discussions of the intervening years are summedtip and reviewed ; and a few changes, which experience has shown the need of, are proposed. In the second part the author takes up certain questions which the Paris Congress left un- touched, such as the nomenclature of organs (which he treated in his recent " Phytographie "), the nomenclature of fossils, and the rules according to which names and authorities should be cited when old genera are combined or reconstituted ; also some matters of orthography and punctuation are briefly con- sidered. Finally, in the third part, the laws adopted by the Paris Congress are reprinted, with the suggested changes. The alterations and additions are printed in italic type, so that they may be seen at a glance. We are not sure of an English edition ; and in any case it is desirable to make so important a publication as this generally known to our own naturalists. So our abstract and comments may run to some length. They are intended partly to illus- trate and reinforce the author's doctrine, in respect to matters upon which there is still diversity of opinion and practice, perhaps occasionally to offer a criticism or suggestion ; also as helps and guides to botanists in our own country, who are beginning to take interest in such matters and to feel (or at least to show) the need of giving attention to them. As respects all the weightier matters of the law, the most experienced phasnogamous botanists are in general agreement. Having accepted the code of 1867, they will be ready to accept proposed modifications and interpretations which are in accordance with its principles. Among them all, perhaps no one is so well qualified as De Candolle, by the bent of his mind and course of his studies, his opportunity of leisurely consideration, and his long editorial experience — combining, as it may be said to do, that of two generations, — for de- claring what the present consensus of authority is, and for 360 REVIEWS. discussing the cases which still need adjustment, some of them of considerable consequence in phytography. At the head of the publications upon nomenclature ema- nating from or sanctioned by associations or committees of naturalists which have appeared since the year 1867, De Can- dolle places the Report to the American Association for the Advancement of Science, at the Nashville meeting, 1877, on Nomenclature in Zoology and Botany, prepared by Captain Dall, after much conference with leading American natural- ists. The differences between these rules and those of the botanical code relate mainly to questions mooted by the zoolo- gists, wdiose systems and views had been comparatively loose and variant. The first alteration in De Candolle's revised draft is adopted from Dall's report, with high commendation, namely : — "Art. 3. The essential point in all parts of nomenclature is : 1. Fixity in names ; 2. Avoidance or rejection of forms or names which may create error or ambiguity or introduce con- fusion into science." The '' fixity in names " is taken from the American code, and is said to supply a real omission. Although merely declaratory, some practical consequences flow from it. The same idea dominates in the report made by Douville, chairman of a committee of the Geological Congress at Bologna in 1881, and which concerned itself with nomen- clature in palaeontology. This report insists that " The law of priority being fundamental in nomenclature, it appears to be necessary to apply it with all possible generality and to sup- press derogations and exceptions to this law. . . . Contradic- tion betw^een the signification of a name and the characters of a genus or species is no sufficient reason for changing such a name," etc. Another code referred to is one by a committee of the Zoo- logical Society of France, in 1881, M. Chaper, chairman, in which this principle of fixity is said to be less prominent, more exceptions being allowed. On reviewing the whole field, De Candolle assures us that the tendency during the last sixteen years has been : 1, to an increasing agreement of the zoologists with the botanists; BOTANICAL NOMENCLATURE. 361 and 2, an increasing recognition of the law of priority as the fundamental i:)rinciple of nomenclature, and as prevailino- over considerations of elegance, linguistic purity, and precise meaning. But some recent publications of individual bot- anists look the other way; notably so two publications by Saint-Lager, " Reforme de la Nomenclature Botanique," and " Nouvelles Remarques sur la Nomenclature Botanique," pub- lished at Lyons, in the Annals of the Botanical Society of that city ; of which it may be said, that the reflex effect of the multitudinous changes proposed in the view of makino- old and accepted names better, more classical, or more signifl- cant, has greatly strengthened the hands of those who contend for the absolute fixity and unalterableness of published names. Whether and to what extent misspelled or otherwise wrongly formed names may be corrected, and whether in some cases the principle of fixity should not prevail over absolute priority, are matters which may be discussed further on. We take uj) in the author's order the points which we wish to specify or to comment on. Article 6 of the code declares tliat " scientific names should be in Latin. When taken from another lanofuaofe, a Latin termination is given to them except in some cases sanctioned by custom." Here our author asks, " But what is Latin ? " He concludes that the Latin of Linn sens should be the model. It is the classical language of botany, and is much more pre- cise than the Latin of antiquity, in wliich very many words bear two, three, or half a dozen senses, either in the same or in different ages ; while in the technical language of botany each word has but one meaning, and each idea or object is ex- pressed by a single term. De Candolle elaborated this point in his " Phytographie," to which he refers for illustrations ; and he returns to it in special applications when commenting on article QQ^ as w^e shall see. But, in fact, even technical language cannot always avoid ambiguities and the use of words in senses which have to be determined by the context. Folium^ for example, in botanical descrij^tion, may mean the blade of a leaf only, or this along with its petiole, or blade with petiole and stipules together, or it may even mean any homologue of the ordinary leaf. 362 REVIEWS. " Art. 15. Each natural group of plants can bear in sci- ence but one valid designation, the most ancient, whether adopted or given by Linnaeus, or since Linnaeus, — provided it be consistent with the essential rules of nomenclature." De Candolle now adds an article 15^^^ which is purely explanatory, but has a bearing upon subordinate questions. It is: "The designation of any group, by one or more names, has not for its object the enunciation of its characters or history ; it gives merely the name by which we are to call it." He comments upon the tendency which is often shown to mix up the ques- tion of name with other considerations. Before Linnaeus introduced the binomial system, the names of sjDCcies were at the same time names and characters. In separating these two things, Linnaeus rendered a great service, and we should be careful to preserve this advantage. " A name is a name ; characters are characters ; the succession of names is syno- nymy. To mingle such different ideas leads to confusion. In these days there is a disposition to attribute too much im- portance to the meaning of names, and also to intermix the synonymy — i. e., the bibliographical history of the groups — with the names, at least mth the indication of the author, which being commonly annexed comes to be almost a part of the same. Such complications are contrary to the general principle that different ideas should be expressed separately. If this rule is neglected, we may be led into attempts to ex- press in the name, or with the name, the phytogenitic history of a group, that being just now one of the ideas in vogue." Names that have an appropriate meaning are very well, and botanists always endeavor to make such ; but experience shows that meaningless names are in some respects better — are generally better than names founded on ideas, which in the progress of knowledge often become false.^ De CandoUe's main remarks upon article 15 relate to the ^ Or may be essentially false from the beginning. One of our common Maples has two names, Acer dasycarpum and A. eriocarpum, both signify- ing that the fruit is woolly, whereas it is perfectly glabrous ; only the ovaries are woolly, yet no botanist has ever proposed to change the re- ceived name, — which is remarkable. BOTANICAL NOMENCLATURE. 363 point of departure for the law of priority in botany. The names of the two great classes, Monocotyledones and Dicoty- ledones, are of Ray in 1703, who scientifically distinguished and named them. The counterpart, Acotyledones, is of Jus- sieu (1789), and to all three names our author would apply the law of priority. That the name Acotyledones fails to express the true character is a small objection. This is only an example of the disadvantage of significant names, which may lose aptness in the advance of knowledge. Far better are such names as Aves, Pisces, Vermes, etc., which time and discovery can never falsify. Cryptogamia as the name of a class (which for meaning is hardly so good as Acotyledones) was introduced by Linnaeus in 1735, and has maintained its place.i The counterpart, Phauerogamia (or Phsenogamia), is not of Linnaeus, but much later. Natural Families or Orders date from Jussieu's " Genera Plantarum," in 1789 ; Cohorts and Tribes, from A. P. De CandoUe's " Systema," 1818. Subgenera begin with R. Brown, in 1810, according to our author. There are a few instances (without the name) in the " Prodromus Florae Novae Hollandiae." But it is in the Oudney and Denham-Clapperton paper, in 1826 (p. 16), that their use is discussed, and the mode of designating them in citation by interpolation between generic and specific name in parenthesis, is introduced. Species (as distinguished from varieties) and the actual binomial nomenclature date from 1753, and the first edition of the " Species Plantarum." There is substantial agreement among botanists as to this point of departure ; and the fact that the specific phrase of earlier authors is occasionally of a single substantive does not militate against it. Galega vulgaris^ Lci'p'pa majoi\ and Trifolium agrarium of the old herbalists were in good binomial form ; but the adjectives are phrases, not specific names. Generic names bring in a question of interpretation and usage. In his table De Candolle makes the point of departure for priority, Linnaeus, '' Genera Plantarum," ed. 1, 1737. All 1 On p. 58 De Candolle has collected nineteen synonyms of the name Cryptogamia, all of later date, and specified the objections which may be brought against each of them, besides that of want of priority. 364 REVIEWS. agree, or should agree, that no anterior name has right of priority to a Linnaean name or to a name adopted by Lin- naeus. But as respects generic names adopted by him, are we to follow Linnseus or are we not ? He says, " Tournefortius primus characteres genericos ex lege artis condidit." And in the " Genera Plantarum " : " Ipsi non immerito inventionis gloriam circa genera concedere debeam," — and so he uni- formly accredits to Tournefort the generic names adopted from him ; and the same as to " Plumerius, . . . Vaillantius, Dillenius, . . . Michelius et pauci alii," "qui ejus vestigia presserunt." De Candolle remarks that Tournefort had the merit which Linnseus ascribes, but that " he kept a good many adjective names for genera (Acetosa, Bermudiana, etc.)." Since Linnseus did not adopt these, they are out of the present question. Moreover, not to speak here of a score or two of really adjective generic names, Linnseus himself adopted two which Tournefort had discarded, Mirabilis and Impatiens, and deliberately made another, Gloriosa, in place of a proper name, Methonica, of a sort which, though not of the best, is now regarded as next to the best. But it is com- pletely understood that Linnseus is not to be corrected ; so Gloriosa, Impatiens, etc., remain. Are we equally to follow Linnseus in regard to names which he adopted from Tournefort and a few later authors, some of them his own contemporaries ? If so, we shall continue to write Salicornia, Tourn., Corispermum, A. Juss., Olea, Tourn., Justicia, Houst., Dianthera, Gronov., Lycopus, Tourn., Lin- nsea, Gronov. The practice of the leading botanists has been essentially uniform in this respect, from Jussieu down to De ^ Candolle, father and son, even to the latest volume of the " Monographia," published during the current year. It seems perfectly clear therefore — although we believe that the ques- tion is not raised in this revision — that such genera are ex- pected still to be cited as of their respective founders. And, as hardly any one doubts that Tournef ortian genera sujDpressed by Linnseus but restored by modern botanists (such as Fago- pyrum) are to be cited " Tourn.," it follows that only in a restricted sense do genera begin with Linnseus in the year 1737. This case, indeed, is governed by the principle in BOTANICAL NOMENCLATURE. 365 citation, so well insisted on by M. De Candolle, that no author is to be represented as saying the thing that he does not say. The alternative course is to write " Linngea, Linn.," which is certainly what Linnaeus has not said. The only authors we know of who have on principle followed this alternative — and a notable exception it is — are Bentham and Hooker in the new " Genera Plantarum," against which a protest was made in this Journal when the fii-st part was issued. They have followed the rule that botanical genera began with Lin- naeus so strictly as to cite even authors as recent as Ga^rtner for Tournefortian genera and to ignore botanists like Gro- novius, contemporary with Linnaeus, and publishing since the year 1737 ; and it is only by an infraction of their rule that they have avoided writing Linngea, Linn. No change of rule 15 seems actually required to bring it into unison with the almost universal practice in citation. We have only to understand that genera adopted by Linnaus from Tournefort, etc., and so accredited, should continue to be thus cited ; that the date 1737 (Linn^us, " Genera," ed. 1) is, indeed, the point of departure from which to reckon priority, yet that botanical genera began with Tournefort ; so that Tournefortian genera which are accepted date from the year 1700. That is the limit fixed by Linnaeus, and it defi- nitely excludes the herbalists and the ancients whose writings may be consulted for historical elucidation, but not as author- ity for names. Upon articles 21 and 22, which give rules for the names of orders and other supra-generic groups, our author offers no new remarks. We venture to offer two. It being the general rule that acece is the proper termination for ordinal names which take their appellation from a typical genus, it is desirable to conform to it as fully as well may be.^ Since Saxifragacece^ 3fyrsineacece^ StyracacecB^ GentianacecB, Nyc- taginacecE^ and even Lauracece and Juglandacece were adopted - Crucifer(By Leguminosce, [/mbelli/ercp, Composite^, Lahiatce, and the like, are no exception to the rule, rightly stated, as they are not named from typical genera. "We shall not have any more of them, but the old ones in use are among the best. 366 REVIEWS. in the " Proclromus," it seems to us retrograde and unadvis- able to have gone back in the new " Genera Plantarum " to Saxifnif/ecv, and the like ; and this upon no obvious prin- ciple, as we have jSamydacece, Oornacece, etc., brought into harmony with the rule. And if in the " Prodromus " we have Sti/)'acacece, why not also Salicacece f And if, in the " Genera Plantarum," Styrax could take the acece termina- tion as Sti/racece, why not as Styracacece ? If it be objected that some such terminations have an unclassical aspect, this objection applies all the more to the cases under our second remark : Xamely, that too rigid adherence to the rule that names of suborders, tribes, etc., shall end in -ece and the like, gives us nearly unpronounceable words of four or five con- secutive vowels, or, when the diphthongs are printed in sepa- rate letters, according to a prevalent fashion, one or two more. Of these — the diphthongs written out — Sauraujeac, Spiraeeae, Cateshaeeae^ Jaumeeae^ Thymleeeae^ and Moraeeae are the worst instances, and would justify any infraction of rules.^ The last, and one of the worst, would have been avoided by writing the ordinal name IridacecE when that of the tribe would have been Iridcoe. Names are to be spoken as well as read, and botanists who have to teach think more of these things than those who only write. At the head of his remarks upon generic names (art. 25 et seq.), our author commends to other naturalists the very clear directions given in the rules for Zoological Nomencla- ture, edited by Dall, for rendering Greek letters into Latin in the construction of generic and specific names. He notes, however, that the rendering of -7 by e is not in full accordance with Latin usage, as witness hihiiotheca, dialectica, Hecuhea^ etc. 1 Far better to write SpirceacecE, with De CandoUe. The use of this termination for tribal names need not be objected to by those who take little pains to use it for orders. And those of us who are careful so to employ it, would prefer its occasional use for tribes and suborders to the concatenation of vowels, which it is not easy to write and almost impos- sible to pronounce. Some quite unnecessary tribal names in acece, such as Vernoniacece and Eupatoriacece, adopted by De Candolle from Lessing, are kept up, although exceptional. BOTANICAL NOMENCLATURE. 367 Article 28 of the code, which is reprinted without altera- tion or comment, consists of a series of recommendations of points to be attended to in the construction of generic names. As such names are to be " in Latin," it would have been well to recommend that in their formation from the Greek, from which most of them are nowadays drawn, the principles of Latin prosody should not be wholly ignored. Such names as Trichocliidus and Ancistrocladus would have been euphonious as Trichocladia and Ancistrocladia, and very little longer. Acanthopanax, Didymopanax, Dimorphochlamys, Trigono- chlamys, Aulacocalyx, Pellacalyx, Microcharis, and tbe like, are harder to pronounce than the makers probably thought ; and so of many others. One of the actual recommendations is " Eviter les noms ad- jectifs." This in Weddell's version is translated, "To avoid adjective nouns:" doubtless a wrong translation. Adjective nouns we take to be substantives which are directly formed from adjectives. Not many such are likely to be made for genera ; but if such good ones can be constructed as those we already have in Nigella, Amarella, Flaveria, Chlora, Rubia, Leucas, and Hyptis, they will not be objected to. Clearly the recommendation is to avoid adjective names for genera. That may be done for the future, but has not been done in the past. They are contrary to the rule of Linnaeus, but not to his practice. Not to refer again to Gloriosa, Mirabilis, and Impatiens, at least two score of obviously adjective names for genera may be counted in the first edition of the " Genera Plantarum," — such as Arenaria, Stellaria, Utricularia, Den- taria, Asperula, Angelica, Trientalis, Pedicularis, Digitalis, and from the Greek such as Polycarpon. Amphicarpum and Mitracarpum are recent names of this kind. To conform the rule to the fact it were better to state that : generic names are either substantives or adjectives which may be used as sub- stantives, the latter mostly feminine in gender. Angelica is understood to be Planta Angelica, Sanguinaria, Planta san- gidnaria, etc. It is recommended " to avoid making choice of names used in zoology." But it has become nearly impossible to follow this advice, nor is it now thought to be important. 368 REVIEWS. Article 33 is suppressed. It was no more than a statement of the custom that personal names for species were to be nouns in the genitive (e. g.^ Clusll} when the person com- memorated was a discoverer, describer, or an illustrator of the species, but were in adjective form (e. ^., Clusianci) when the name was merely complimentary. The rule sometimes worked awkwardly; for many personal names do not take kindly to latinization in the genitive form, which are suffi- ciently euphonious as adjectives in -ana or -anum ; and it is well to do away with a needless restriction. Article 34, which recorded the fact that many names of spe- cies are substantives, is the subject of a few remarks, called out by the publication of Saint-Lager, in which it is proposed to change all these into adjectives, — a proposition which bot- anists are not likely ever to adopt. Some of them are among the best of descriptive names, e. ^., Stellaria neinorum^ Con- volvulus sepiu7n, Rhus vernix^ Chamcerops Jiystrix (and such should be written without a capital initial) ; those which are proper names, either old names of genera or of the her- balists, are rightly said to be significant either of the absorp- tion of a former genus, or of a transference, or as preserving a native appellation, or as indicating a likeness. " Digitalis Sceptrum means a Digitalis which had been called Seep- trum ; " Ardisia Picheringia^ a species of this genus which, mistaken by Nuttall for a new one, had been named Picker- ingia ; Hudheckia Heliopsidis, a Rudhechia facie Ileliopsi- dis, from its resemblance to a Heliopsis. Linnaeus gave us many such names ; and no sufficient reason appears either for discarding these, or for forbidding the discreet adoption of new ones. But we cannot commend such a name as Senecio Bhot for a species indigenous to Bhotan. Article 36 consists of a series of recommendations for the formation or adoption of specific names. Its fifth sub-article we may refer to in another connection, namely, along with arti- cle 48. The recommendation to " name no species after one who has neither discovered, nor described, nor figured, nor studied it in any way," should be respected ; yet there are oc- casions for departing from it, especially in case of new species BOTANICAL NOMENCLATURE. 369 in very large genera. Excellent and sometimes needful is the advice to " avoid names designating little known or very limited localities." We are obliged to cite — haj^pily as a synonym — Helenium Seminariense., published by a Pro- fessor who thought he had discovered a new species of Hele- nium in the vicinity of the " seminary," in one of our south- ern States, where he taught botany. Article 40 suggests that names of varieties originated in cultivation, and still more half-breeds and sports (so impor- tant for horticulturists to distinguish), should have only fancy names, generally vernacular, and in some form as different as possible from the Latin specific names of botany, — names which, when needful, may be appended to the botanical name of the species, when that is known, e. g.^ Pelargonium zonule^ Mrs. Pollock. This has been seconded by the editor of the " Gardeners' Chronicle " and other judicious experts, and is slowly making its way. Article 42, treating of the conditions of publicity, is the subject of additional remarks. The rule is, that " Publication consists in the sale or the distribution among the public of printed matter, plates or autographs. It consists, likewise, in the sale or distribution, among the leading public collec- tions, of numbered specimens, accompanied by printed or auto- graph tickets, bearing the date of the sale or distribution." De Candolle now remarks that distribution among the mem- bers of an exchange club, of collections not offered to the public, does not come up to the rule ; also that, as Dr. J. Miiller states, the distribution of specimens without charac- ters or any indication of the reason for calling it new, is nearly tantamount to announcing a species or genus in a pub- lication, but without characters ; which article 46 declares is not publication. But the cases are not quite alike. The possession of the named specimen enables a botanist to ascer- tain its distinctions. A published description without access to specimens may or may not serve the same purpose, very often does not. Unfortunately an insufficient or even a misleading description — and we have many such to deal with — claims the same right of priority that a good one does. 370 REVIEWS. It is well, therefore, that publication by sufficient distribution of named sj^ecimens should be recognized. But the remark is true that, in fact, very few distributed collections fulfill all the requirements of article 42. Article 47, sect. 2, recommends botanists " to publish no name without clearly indicating whether it is that of an order or of a tribe, of a genus or of a section, of a species or of a variety — in short, without giving an opinion as to the nature of the group to which the name is given." Unless this is attended to, and unless citations are equally precise, — for in- stance, unless subgeneric names are cited as such and not as generic, and vice versa., — much confusion in synonymy and in indexes will ensue. Article 48, on the citation of the authority for generic and specific names, and matters herewith connected, involves ques- tions which have been more disputed than any other. In the revised article the phrase printed in italic type is interpo- lated : " For the indication of the name or names of any group to be accurate and complete, and for the ready verification of the date, it is necessary to quote the author who first published the name or combination of names." ^ The statement miijht ^ M. De Candolle appears to insist upon this verification (or incipient verification) for higher groups as well as for genera and species, and would deprecate the not unusual custom in compendious Floras, Cata- logues, etc., of omitting to cite the authority for orders, suh-orders, tribes, etc. We should agree with him if the omission was held to signify that the names of the groups in question, when thus simply given, were proposed as new. But in fact, authority is omitted, not because the groups are new, but because they are old and entirely familiar. Nobody will ever suppose that RanunculacecB, Clematidece, etc., nakedly written, are novel- ties. In this regard, the nature and plan of the publication are to be con- sidered ; what is necessary in a Systema or a Genera Plantarum may be superfluous in a local or a compendious work. Indeed it may be nearly impossible to assign the authority for the name of an order correctly, without explanation and extended references. One would wish to write, succinctly, Cornacece, DC, as De Candolle founded the order, but it was in the form Cornece. A well-known order was instituted as Onagrce, Juss., "Genera Plantarum," which the founder altered to Onagrarice ; for reasons referred to in this article we may wish to adopt the form OnagracecE. As all this is most familiar matter, yet may not be correctly BOTANICAL NOMENCLATURE. 371 be simplified by omitting "or names" and "or combination of names," on the ground that the name of a plant is one, that it has a name not names, its name being the generic followed by the specific appellation. Ranunculus hulhosus is one name of two words. Our veteran botanist, Bentham, has in- sisted upon this ; and it has a bearing upon the mooted ques- tion of mode of citation of authority. The governing principle for the citation of authorship, etc., is well declared by De Candolle : " Never make an author say that which he does not say." It is difficult to go wrong when this principle is kept in mind, and when it is also understood that the appended name of an author, or its abbreviation, makes no part of the name of the j^lant, but is only the initial portion of its bibliography. Those who take a different view seem to have fallen into it by failing to distinguish strictly between name and history, and especially by mixing the his- tory of a preceding with the statement of an actual name. A single example may illustrate this. When we write " 3Iatlii- ola tristis, Brown," we give the name of a certain kind of Stock and the original authority for it ; and we may, when needful, complete the citation by adding the name of the book, with the volume and page, where it was first published. If, with some, we write " 3fathioIa tristis, Linn.," we make an untrue statement. Linnaeus had a wholly different genus 3fathioIa, and no 3f, tristis. If we add " sp.," and somewhere explain its import to be that the latter half of the name was given by Linnaeus, the other half remains unaccounted for. And we have still to seek in the synonymy for the name of the genus under which Linnaeus knew the plant, and also for that of the author who transferred it to Mathiola. If, with others, we write " Mathiola tristis, Linn. ( Cheiranthus^,'' or " Mathiola tristis, Linn. (^suh-Cheirantho'),'' our longer phrase still wants stated without somewliat detailed exposition, why not in a local or con- densed botanical book write simply Onagracece ? The proper exposition is in place in a Genera Plantarum ; and it would have been better if Ben- tham and Hooker had critically attended to this, instead of referring merely to the preceding work of Endlicher. It would have added some- what, yet not very much, to their great labor. 372 REVIEWS. the essential part of the citation. If, to secure this, we write " JIathioIa tristis, Linn. (^Cheircmthus, Brown)," our name, if it may be so called, now extended to five words and two signs in print, or of seven words when spoken, is still ambiguous and confused. It is a jumble of synonymous names and author- ities, which become explicit and clear only when we translate it into '"'' JlatJiiola tristis, Brown (^Cheiranthus tristis, Linn.)," that is, into name and synonym, with respective authorities. This is clear and literally truthful ; the injection of the sy- nonymy into the name is neither. Linnseus reformed nomen- clature by freeing the name from the descriptive phrase. The school in question would deform it by rebuilding, in another way (as De Candolle observes), ante-Linnsean phrases, only making them historical instead of descriptive. The practice of appending the authority to the name when- ever the species is mentioned has been so strictly and pedan- tically adhered to that many take the former to be a part of the name. To obviate this impression, it might be well to treat the names of common plants as we do those of genera ; that is, to omit the reference to authorship in cases where there is no particular need of it. Not, however, so as to cause any confusion with the cases referred to in the following j)ara- graph : — " When a botanist proposes a new name ... it is impos- sible for him to cite an author ; consequently the absence of such citation suffices to show that the name of the species or other group is new. Linnaeus, Lamarck, De Candolle, R. Brown, Martins, etc., followed this course. It is then a useless complication of many modern naturalists to append ' milii,' ' nobis,' 'sp. nov.,' 'gen. nov.,' etc., to a new name. A large majority of species, genera, and families w^ere published with- out these wholly personal indications." This is good as a general rule ; but the " gen. nov." and an indication of the order or tribe are often needful. No new comments are made upon article 49, probably be- cause the practice of botanists generally is conformed to it. The article reads : " An alteration of the constituent charac- ters, or of the circumscription of a group, does not warrant the BOTANICAL NOMENCLATURE. 373 quotation of another autlior than the one that first published the name. . . . When the alteration is considerable, the words ' mutatis char.,' or ' pro parte,' or ' excl. syn.,' ' excl. sp.,' etc., are added," etc. The translation would have been better Vv^orded, " does not warrant the quotation of another author in place of the one that first published the name." For, in fact, the addition of the reforming author's name to the citation is often warranted and helpful, sometimes is almost a neces- sity, in the case of genera. It appears that R. Brown began, in an oblique way, the practice objected to, and for wliivh there is often a plausible excuse ; and the elder De Candolle sometimes followed it. It was only when the practice was systematically carried out by one or two authors that the con- sequences became apparent ; for few genera or species have now their Linnsean limits or signification, and the new rule was practically proved to be a necessity. Among the recommendations contained in article 36 was the following : " Readily adopt unpublished names found in travellers' notes or in herbaria, unless they be more or less defective." Guided by the practice of the elder De Candolle and his contemporaries, it used to be thought a duty, or at least a part of common courtesy, to do this, in all cases in which the author's approval could fairly be supposed. But certain inconveniences and misunderstandings have resulted, especially as to mode of citation, which have suggested its withdrawal or modification. In this Revision, M. De Can- dolle only adds the restriction, " or unless the author has not in advance approved the publication." This does not alter the case except for living authors : their approval ought to be obtained or counted on ; and in respect to authors no longer living a botanist takes up only such names as in his opinion ought to be published, and which he supposes the posthumous author would have approved. On the whole it were probably better not to take up names left unpublished by a deceased botanist ; and De Candolle assigns good reasons for letting them alone. If he had modified the article decidedly in this sense he woidd have more fully expressed his own view, and probably have been sustained by prevalent opinion. More- 374 REVIEWS. over, he might have distinguished this practice from an essen- tially different one, namely : — The case of plants sent under manuscript names by a dis- coverer or an investigator to some botanist engaged in publi- cation, and with a view to their publication; of which the sending by Nuttall of new Umhelliferm to the elder De Can- dolle when elaborating that order for the " Prodromus," is a marked and not unusual instance. For this is a practice that need not be discouraged. Any small inconvenience that may arise as to mode of citation is counterbalanced by the greater concentration of publication, new genera and species thus appearing in monographs, floras, or in the papers of leading botanists, which otherwise would have dimly seen the light in obscure or local periodicals. And they are more likely to have proper characters assigned to them, instead of vague descriptions, by incompetent or unpractised hands, such as often try a botanist's patience. Article 50 treats of the mode of dealino: with such names as the above mentioned after they have been published, i. e., " names published from a private document, an herbarium, a non-distributed collection, etc." It declares that such names "are individualized (Fr. precises) by the addition of the name of the author who publishes them, notwithstanding the contrary indication that he may have given." This is found to mean that, although the elder De CandoUe gives us " Eu- lophus, Nutt.," as the name of a genus communicated by Nuttall, with a specimen, for the purjDose of its being so pub- lished in the fourth volume of the " Prodromus," yet subse- quent writers, looking only to the work it was published in, are to cite it as Eulophus, DC. And that the genus which Linnseus published as " Linnaea, authore Clariss. Dr. Grono- vio," we are to cite as Linnaea, Linn. This is not only quite contrary to the practice of botanists from Linnaeus down to De CandoUe and later, but is also contrary to the golden rule of citation, already referred to, never to make an author say something different from or opposed to that which he does say. Appreciating this, the author of the code has now recast BOTANICAL NOMENCLATURE. 375 article 50, so as to read, " When an inedited name has been published (by another botanist), in attributing it to its author, those who afterwards mention it ought to add the name of the person who published it ; for example, Leptocaulis, Nutt. in DC. ; Oxalis lineata^ Gillies in Hook." This is reasonable, and in the first instance such names will almost of necessity be so cited, must always be so cited when work, volume, etc., are specified. But De Candolle remarks that the addition will soon vanish ; for instance, *' Cynoglossum ciliatmn, Douglas, Mss.," published by Lehmann in " Pugil- lus," etc., and in Hooker's " Flora Boreali Americana,*' will soon come to be quoted simply as " Cynoglossum ciUatum, Dough," that is, just as other names are quoted. And why not ? Because, it is said, the name dating only from the pub- lication, it is necessary to know when and where this vicarious publication was effected. For this " Nutt. in DC." may fairly serve, nearly all names published by De Candolle being con- tained in the " Prodromus." Not so, however, with " Gillies in Hook." Sir William Hooker published very widely, in periodicals, in the " Botanical Magazine," and in numerous independent works. In such cases the double citation gives little help. The experienced botanist may know where to look ; the inexperienced must turn to indexes at once ; for both these must be the final and the usual resort ; and in them the double has little if any advantage over the single citation. Moreover, if this principle is fully applied, the number of double-cited names may be inconveniently numerous. The first volume of Torrey and Gray's " Flora of North America " abounds in species and genera published by them for Nuttall. If these have all to be permanently quoted " Nutt. in Torr. & Gray," why not also the many species published, say by Bentham in De Candolle's "Prodromus," in the "Flora Brasiliensis," etc., and even the species published by Brown in the second edition of the " Hortus Kewensis," and else- where? On the whole it seems probable that these double citations will be used only in first or in early quotations, or in special instances; that it will not be deemed necessary to retain them when the names become settled in Floras or 376 REVIEWS. general works, except in the bibliography or full reference ; when of course the " Lei)tocauUs inermis^ Nutt. in DC. Coll. Mem. V. 39, x. 10, et Prodr. v. 107," will fully appear. But so long as the abbreviated citation of the author and publisher together is requisite, the mode or citation recom- mended by De Candolle is the one to be employed. A quite different case is that of citing, as authority for a genus or species, the name of a botanist which is not upon the record. There is reason to believe that L. C. Richard edited the "Flora Boreali- Americana " of Michaux, and drew up the excellent generic and specific characters of the new plants in it. There is equal reason to believe that he pur- posely withheld his name. Upon no just principle of cita- tion, therefore, can the name of Richard be quoted, as the 3^ounger Richard and Kunth essayed to do. The same holds for the work of Solander in the first edition of Alton's " Hortus Kewensis." And if it does not hold for the con- tributions of Brown to the second edition, it is because he claimed them in his lifetime, rather than because they have been collected and republished under his name since his death. Only confusion will come from the admission of hy- pothetical constructive authorship. The old rule, that what does not appear is no better than non-existent, must apply to all such cases. In the comments uj^on article 52, the duty of abbreviating authors' names in the normal way is insisted on, and the bad practice of doing so by leaving out the vowels is deprecated. " Michx." for Michaux, which is partially shortened in this way, was a necessity on account of the ancient botanist Micheli. But "Crn." for Crouan is intolerable. Such a name need not be abbreviated at all. Monosyllabic names should rarely if ever be curtailed. "R. Br." has so long been used for Robert Brown that it may continue to be used, although " Brown " is better. In the other form, it may be counted among the few cases in which initial letters are used instead of the first syllable and first consonant of the second, — cases which should probably be restricted to the** "L." for Linnseus, "DC." for De Candolle, "H. B. K." BOTANICAL NOMENCLATURE. 377 for Humboldt, Bonpland, and Kimth. We are not sure that De Candolle would favor the latter. A series of remarks is made upon articles 29-6G, taken together, — relating to names which are to be rejected or modified, and those which are to be maintained notwith- standing certain faults. As already mentioned, the tendency among working naturalists is to preserve names in spite of faults; while a few linguistic reformers, such especially as M. Saint-Leger, propose changes which would affect 733 re- cognized names of species in Europe alone, and ten or twelve thousand in the vegetable kingdom at large, and " this after all the endeavors of botanists for the past half century to establish the law of priority and to have more stability of names. Sagittaria safjittifoUa and Psamma arenaria must be changed, forsooth, because they are pleonasms ; all substan- tive specific names, because a great majority of specific names are adjectives, and many others because they are not suffi- ciently classical." For instance, Dianthus he would change to Diosanthus, Mentha to Miuthe, Hydrocotyle hrevijyes to J7. hrevipedata^ Cactus to Cactos, Arum to Aron, and so on. De Candolle adds, that Cicero was not so particular in Latin- izing Greek words, as witness " barbarus," " machina," " em- porium." As to Pirus, our author insists that even if Pyrus is not Latin, it is the botanical name of the genus as adopted by Linnseus (from Tournefort and from all the herbalists, and it is old enough to be entered as an alternative form in the dictionaries), and so is to be preserved under the law of pri- ority. There is little danger that the reform of Saint-Leger will prevail. There is some danger that the reaction will so stiffen the rule of priority as to forbid the correction of ob- vious mistakes. See, for instance, the form in which article 60 is now recast by De Candolle : " A generic name should subsist just as it was made, although a purely typographical error may be corrected. The termination of a Latin specific name may be changed to bring it into accordance with its generic name." From this it would seem that a slip of the pen and a mistaken orthography of a man's name may not be corrected. We trust that, when the change would not 378 REVIEWS. sensibly affect the place of a name in an index, such obvious ■coiTections as of Wisteria to Wistaria may prevail. We may assume that the error was typographical ; for Dr. Wistar was at the time too well known in Philadelphia for Nuttall to have been ignorant of the orthography of the name. The correction of Balduina into Baldwinia brings it into accord- ance with the rule that personal names used for genera sliould be written as near as may be with the original ortho- graphy of the person's name. "'Astragalus ahorlgi?io7mm" is neither a typographical nor a clerical error. It is a hard rule that forbids us to write " aboriginum," still retaining Kichardson's name as authority. Botanists may take more kindly to the rule when apj^lied to such names as Eleocharis and Aplopappus, in the forma- tion of which the Greek aspirate was neglected. We cannot well suppose this to have been a typographical or clerical over- sight on the part of Kobert Brown or of Cassini. Perhaps a majority of botanical authors have preserved the original or- thography, on the ground that the right of priority, like that of a certain king, is super grammaticum^ — while the re- mainder have written Heleocharis and Haplopappus ; whence some confusion in the indexes. The requirement to preserve the original form of the generic and specific names and to abide by the Latin of Linnaeus and Lis contemporaries, notwithstand- ing classical faults, enables us to retain such familiar names as Ranunculus acris, Lathyrus palustris and sylvestris (in- stead of R. acer^ L. paluster and silvestris^^ and to keep up " laevis " for smooth, — probably to the disgust of classical scholars. M. De Candolle has a note on Diclytra of Borckhausen, changed into Dielytra to make it conformable to a conjectured meaning, and then into Dicentra that it might agree with the etymology given by Borckhausen himself : he gives it as a case in which an excess of erudition has loaded the genus with three names in place of one ; and he concludes, as do we, that it were better to have kept the original orthography, and have treated it as a name which had, through some mistake, failed of meaning. But the name having been changed into BOTANICAL NOMENCLATURE. 379 Dicentra on the ground that the right word Kevrpov, and not the impossible word KXvrpov, must have been intended by Borckhausen, we think it should now be maintained, althou"-h it might have been left in the original form. Moreover, the doctrine that names must not be mended and that sense is un- important, however good and needful, is so recent that it must not be too rigidly applied to long-standing cases. This consideration should not be wholly overlooked in the case of old and long-established genera, especially those of numerous species for which some obscure older name has come to light. Since it is impossible to make rules for the infraction of a rule, such cases must be left to sound discre- tion. In our opinion such discretion would forbid the trans- ference of the name Stylidium from Swartz's genus to Marlea, and the revival of Labillardiere's transient first Candollea for Swartz's Stylidium. The fourth section of article 60, which enjoined the rejec- tion " of names formed by the combination of two languages," is now suppressed. Nothing is put in its place; but let us hope that we shall not be driven to the acceptance of the specific name " acuticarpum " which one of our fellow-bota- nists has recently perpetrated. Although hybrid names are to be avoided, yet, as De Candolle remarks, they cannot con- sistently be outlawed by people who accept "centimetre," " decimetre," " bureaucracy," " terminology," and the like, nor by botanists who raise no objection to " ranunculoides," " scirpoides," " linnseoides," " bauhinioides," etc. Names of identical meaning but of different orthography, as our author insists, may well enough co-exist. In a vast genus it might be neither inconvenient nor harmful to main- tain species named respectively " fluviorum," " fluvialis," and " fluviatilis," at least if they belonged to different parts of the world. We pass to some brief annotations upon the second part of the publication before us, which deals with questions not taken up by the congress of 1867. The first topic is that of the nomenclature of organs, wliich was treated with some fullness in the " Phytographie." The 380 REVIEWS. remark is here repeated tliat tlie greater part of the so-called names of organs are only terms, that is, names indicative of the condition of organs or parts of the plant. For some of these substantive names are necessary or highly convenient, yet most were better provided with adjective terms onlj^, which belong to terminology, not to nomenclature. Doubtless principles of fixity and the rule of priority should apply to these, both to names and to terms. But it seems unlikely that the phytotomists will at present heed the counsels of the phytographers in this matter. Yet the latter may insist that established names used in descriptive botany shall not be displaced on the pretense of getting more appropriate ones. For instance, the long recognized name " testa " for the outer seed-coat is to be discarded, because, forsooth, this covering is not always or even not generally a shell, or of the texture of earthenware. As well ask the French to discard the word "tete " (or "teste"), because the human head, or the skull which gave the name, does not really resemble a brick or earthen pot. The second is upon the nomenclature of fossils. And the rule is that they are named according to laws which apply to living plants. The Bologna congress of palaeontologists or- dained that, to secure priority for specific names of fossils, they should not only be described but figured. De Candolle, after consultation with Heer (whose recent death we have to deplore), concludes that this rule is too absolute. It seems to us that so long as a large part of the names of fossil plants are merely tentative and provisional, we should be content with a general approximation to the received rules of botany. The nomenclature of groups inferior to species (varieties, sub-varieties, variations and sub-variations) is considered ; but no new rules are proposed ; nor is the question of sub-species discussed. Although it is not exactly a matter of nomenclature, we should have liked that our author had considered the two modes of disposing of varieties, and had expressed an opinion as to whether the character of the species should or should not completely cover the variety or varieties assigned to it. In BOTANICAL NOMENCLATURE. 881 tlie former case there is a variety a, followed by /?, etc. In the latter, the species is defined upon its type, without any special regard to the appended variety or varieties, which are then characterized as to the points in which they differ from the type. We prefer the latter method, as being on the whole clearer, and as a saving in names ; avoiding the awkwardness or the superfluity of a varietal name for the type of the species. Some noteworthy observations are introduced in respect to the plight which systematic botany is threatened with by what De Caudolle would call micromorphic botanists, like Jordan and Gandoger, who abandon the Linnsean idea of species altogether, and give this name and rank to what ordinary botanists take for sub-varieties. For example, we are in- formed that M. Gandoger divides the Roses of Europe and North America into 4600 species, or groups provided with names similar to those of species, under numerous subgenera, which in effect take the place of genera. Mentha has already undergone a similar micro-metamorphosis. If this goes on, and the names should be written every one, I suppose that even the world itself could not contain the books (or indexes) that should be written. The obvious and only remedy is to rele- gate this kind of botany to a world of its own, with which the legitimate science need have nothing to do. Questions having been raised as to the proper use of capital initials in certain specific names, M. De Candolle has devoted two or three pages to this and related topics. Linnaeus used capital initials only for substantive names ; Lamarck employed them for personal and some geographical names, seemingly without system. A. P. De Candolle used the initial capital systematically for all three, and even for " Alpina " when used to designate a plant of the Alps. His example has generally been followed until recently ; and this is in accordance with the custom of the English language. To the objection that it is contrary to the customs of the Latin language, our author replies at some length, substantially as follows. He finds that in the matter of orthography, etc., classical writers distinguish nine phases or periods of the Latin language, of which the most classical is the seventh period, that of Augustus ; and 382 REVIEWS. there is no foundation in classical Latin for either punctua- tion (the points distinguishing words, not phrases) or accen- tuation by signs, and that the distinction between capitals and small letters was made since the dark ages by scholars whom a purist of our day might tax with ignorance of the proper way of writing Latin ; that the object and result of all these and other innovations was greater clearness and pre- cision ; that the question is not at all one of ancient latinity, but of modern usages, both of the philologists and the natu- ralists ; and these have happily modified classical Latin into a medium of greater precision and clearness and better adapted to the needs of science. Finally, we have a brief discussion of the question : " When an author has comprehended one genus in another without naming a species, can he be cited for the names of the species which implicitly result from this union?" The answer he decisively gives is : " This would be neither right nor possible, nor ^practically convenient. To be correct, one should attrib- ute to an author exactly what he has published. When it is said that the genus B should be united with the genus A, this is not saying that the species of the genus B should be called by such and such names in the genus A. To name them cor- rectly it is necessary to examine them one by one. A glance at the Genera of Bentham and Hooker, or at the works of Baillon, will show how impossible it is to attribute the desig- nation of the species to the authors who have changed the names of the genera, without an explanation under each species." This is illustrated by the supposed case of three genera combined into one, each of which has a species " lanceo- lata " ; by the case of a species " minor " transferred to a genus of which it may be the largest species ; and by reference to the state of all large and many small genera, full of obscure, mis- understood, or debatable species, the arrangement and naming of which can be effected only by patient and prolonged study. W^hen this work has not actually been done by the reformers of genera, it should be left to monographers and the editors of Floras. If, by article 45, " a species is not looked upon as named unless it has a generic name as well as a specific one," BOTANICAL NOMENCLATURE. 383 neither is it named unless a specific as well as a generic name is assigned to it. Besides tlie instances in which the old spe- cific name is impossible under another genus, there are very many in which it would be improper or questionable, and in respect to which particular consideration is required. Between these cases and the plain ones in which implied namino- could not go wrong, who is to draw the line? Perhaps it might be drawn at monotypic new genera with old specific names. But how to do even this upon recognized principles is a problem. A fatal objection to the principle of names by implication is that all such names, if they are existent, must be indexed in the new "Nomenclator Botanicus" now in preparation. To transcribe under Senecio the specific names pertaining to all the genera which Bentham has referred to that already vast genus is no small matter, and a part of the work will prove superfluous if — as we suppose to be the case — some of these genera, such as Cacalia, ought to be maintained. But that is only the beginning. A more recent author, Baillon, has reduced the genera of Comjyositce nearly one half. For example, to Helenium he has referred Gaillardia, Actinella, Cephalophora, etc. ; to Tagetes he has referred D^^sodia, Ni- colettia, Hymenatherum, and others ; to Helianthus, a greater number of genera, most of them prolific in species. In all probability, most of these reductions will not be approved. Yet, if the principle of constructive naming is adopted, the '^ Nomenclator " must burden its columns with these hosts of inchoate specific names of Baillon, either as received names or as synonyms. It is plain that the principle referred to, besides its incongruity with the leading ideas of received nomenclature, breaks down with its own weiglit. There are, nevertheless, taking arguments in its favor, which need not here be recapitulated ; and the common system has its disad- vantages and liability to abuse ; yet it a2:)i)ears to be the only workable system. As already intimated, the riglit assignment of specific names in reconstructed genera requires particular knowledo'e and careful investigation. And the botanist who reconstructs genera should himself adjust and state the spe- cific names as far as he can. 384 HE VIEWS, BALL'S FLORA OF THE PERUVIAN ANDES. The personal observations and the collections upon which tins essay ^ was founded were made in April, 1882, in an excursion by railway from Lima up to Chicla, which although only seventy-live miles in distance, is at the elevation of 12,220 English feet. Much to his surprise, Mr. Ball found that at this elevation he had not yet reached the alpine region, which really begins about 2000 feet higher. This is three or four thousand feet higher than Grisebach had placed it, on the authority of Tschudi and Humboldt; yet is only what we should expect, since the proper alpine vegetation of the Rocky Mountains in lat. 40"" N. hardly descends below 10,000 feet, and the oscillations of temperature in the Peruvian Andes are small. Equally mistaken, Mr. Ball suspects, must be the common view that the flora of the tropical Andes is scanty in species as compared with high-mountain floras in general. He makes some comparisons from which he infers that the paucity is apparent rather than real, and may be attributed mainly to the paucity of collections in the Andes, since these vast re- gions have been visited at very few points and far between. About a quarter of the Andean phaenogams is of Com- positce, which is double their ratio in North America, which again is greater than that of any other continent. The char- acteristic and the most abundant Andean Com^jositm are the Mutisiacem. Mr. Ball, referring to Bentham's indication of the complex affinities of this group, ventures " to believe that under Mutisiacece are included very many different lines of descent, but that among them there are some minor groups distinguished by very high relative antiquity." And in another connection he opines " that the arguments that have led some distinguished botanists to consider the great family 1 Contributions to the Flora of the Peruvian Andes, with Remarks on the History and Origin of the Andean Flora. By John Ball ; Journal Linnsean Society, xxii. London, 1885. (American Journal of Science and Arts, 3 ser., xxxi. 231.) BAirS FLORA OF THE PERUVIAN ANDES. 385 of Comjjositce as of comparatively recent origin to appear to me altogether inconclusive. When I consider the vast variety of forms which it includes, the degree in which some large groups are localized in different regions of the earth, while others, such as Senecio, have representatives in every zone, I shrink from the conclusion that their origin can be, even in geological language, at all recent. It is, of course, not in- conceivable that plants which we class together under the name Coinpositoi may have come into existence by different lines of descent through gradual modification from different ancestral types. But when we consider the general agree- ment in the structure and arrangement of the essential organs, I think that the balance of probability inclines decidedly to- ward the belief in a community of origin of all the various existing forms. Be that as it may, we are, I think, justified in looking to the mountain region of South America as the original home of many large groups, such as the genus Bac- charis, most of the Mutisiacece, and many genera of other tribes." As to these two suggestions, although it is practically con- venient, and perhaps necessary, to bring all the Labiati- florous Corajpositm under one tribe, as Bentham has done, it seems to us altogether probable that the existing forms are descended from different lines of ancestry. Indeed, by such a conception we can more naturally understand their diverse affinities. But as to the great order they belong to, if there is any large group in which the structure suggests community of origin, it is the ComposltcG. And we suppose that system- atic botanists of large experience would entirely agree with Mr. Ball, that the wide differentiation and distribution of this vast order indicated its high antiquity. Our author has assigned some strong reasons for this opinion. The only argument to the contrary that we know of is an ideal one, based upon two suppositions : one, that Dicotyledons culmi- nate in the Composltm and in such-like orders ; the otlier, that the highest ideal type of plants must be of the latest evolution. But, indeed, the vegetable commonwealth shows no tendency to culminate in any one group or set of groups ; 386 REVIEWS. and it is a questionable morphology which would promote the capitulum of a Thistle or a Dandelion to the head of the class. We remember an interesting^ lecture, in which, recoofnizino- the dominant part which the northern hemisphere and its boreal lands, with their favorable configuration, have un- doubtedly played in floral distribution, it was inferred that the role of the southern had always been comparatively in- significant. But a great deal may have happened in the austral regions before this boreal supremacy was established. Mr. Ball several years ago brought forward his doctrine of the very high antiquity of our actual temperate and alpine floras, of their coexistence in highly elevated regions of low latitudes even in early times. So now, applying his former conclusions to the southern hemisphere, and to " a period remote even in geological language," he notes that " the spe- cial generic types of the antarctic flora " " belong without exception to the great groups or natural orders which are now almost universally diffused throughout the world ; and the ancestral types from which they originated were probably carried to that region at a remote period, when the physical conditions of the earth's surface were widely different from those now prevailing." " Various considerations tend to the conclusion that the dispersal of the chief cosmopolitan genera of plants may have coincided with the period of the older secondary rocks ; and at that period physical agencies far transcending those of our experience prevailed throughout the earth. If the ancestors of the antarctic types of vegeta- tion were then established in a south polar continental area, and were developed from them by gradual modification, I see no difficulty in believing that they may have maintained themselves through successive gradual changes of physical conditions within the same region, and even that some may still survive within the Antarctic Circle." Whether or not one accepts the idea of such high geo- logical antiquity which Mr. Ball claims for what he calls Cos- mopolitan types, we must wholly agree with him in his use of this name for them, in preference to that of the Scandinavian. The latter term was used by Hooker before the relation of BALUS FLORA OF THE PERUVIAN ANDES. 387 the present flora of our temperate zone to a former lilo-lj. northern vegetation was made clear, and before the types in question could " with more reason be referred to North America than to Scandinavia." Mr. Ball's remark that, as to many of them, "the balance of evidence points to an original home in the high mountains of lower latitudes " chimes in with his favorite and original doctrine. And this indeed seems likely to gain ground the more it is considered and applied, as he is applying it, to the explanation of actual distribution. The interesting problem is to discriminate, as well as may be, the two commingled elements of the northern temperate floras, one of arctic, the other of more endemic mountain origin. An interesting presentation, as concerns central Europe, is made in Heer's " Nival Flora of Switzerland," a posthumous work published by the Societe Helvetique des Sciences Naturelles, of which a summary is given in " Nature " for December 31, 1885. The following idea is extremely suggestive. " In a zoo- logical as well as a botanical sense Brazil is one of the most distinct and separate regions of the earth. It is in large part a granitic region, from which vast masses of superincum- bent strata have been denuded, and where the granite itself has undergone a great amount of decay and ablation. We there see the ruins of one of the greatest mountain masses of the earth, where a very ancient flora and fauna were devel- oped, of which portions were able to migrate to a distance, while others have been modified to adapt themselves to the gradual changes of the environment. Many vegetable groups, which are but slightly represented in the higher region of the Andes, such, for instance, as the 3Mastomacecv, probably had their origin in the mountains of ancient Brazil." We are now only beginning to reach some conception of the role which the Andes and their prolongation througli Mexico have taken in determining the character of no small part of the North American flora. Following up some ideas which were touched upon in this Journal (vol. xxvli., Nov., 1884, p. 340) and elsewhere, Mr. Ball writes : — 388 REVIEWS. " When we consider that, although subsidence has prob- ably at various times separated the two portions of the con- tinent, the highlands of Mexico and Central America have, in all probability, served during long periods as a bridge over which some portions of the mountain vegetation may have been transferred from north to south, and vice versa, we are led to feel surprise rather at the separations now ex- isting than at the presence of many genera and of a few iden- tical species in the flora of the Andes and that of the Kocky Mountains. It is true that I have reckoned as Andean genera and species many that extend northward as far as Mexico ; and it may well be that that region, so rich in varied forms of vegetation, is the original home of some that now appear to be more fully developed in the mountain ranges of western North America. Among the widespread Ameri- can types we must note two natural orders whose original home may with some confidence be placed in the north- western part of the continent. The Polemoniacece, of which about 140 species belong to that region, are represented in the Andes by five species of Gilia, one of CoUonia, and by the endemic genus Cantua. They have sent to the Old World two or three species of Phlox in northern Asia [we believe only one, and that not far over the border], and a single emigrant which has reached Britain, — the Jacob's Ladder of old-fashioned gardens, — which maintains a strug- gling existence in several isolated spots in Europe. The other specially American family is that of the HydropJiyllacece., of which 12 genera are known in North America, but which is represented in the Andean chain by only four species of Phacelia." The Loasacece illustrate the opposite course of migration. A list of the plants which Mr. Ball collected in the upper valley of the Kimac in the Peruvian Andes, with various an- notations and the characters of some new species, concludes the present interesting contribution to Andean Botany. We believe that a second paper upon the subject may be ex- pected. Two or three comments upon individual plants of the list will bring our review to a close. BALVS FLORA OF THE PERUVIAN ANDES. 389 Erodium cicutarmm. — Although Mr. Ball notes the wide diffusion of this Old World species in South America, and that it attends the distribution of cattle, he seems at a loss to account for its presence in the Peruvian Andes at the heio-ht of 12,500 feet. He supposes that it has not shown the same readiness to establish itself in North America. This is true of the Atlantic but not of the Pacific side. In California and the adjacent districts the Alfilaria, as it is popularly called, has taken such full possession that we can hardly convince even the botanists that it is an introduced plant. The authors of the " Botany of California " speak of it as " more decidedly and widely at home throughout the interior tlian any other introduced plant, and, according to much testi- mony, it was as common throughout California early in the present century as now. ... It is a valuable and nutritious forage plant, reputed to impart an excellent flavor to milk and butter." At Santa Barbara and other parts of southern California it is used for lawns around dwellings, and it seems to be the only resort. It makes a passable substitute for grass so long as the rainy season lasts or irrigation is kept up. It must have been brought in with tlie earliest cattle, and have found on the Pacific coast a perfectly suit- able climate. Caldasia of Lagasca, Mr. Ball shows us, must be restored as the name of the genus named Oreomyrrhis by Endlicher. Relbunium, upon a general survey of the species, will in our opinion be found quite untenable as a genus. Phacelia circinata^ which extends almost from one end to the other of the American continent, is said to be singularly constant, exhibiting no marked varieties. But we have in North America a remarkable diversity of forms, the ex- tremes of which, by themselves, no botanist would refer to one species, although intermediate forms inextricably com- bine them. INDEX. AcanthacecB in the Prodromus, 22. African, South, Plants, Harvey's, 36. Agassiz's Zoological Nomenclator, 41. Aldrovanda vesiculosa, insectivorous habits of, 208. Algce, fecundation of, 91. Alligator Pear, history of the, 342. Allium, North American, 281. Allium stellatum, 281. Allium vineale, 281. Amarantacece in the Prodromus, 27. Amelanchier, 195. Amole, 198. Ampelidece, Planehon's Monograph of, 250. Ampelopsis, characters of, discussed, 251. Ancestry of plants, 269. Andes, Peruvian, Ball's Flora of, 384. Anemophilous plants, 234. Anthephora axillijlora, 113. Apple, effect of foreign pollen on an, described (note), 189. Aquilegia Jlavescens, 185. Archaeology, plant, 269. Arctic Plants, distribution of, 122. Asclepiadece in the Prodromus, 19. Aspirate, use of, in forming botanical names, 262. Australia, Bentham's Flora of, 246. Authors' names, citation of, discussed, 370. Balanophoreoe, structure and affinities of, 94. Ball, John, and J. D. Hooker, Tour in Marocco, 255. Ball's Flora of the Peruvian Andes, 384. Banana, history of the cultivation of the, 344. Basin, the Great, character of, de- scribed, 181. Basin, the Great, statistics of the vege- tation of, 182. Batatas vulgaris, origin of , discussed, 317. Beans cultivated by North American Indians, 348. Bentham on Euphorhiacece, 259. Bentham's Flora of Australia, 246. Bentham's Flora of Hongkong, 117. Bentham's Hand-Book of the British Flora, 104. Bentham and Hooker's Genera Plan- tarum, 355. Bignonia capreolata, mode of growth of, 177. Bolandra, 282. Borrea ciliaris, black dots on ^note), 58. Botanical descriptions in various lan- guages contrasted, 302. Botanical riddles, 291. Botanical works, methods of publica- tion of, discussed, 285. Botanists, characteristics of, 284. Botany, fossil, discussion of, 269. Botany, Henfrey's, 72. Botany, Natural System of, 1. Botany, Watson's, of the 40th Parallel, 180. Bottle-gourd, history of the, 330. Boussingault on Influence of Nitrates, 100. Brazil, antiquity of the flora of, dis- cussed, 387. 392 INDEX. British Flora, Bentham's Hand-Book of, 104. British flora, statistics of, 105. Buchloe dactyloides, 113. Buffalo-grass, 112. Bulbochcete, nature of, 92. Calanthera dactyloides,llS. Cassava-plant, origin of, discussed, 320. Cell-formation, 54. Cell-markings, 53. Cell, physiological condition of, 55. Cell, the vegetable. Von Mohl's, 51. Cerojjegia, revolution of growing stem of, 162, 167. ChamcElirium Carolinanum, 281. Champlain, Les Voyages du Sieur de, 84, 316 (note), 347. Circumnutation defined, 304. Circumnutatory movements in plants explained, 306. Citation of authors' names discussed, 49, 370. Clematis, mode of growth of stems of, 168. Climates described, 272. Climbing, advantage of, in plants de- scribed, 178. Climbing Plants, Darwin's Movements and Habits of, 158. Climbing plants, irritability of growing stems of, 164. Close-breeding, results of, 266. Close-fertilization, advantages of, dis- cussed, 229. Cobcea scandens, mode of growth of, 177. Color in flowers, purpose of, described, 227. Comandra, 29. CompositcB in the Prodromus, 16. ComjJositcB of the Peruvian Andes, 384. Connubial relations of plants, 241. Crescentia Cujete, note on derivation of the name of, 330. Cross-fertilization in the vegetable kingdom, 217. Cross-fertilization, benefits of, 245. Cross-fertilization, methods of securing, described, 230. Cross-fertilization, remarks on, 225. Cucurbit a, Naudin on the genus, 83. Cucurbita maxima, 83, 332. Cucurbita moschata, 83, 332. Cucurbita Pepo, 83, 332. Cucurbita verrucosa, 85. Cucurbitaceve, cross-breeding of, 86. Cucurbitacece, origin of cultivated species of, 329. Cucurbitacece, tendrils of, 174. CvJtivated Plants, De CandoUe's Origin of, 311. Cultivated plants, improvement of, 109. Cultivated plants, origin of, 69. Cupuliferce, present condition of, in Europe, 137. Curtis' Trees of North Carolina, 115. Darwin's Cross and Self -Fertilization in the Vegetable Kingdom, 217. Darwin's Different Forms of Flowers on Plants of the Same Species, 241. Darwin's Insectivorous Plants, 206. Darwin, Movements and Habits of Climbing Plants, 158. Darwin's Power of Movement in Plants, 304. Decaisne's Monograph of the Genus Pyrus, 186. De CandoUe's Prodromus, 15. De CandoUe's (A.) G^ographie Bota- nique, 67. De CandoUe's (A.) Monograph of Smilacere, 249. De CandoUe's (A.) New Monographs, 248. De CandoUe's (A.) Nomenclature, 358. De CandoUe's (A.) Origin of Cultivated Plants, 311. De CandoUe's (A.) Phytography, 282. De CandoUe (A.) on Species, 130. Descriptions, abridged, 290. Descriptions, developed, 290. Descriptions of plants, discussed, 290. Dextrorse and sinistrorse, 297. Diagnosis of plants discussed, 290. Dianthera, 22. INDEX. 393 Dicentra, origin of name discussed, 878. Bidytra, origin of name discussed, 378. Dielijtra, origin of name discussed, 378. Bioncea, insectivorous habits of, 207. Dioscorea sativa, origin of, discussed, 322. Bouglasia, 128, Drosera rotundifolia, insectivorous habits of, 207. Drosophyllum, insectivorous habits of, 208. Echinocystis lobata, movements of, 174. Embryo, direction of, explained, 13. Emerson's Trees and Shrubs of Massa- chusetts, 204. Endlicher's Genera Plantarum, 33. Engelmann's Notes on the Genus Yucca, 196. Engelmann on the Buffalo-grass, 112. Epiontology defined, 131. Epping Forest, 253. Erodium cicutarium, distribution of, 389. Euphorbiacece, Bentham on, 259. Fecundation, Process of, in the Vege- table Kingdom by Radlkofer, 91. Fertility, degrees of, in crossed seed- lings of Ipomoea j)urpurea, 221. Fertilization, self-, Henslow on, 263. Fertilization of flowers, relation of in- sects to, 235. Fertilization of Red Clover, 226. Flora of Australia, Bentham' s, 246. Flora, British, Bentham's Hand-Book of, 104. Flora of Hongkong, Bentham's, 117. Flora, Indian, Hooker and Thomson's, 62. Flora of the Peruvian Andes, Ball's, 384. Flora, Tertiary, vegetation of, in south- east France, 145. Floras discussed, 292. Flowering and Flowerless Plants, 5. Flowerless and Flowering Plants, 5. Forms of flowers different on plants of the same species, 242. Fossil botany, discussion on, 209. Fossil plants, nomenclature of, 380. Fruit, ripening of, 78. Fruits, nomenclature of, 78. Eucacece, fecundation of, 93. Euciis, note on Decaisne and Thuret's Memoir of, 59. Eungi, note on Ldveille's paper on, 59. Gaura parvijlora, 207. Genera, number known at different epochs, .357. Genera Plantarum, compared, 34. Genera Plantarum, Bentham and Hooker's, 355. Genera Plantarum, description of the different, 292. Genera Plantarum, Endlicher's, 33. Genera Plantarum, history of the dif- ferent, 355. Generic names, priority of, discussed, 363 ; duplicate use of, 45. Gentiana Andrewsii, fertilization of, 267. Geographic Botanique, A. De Cau- dolle's, 67. Glaucus defined, 297. Granadillas, history of the cultivation of, 342. Gravity as a factor in plant-growth discussed, 309. Greenland, origin of the flora of, 123. Groups, methods of arranging, 289. Groups, natural arrangement of, 294. Gynsecium, use of the term discussed, 297. Hamamelis Virginica, properties of, 79. Harvey's South African Plants, 30. Hastingsia, 281. Heer, appreciation of, 270. Helianthiis argop/iylltis, experiments on growth of, 101. Helianthus giganteus, origin of Jerusa- lem Artichoke discussed, 315. Helianthus tuberosus, origin of, dis- cussed, 314. 394 INDEX. Henfrey's Botany, 72. Henslow, George, Self -Fertilization of Plants, 263. Herbaria, their history, 304. Heredity and variability in plants, nature of, 212. Hesperoenide teiiella, 90. Ileteromeles, 195. Ueterosmilax, 2-19. History of herbaria, 304. Hollisteria, 2S2. Hongkong, Bentham's Flora of, 117. Hook-elimbers, mode of growth of, 171. Hooker on Balanophorea, 94. Hooker on distribution of Arctic plants, 122. Hooker on Welwitschia, Ibl. Hooker and Ball's Tour in Marocco, 255. Hooker and Thomson's Indian Flora, 62. Hop, revolutions of growing stem of, 160. Hops, native in North America, 328. Ilumulus Lupulus, native in North America, 328. Hybrids, characters of, 212. Hybridization of Lilies, 238. Hybridization, views upon, in Flora of India, 63. Hydrophyllacece in the Prodromus, 20. Imbrication, quiucuncial, 120. Improvement of cultivated plants, 109. Indian Flora, Hooker and Thomson's, 62. Initials, capital, use of, discussed, 48, 381. Insectivorous Plants, Darv\'in's, 206. Insects in relation to the fertilization of flowers, 235. Ipomoea purpurata, Dai-Avin's experi- ments in fertilizing, 218. Japan, Siebold's Flora of, 37. Japanese flora contrasted with North American, 38. Jerusalem Artichoke, origin of, dis- cussed, 314 Kidney Bean, history of the, 344. LabiatcE in the Prodromus, 23. Lagenaria, history of, 330. Lanceolate, defined, 297. Language for botanical works dis- cussed, 286. Lasiostega humilis, 114. Latin, botanical, discussed, 286, 361. Leaf -climbers, mode of growth of, 168. Leaf, definition of the, 295. Lesquereux, appreciation of, 270. Lichens, Antheridia of (note), 58. Lichens, note on Tulasne's Memoir of, 58. Lilac, terminal bud of, 75. I LiliacecE, North American, 278. LiUace(B,V^ atson's arrangement of, 278. ' Lilies, hybridization of, 238. I Lillium Grayi, 281. I Lillium Parkmanni, 238. Lima Bean, origin of, 351. Lindley's Natural System of Botany, 1. Literary material, methods of collect- ing, discussed, 287. Longevity of trees, 82. j Lophosjjermum scandens, mode of I climbing, 170. I Lycopersicuvi esculentum, history of the cultivation of, 341. Lygodium scandens, revolution of growing stems of, 163. Maianthemum bifolium, 281. Mammoth, discussion on the origin of the, 142. Manihot iitilissima, origin of, discussed, 320. Manioc, origin of, discussed, 320. Marocco, Hooker and Ball's Tour in, 255 ; early travels in, 255. Massachusetts, trees and shrubs of, 204. Maurandia semperflorens, mode of growth of, 171. Meliacem, monograph of, 249. Mohrs Vegetable Cell, 51. MonanthochJoe, 114. Monographs of plants discussed, 290. INDEX. 395 Monstrosities, vegetable, 40. Moquin-Tandou's Vegetable Monstrosi- ties, 40. Musa, history of the cultivation of, 344. Names for garden-plants discussed, 369. Names, formation of personal, for plants discussed, 368. Names, methods of abbreviation dis- cussed, 376. Names of natural orders, 11, 305. Names of organs, rules for choosing, 296. Names of species in combined genera discussed, 49, 382. Names, permissible changes in, 377. Names, popular, of British plants dis- cussed, 106. Names, substantive specific, for plants discussed, 48, 368. Names, use of unpublished, 373. Natural orders, analogies of, 80. Natural orders, names of, 11. Natural System of Botany, 1. Naudin's Nature of Heredity and Variability in Plants, 212. Naudin on the genus Cucurbita, 83. Nitrates, influence of, in production of vegetable matter, 100. Nitrates, sources of, in soil, 103. Nolinece, 280. Nomenelator, Agassiz's Zoological, 41. Nomenclature, A. De Candolle's, 358. Nomenclature, Bentham's views on, 206. Nomenclature of fossil plants, 380. Nomenclature of organs, 294. North Carolina, trees of, 115. Nyctaginacece in the Prodromus, 28. Nyctotropism explained, 307. Oakesia sessilifoUa, 280. Oaks, A. De Candolle on distribution of the, 130. Oaks, history and origin traced, 136. Oaks, variations in, 132. Oca, history of cultivation of, 328. Odor in flowers, purpose of, described, 227. (Edogonium, nature of, 91. Ordinal names, proper terminations of, discussed, 11, 365. Organography discussed, 293. Organs, nomenclature of, discussed, 294. Organs of plants, nomenclature of, dis- cussed, 379. Organs, rules for choosing names of, 296. Origanum vulgare, Darwin's experi- ments in crossing, 223. Orobanche, 21. Oxalis crenata, history of cultivation of, 328. Oxalis tuberosa, history of cultivation of, 328. Papilionaceous plants, fertilization of, 267. Parkman, F., on the Hybridization of Lilies, 238. Parrya macrocarpa, 185. Parthenocissus, 252. Passijlora gracilis, movements of, 173. Passiflora, history of cultivation of, 342. Passijlora incarnata, 343. Peach, origin of the, 329. Pears, history of the cultivation of, 193. Pears, method of producing large, described, 191. Pears, number of cultivated species of, 188. Pears, races of, 192. Pears, stock for, 191. Peraphyllum, 195, Persea gratissima, history of, 342. Phacelia circinata, 389. Phaseolus lunatus, origin of, .351. Phaseolus vulgaris, histoi-y of. 344. Photinia, 195. Phyllotaxy, 76. Physiology of plants, 81. Phytogamy, 241. Phytography, A. De Candolle's, 282. Pickeringia, 15. Pilea pumila, 90. Pinguicula, glandular hairs of, 209. Pirus, 89, 329. 396 INDEX. Pistillum, use of the term disciassed, 297. Planchon's Monograph of Ampelidexje, 250. Plant Archaeology, 269. Plants, ancestry of, 209. Plants, aneniophilous, 234. Plants, Climbing, Darwin's Movements and Habits of, 1.58. Plants, Cultivated, A. De CandoUe's Origin of, 311 ; improvement of, 109. Plants, connubial relations of, 241. Plants, Darwin's Power of Movement in, 304. Plants in Glazed Cases, Ward's Growth of, 59. Plants, history of their distribution, 70. Plants, Naudin on the Nature of Hered- ity and Variability in, 212. Plants, physiology of, 81. Plants, relation of, to surrounding con- ditions, 68. Plants, Self-fertilization of, by George Henslow, 263. Plants, self-sterile, 224. Poinsettia, adaptation to insect-fertil- ization, 234. Pollen, direct action of, on ovary and ovules, 87 ; value of, discussed by Darwin, 228. Popular names of British plants dis- cussed, 106. Portulaca oleracea, introduction into North America, 326. Potato, origin of, discussed, 09 note, 317. Primula Sinensis, glandular hairs of the leaves of, 209. PriviulacecB in the Prodromus, 18. Prodromus, De CandoUe's, 15. Proserpina, Ruskin's, 199. Pruinosus defined, 297. Publication of botanical works dis- cussed, 285. Publicity of publication discussed, 369. Purslain, introduction into North America, 326. Pyrularia, 28. ! Pyrus, characters of, 193. Pyrus, Decaisne's monograph of the genus, 186. Pyrus, division of the genus, 194. Pyrus, limitation of the genus, 189. Pyrus, morphology of the gynsecium of, 190. Pyrus, orthography of, 329. Quaternary climate described, 275. Quercus Ilex, 139. Quercus Bobur, 138, 139. Quincuncial imbrication, 120. Races, fixity of new, in cultivation, 110. Radicle, character of the, 121. Radicle of the embryo, true character of, 74. Radlkofer's Process of Fecundation in the Vegetable Kingdom, 91. Panunculus alismcefolius, 184. Ranunculus fascicularis, 184. Panunculus orthorhynchus, 184. Red Clover, fertilization of, 226. Pestiacece, monograph of, 248. RhamnacecE, stamens of, 76. PJdlJogonum, 249. Riddles, botanical, 291. Root, mode of penetration into the son, 306. Root-hairs, 75. Root-tip, functions of, discussed, 307. Posa setigera, mode of growth of, 171. Ruskin's Proserpina, 199. Salsolacece in the Prodromus, 26. Santalacece in the Prodromus, 28. Saporta's Monde des Plantes avant I'Apparition de 1' Homme, 269. Sarracenia, history of discoveries re- lating to, 210. Saxifrages, glandular hairs of the leaves of, 209. Schoenolirion album, 280. Seeds, reproduction by, 58. Self -fecundation discxissed, 267. Self-fertilization in the vegetable king- dom, 217, 263. Self -sterile plants, 224. Sequoia, history of, 277. INDEX. 397 Shortia glacifolia, 16. Siebold's Flora of Japan, 37. Sinistrorse and dextrorse, 297. Sisymbium, 186. Smilacece, A. De CandoUe's Monograph of, 249. Smilacece, distribution of, 250. Smilacina, nomenclature of, 261. Smilax, 249. Soil, soluble matters in, 103. Solanum tuberosum, origin of, discussed, 317. Species, A. De €andolle's definition of, 149 ; Linnaeus' definition of, 148. Species, origin of, Naudin's views on, 216. Species, variation and distribution of, 130. Species, various views on the, 147. Species, views on the derivation of, 140. Specimens, incomplete, difi&culties of describing-, 292. Sphceroplea, nature of, 92. Statistics of the British flora, 105. Stem, the initial, defined, 310. Stems of climbing plants, revolutions of, 159. Streptanthus, 185. Style in botanical writing discussed, 302. Sullivantia Ohioensis, 282. Sweet Potato, origin of, discussed, 317. Tendril-bearing plants, origin of, 159. Tendril-climbers, mode of growth of, 172. Tendrils, movements of, described, 172. Tendrils of Cucurbitacece, 175. Terms, contradictory use of, 296. Terms, multiplication of, discussed, 296. Thalictrum Fendleri, 184. Thomson and Hooker's Indian Flora, 62. Tomato, history of the cultivation of the, 341. Tree-culture, possibilities of, in Eng- land, 254. Trees, longevity of, 82. Trees of North Carolina, 115. Trees and Shrubs of Massachusetts, Emerson's, 204. Urticacece, geographical distribution of, 89 ; relationships of, 88. Urticacece, Weddell's Monograph of 87. Utricularice, North American in the Prodromus, 18. Uvularia, 281. Variability and heredity in plants, na- ture of, 212. Variation of species discussed, 214. Varieties, importance of describing, 293. Varieties, method of treating, 288. Varieties, their relations to species, 287. Vegetable kingdom, cross and self- fertilization in, 217. Vegetable kingdom, Radlkofer's pro- cess of fecundation in, 91. Vegetable monstrosities, 40. Vegetable world, history of, 274. Vilmorin on the improvement of cul- tivated plants, 109. Virginia Creeper, tendrils of, 176. Vitis, characters of, discussed, 251. Wallace, A. R. , on Epping Forest, 253. Ward's Growth of Plants in Glazed Cases, 59. Wardian cases, growth of plants in, 59. Watson's Botany of the 40th Parallel, 180. Watson on North American Liliacc(C, 278. Weddell's Monograph of Urticacece, 87. Weed, Henslow's definition of a, 268. Weeds, fertilization of, 268. Weldenia, 281. Welwitschia, J. D. Hooker on, 151. Yam, origin of the, discussed, 322. Yucca, Engelmann's notes on, 190. Yucca, anthesis of, 197. Yucca, nature of the fruit of, 198. f-H' 2K:'-' m ^y^m im: wm: